Modeling human cortical development in vitro using induced pluripotent stem cells

Jessica Mariania,b, Maria Vittoria Simoninia,b, Dean Palejeva,b, Livia Tomasinia,b, Gianfilippo Coppolaa,b, Anna M. Szekelya,c,d, Tamas L. Horvathe, and Flora M. Vaccarinoa,b,f,1

aProgram in Neurodevelopment and Regeneration, bChild Study Center, and Departments of cGenetics, dNeurology, eComparative Medicine, and fNeurobiology, Yale University School of Medicine, New Haven, CT 06520, USA

Edited* by Pasko Rakic, Yale University, New Haven, CT, and approved June 11, 2012 (received for review February 21, 2012)

Human induced pluripotent stem cells (hiPSCs) are emerging as described method for neural differentiation that allows formation a tool for understanding human brain development at cellular, of serum-free, floating embryoid body-like, quick aggregates molecular, and genomic levels. Here we show that hiPSCs grown (SFEBq) (16) we derived neurally differentiated SFEBq-like in suspension in the presence of rostral neuralizing factors can multilayered structures from hiPSC. We also performed detailed generate 3D structures containing polarized radial glia, interme- immunophenotyping of these cultures, analyzed their tran- diate progenitors, and a spectrum of layer-specific cortical scriptome, and compared it with transcriptomes of hESCs, hiPSCs, reminiscent of their organization in vivo. The hiPSC-derived mul- human neuronal progenitors, and developing human brain tissue. tilayered structures express a expression profile typical of These combined analyses demonstrate that hiPSC-derived multi- layered structures display a gene expression profile typical of the the embryonic telencephalon but not that of other CNS regions. embryonic telencephalon, and that neurons within these structures Their transcriptome is highly enriched in transcription factors encompass both lower and upper cortical layer fates. controlling the specification, growth, and patterning of the dorsal telencephalon and displays highest correlation with that of the Results early human cerebral cortical wall at 8–10 wk after conception. Characterization of hiPSC Lines. We used two hiPSC lines: PGP1-1 Thus, hiPSC are capable of enacting a transcriptional program (17) and i03-01#9, a line derived in our laboratory. Both lines specifying human telencephalic (pallial) development. This model were established from healthy control adult male skin fibroblasts will allow the study of human brain development as well as dis- through retroviral reprogramming vectors expressing OCT4, NEUROSCIENCE orders of the human . , , and c- (18). These hiPSC lines exhibited typical hESC morphology and expressed canonical pluripotency human embryonic | | differentiation | cortical layer markers by immunocytochemistry as well as NANOG, DNMT3B, and other endogenous mRNAs that characterize pluripotent merging data highlight the complexity and dynamic nature of cells (Fig. S1 A and B). Pairwise correlations of global gene ex- Egene expression in the central nervous system (CNS) and the pression data obtained by microarrays (Dataset S1) showed high divergence between human and other mammalian species, which is similarity in gene expression between PGP1-1 and the hESC line especially pronounced in the developing brain (1–4). Exploring H1 (Pearson’s r > 0.978), i03-01#9 and H1 (Pearson’s r > 0.975), ’ > such differences may reveal the genetic underpinnings of the larger and PGP1-1 and i03-01#9 (Pearson s r 0.979). size and complex architecture of the human brain and elucidate the We then tested the hiPSC lines with Pluritest (http://www. fi molecular and cellular substrates of higher cognitive functions, as pluritest.org), an approach that classi es samples according to fi well as of our vulnerability to neurodevelopmental and neurode- more than 450 genome-wide transcriptional pro les, including generative disorders. To understand the genetic programs that 223 hESC and 41 hiPSC lines from multiple laboratories (19). drive cell specification and differentiation in the human brain, it is The i03-01#9 and PGP1-1 displayed Pluritest scores consistent important to develop model systems that recapitulate dynamic with normal human pluripotent cells and clustered together with the H1 hESC line (green arrowhead, Fig. S1D). Taken together, aspects of neural development, in addition to making inferences fi fi from commonly used models of lower mammalian species. these results con rm that PGP1-1 and i03-01#9 ful ll the Recapitulating human neural development in vitro using human established criteria for completely reprogrammed iPSC lines. induced pluripotent stem cells (hiPSCs) can provide our first un- derstanding of how genetic variation and disease-causing muta- Excitatory Neuronal Precursor Cells Are Generated Within Multilayered tions influence neural development. Human iPSCs generated from Structures Derived from hiPSCs. Undifferentiated PGP1-1 and i03- 01#9 colonies were dissociated into single cells and cultured in reprogrammed cells can be differentiated into any tissue, including – the CNS, while maintaining the genetic background of the in- suspension in the presence of FGF2 (5 20 ng/mL) and inhibitors of the bone morphogenetic (BMP), Wnt/β-catenin, and dividual of origin. These critical features have been exploited to β model monogenic forms of neurodevelopmental disorders, such as TGF- /activin/nodal pathways to induce forebrain fate. The resulting 3D aggregates were then transferred onto a coated surface Rett and Timothy syndromes, and even psychiatric disorders with – complex inheritance, such as schizophrenia (5–7). and maintained without any additional factors until day 45 70. On The brain and spinal cord develop according to distinct dif- day 25, cells within the SFEBq-like structures expressed early ferentiation programs from the earliest stages of CNS develop- neuroepithelial markers, including BLBP, N-CADHERIN, PAX6, ment (i.e., at the progenitor stage during gastrulation) (8, 9). and NESTIN (Fig. S1C). Regional differences in gene expression within stem and pro- genitor cells appear at the onset of the formation of both mouse (10, 11) and human CNS, as shown by recent studies of the human Author contributions: J.M., A.M.S., and F.M.V. designed research; J.M., M.V.S., L.T., and transcriptome using postmortem tissue (4). T.L.H. performed research; J.M., M.V.S., D.P., L.T., G.C., and F.M.V. analyzed data; and J.M., M.V.S., A.M.S., and F.M.V. wrote the paper. Neural cells are thought to differentiate by “default” into an fl anterior, forebrain-like fate (12), and indeed monolayer neuronal The authors declare no con ict of interest. cultures derived from hiPSCs and human embryonic stem cells *This Direct Submission article had a prearranged editor. (hESCs) can express some forebrain markers (6, 13, 14). However, Freely available online through the PNAS open access option. they have not yet been shown to recapitulate the transcriptional 1To whom correspondence should be addressed. E-mail: fl[email protected]. program that gives rise to the mammalian telencephalon, and This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. their regional specification remains unclear (15). Using a recently 1073/pnas.1202944109/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1202944109 PNAS Early Edition | 1of6 Downloaded by guest on September 28, 2021 hindbrain (Fig. 1 P and Q). Additionally, cells did not express ISLET1 and showed barely detectable levels of OLIG2 (Fig. 1R), TFs both expressed by more posterior and ventral regions of the brain and in the spinal cord. We next investigated the potential of these 3D multilayered aggregates to generate inhibitory neurons. On day 50 of neuronal differentiation, we found presence of GABAergic precursor cells (DLX1/2+, MASH1+) as well as more differentiated GAD-67+ inhibitory neurons (Fig. S3). Strongly positive DLX+ cells ten- ded to be rounded and devoid of GAD-67+ processes (Fig. S3B, arrowheads), whereas cells weakly stained for DLX acquired long GAD-67+ processes (Fig. S3 A and C, arrows) and seemed to be migrating within the structure. Importantly, cells positive for inhibitory markers did not express the excitatory neuronal marker TBR1, suggesting that excitatory and inhibitory neuronal lineages are distinct and coexist in the preparation. The GABAergic component was always segregated from excitatory cells, in the center (Fig. S3A) or at the periphery (Fig. S3 D–F). The overall

Fig. 1. Radial glia and excitatory progenitors differentiate from hiPSCs at day 50 in vitro. Day-50 forebrain-like structures contain radial glial cells immunoreactive for NESTIN, PAX6, and BLBP (A, B, and E), and βIIITUBULIN+ and TBR1+ neurons (B, D, and G). The radial glia express SOX1 and SOX2 (H–L), are mitotically active, as shown by Ki67 expression (M–O, red), and are polarized, displaying N-CADHERIN+ apical end-feet (C and F) as well as apical mitoses (M–O, arrowheads). A layer of TBR2+ intermediate progenitors (C and F) surrounds the radial glial layer and displays pH3+ basal mitoses (M–O, arrows). The neurons express MAP2 but rarely ZIC1 (P and Q) and do not express ISLET1 and OLIG2 (R). (Scale bars, 200 μminA–D and M;5μminE–G and I;10μminH, J–L, N, and O.)

Immunofluorescence analysis of the 3D structures at day 45– 50 in serial cryosections revealed neural tube-like substructures within each aggregate, composed of radially arranged cells with apico-basal polarity, organized around a central lumen (Fig. 1). The radially arranged cells were PAX6+, NESTIN+, BLBP+, and GFAP+ and displayed N-CADHERIN immunoreactivity along their apical edge (Fig. 1 A–C, E, and F and Fig. S2 A–C). These radial glia-like progenitors expressed the neuroepithelial transcription factors (TFs) SOX1 and SOX2 (Fig. 1 H–L). Su- perficially to the radial glial layer were numerous neuronal precursor cells expressing T-box brain 2 (TBR2) (Fig. 1 C and F), a TF expressed by intermediate neuronal progenitors in the subventricular zone (SVZ) of the mammalian cerebral cortex Fig. 2. Evidence for synapses in hiPSC-derived multilayered structures at (20). The proliferative markers Ki67 and phosphohistone H3 50 and 70 d in vitro. (A–D) Day-70 multilayered structures analyzed by EM (pH3) were expressed by both radial glial cells and the sur- show examples of direct apposition between vesicles-containing putative terminals (black arrows) and neuronal cell bodies. (E and F) Synapto- rounding intermediate precursors, and pH3 was expressed at the fi luminal (apical) surface, suggesting that radial glia undergo physin/neuro lament 200 double immunostaining at day 70 showing nu- merous synaptophysin-labeled boutons decorating the neurites. (G–O) interkinetic nuclear migration (Fig. 1 M–O). More mature + + Immunoelectron microscopy showing synaptophysin-immunoprecipitate neurons (βIIITUBULIN , MAP2 ) also expressing T-box brain G–I + within putative axon terminals contacting cell bodies (soma) ( ), dendrites 1 (TBR1) (21), presumably progeny of TBR2 precursors (20), (d) (J–L), and spines (s) (M–O). White arrowheads, synaptic thickening; m, + were found superficially (basally) to the TBR2 layer (Fig. 1 B, mitochondia; a, autophagosome. (P and Q) Day-50 multilayered structures D, G, and P), reproducing the typical architecture of the de- stained for SYN I and βIIITUBULIN (P) or PSD-95 and MAP2 (Q). White arrows veloping mammalian cortex (22). In contrast, we found an almost in P point to SYN I-positive boutons. (Scale bars, 2 μminA and C;0.5μminB complete absence of cells expressing ZIC1, a TF expressed in the and D;10μminE, P,andQ;5μminF;1μminG–O.)

2of6 | www.pnas.org/cgi/doi/10.1073/pnas.1202944109 Mariani et al. Downloaded by guest on September 28, 2021 percentage of cells expressing GAD-67 was 6.9 ± 1.0 and 7.4 ± 0.7 the other reprogramming factors (SOX2, KLF4,andc-MYC)as in two different preparations. well as of other characterizing a pluripotent state (NANOG, DNMT3B, and JARID2) was also down-regulated (Dataset S1). Neuronal Connectivity in the Multilayered Structures Derived from Conversely, mRNAs encoding for known effectors of neuronal hiPSCs. At day 70 of differentiation in vitro, the majority of po- differentiation increased in abundance, including MAP2, βIII larized radial glial cells disappeared from the aggregates, to- TUBULIN, ephrins and their receptors, semaphorins, and neural gether with the PAX6 and BLBP protein expression. Electron cell adhesion molecules (Table S1). Genes involved in neuro- microscopy revealed the presence of immature synaptic boutons SNAP25 SNAP29 – transmission were also up-regulated, including , , containing vesicles (Fig. 2 A D, arrows), which contacted large SNAP91, SYN I, SYP, Bassoon, PSD-95, and glutamate ionotropic neuronal cells via thickenings resembling immature synapses, in and metabotropic receptors (Table S1). that the postsynaptic membrane specializations and associated Ingenuity pathway analysis for differentially expressed genes structures did not appear fully developed. Immunostaining for in day-50 3D structures revealed that the most significantly en- the presynaptic markers synaptophysin (SYP) and synapsin riched biological function categories were Nervous System Devel- I (SYN I), as well as postsynaptic density protein 95 (PSD-95), opment and Function, Cancer, Cellular Assembly and Organiza- revealed extensive synaptic differentiation (Fig. 2 E, F, P, and fi Q). Immunoelectron microscopy demonstrated SYP immuno- tion, and Neurological Disease (Dataset S2). The most signi cant precipitate within boutons containing synaptic vesicles (Fig. 2 terms under the Nervous System Development and G–O), which contacted dendrites and spines via synapses (Fig. 2 Function category were Neurogenesis of Cerebral Cortex and J–O, arrowheads). We also observed extensive immunostaining , Guidance of Neurites, Brain Development, Hippo- for the vesicular glutamate transporter vGLUT1 at both 50 and campal Development, Cortical Development, Neuronal and Glial 70 d in culture, consistent with glutamatergic synapses differen- Differentiation, and Synaptic Transmission, and most of these tiation (Fig. S4 I–K). By day 70, astroglial cells differentiated included up-regulated genes. In contrast, the Cancer category throughout the culture, as demonstrated by the presence of included predominantly down-regulated genes with known in- GFAP/S100β immunostained cells, particularly in the outer re- volvement in the control of cell cycle, suggesting that the ma- gion (Fig. S2 D–I). jority of the neuronal cells at day 50 are likely already postmitotic (Dataset S2). Virtually all terms in the Cellular Assembly and Global Gene Expression Analysis Reveals the Implementation of a Organization category involved neural cell functions, such as Forebrain Neuronal Gene Expression Program. To ascertain whether growth and fasciculation of neurites and synapse formation these hiPSC-derived structures underwent a region-specific dif- (Dataset S2). Significant Neurological Disease associations in- NEUROSCIENCE ferentiation program, we assessed their gene expression profiles cluded genes implicated in disorders of the developing nervous by microarrays. Whole-genome expression data analyzed by the system, such as schizophrenia, and neurodegenerative disorders, Pluritest demonstrated progressively lower pluripotency scores such as Alzheimer’s and Huntington disease (Dataset S2). and higher novelty scores, suggesting ongoing neuronal differ- Consistent with the above characterization, the top DAVID entiation (Fig. S1D). Gene expression was compared between gene ontology clusters were Differentiation/Morphogenesis, 3D multilayered structures at day 50 in vitro and their corre- Synapse Formation, Forebrain/ Development, and Regula- sponding undifferentiated hiPSC lines PGP1-1 (one preparation) tion of Transcription (Dataset S3). Thus, differentially expressed and i03-01#9 (four preparations). Hierarchical clustering of the genes in hiPSC-derived structures at day 50 in vitro converged samples demonstrated a highly consistent pattern of gene ex- pression across all lines and preparations (Figs. S5 and S6). The toward telencephalic development, neurogenesis, neuronal pro- combined dataset, including all hiPSC lines and different 3D cess outgrowth, and synaptogenesis. preparations analyzed together, revealed that 3,363 genes were Comparison of gene expression in our hiPSC-derived cells with normal human neural progenitors (NHNP) derived from differentially expressed in day-50 multilayered structures com- ∼ – pared with the undifferentiated hiPSC stage (P value ≤0.05 to- early human embryo of 8 19 gestational weeks and differenti- fi gether with a fold change of ±2.0), of which 1,629 genes ated in vitro (23) revealed a signi cant similarity of gene ex- increased and 1,734 genes decreased in abundance. The top pression patterns. Among the 4,427 genes significantly changed down-regulated molecules (between −9,110- and −1,384-fold) in NHNP between time 0 and 4 wk of neuronal differentiation were LIN28A, DPPA4, RAB17, CDH1/E-CADHERIN, OCT4/ in vitro (23) and the 3,363 genes significantly changed between POU5F1, ZSCAN10, and GAL,reflecting genes commonly the hiPSC and 50 d in vitro stage in our model, 1,100 genes were − expressed in hESCs and hiPSCs and epithelial-specific genes in common. The P value for such overlap is <10 12 (calculated (boxed areas in Fig. S5A). In addition to OCT4, the expression of using numerical simulations and tail probability estimation).

Fig. 3. Regional and temporal specification of hiPSC-derived neuronal cells. (A) Fold differences in neuronal gene expression compared with undiffer- entiated hiPSCs for dorsally and ventrally enriched telencephalic genes. The gene lists were manually curated according to published data (24, 25, 35). Orange, genes that are statistically different at P < 0.05 and fold change ±2.0; gray, genes that do not fulfill the minimum P value criterion. (B) Number of developing human brain samples in Kang et al. (4) correlating with the day-50 multilayered structures within 95% confidence interval of the maximal correlation. Human brain samples are displayed according to developmental stages (Upper) or brain regions (Lower). PCW, postconceptional week; M, months; Y, years; FC, frontal, PC, parietal, TC, tem- poral, and OC, occipital cerebral cortical wall; HIP, hippocampal anlage; AMY, amygdala; DIE, diencepha- lon; URL, upper rhombic lip.

Mariani et al. PNAS Early Edition | 3of6 Downloaded by guest on September 28, 2021 Regional Specificity of Multilayered Aggregates Is Most Consistent Table 1. Up-regulation of genes involved in cortical with a Dorsal Telencephalic Fate. To further understand whether morphogenesis our 3D multilayered structures were regionally specified as Gene symbol F-I P reflected by their pattern of gene expression, we took into ac- count published datasets of telencephalic gene expression pro- Dorsal telencephalon specification & patterning files (24, 25), literature data, and the Allen brain atlas of gene LHX2 7,477 0 expression database. Overall the expression pattern of the LEF1 2,016 0 hiPSC-derived multilayered structures at day 50 in vitro was PAX6 970 0.00006 more typical of dorsal forebrain than ventral forebrain (Fig. 3A) FEZF2 846 0 and excluded hindbrain and spinal cord fates. For instance, genes NR2F1/COUP-TF1 589 0.00667 that confer hindbrain and spinal cord fates, such as HOX genes, EMX2 243 0 GBX2, HB8, NKX2.2, OLIG2, /2, ZIC1/2, and GCM1, and NEUROG2 180 0 fi transcripts speci c of development, such TBX21, WNT3A 52.1 0.00223 were not expressed or down-regulated (Fig. 3A and Dataset S1). NR2E1/TLX 43.6 0 Among the top 10 genes with the highest degree of up-regulation GAS1 fi 38.8 0 in the day-50 neural structures were those involved in speci - ZEB2/ SIP-1/ ZFHX1B 30.8 0.00631 cation and patterning of the mammalian , including EMX1 25.9 0 LHX2, LEF1, TBR2, PAX6, FEZF2, and RELN (Fig. 3A, Table GLI3 16.4 0 1, and Fig. S5B, boxed area). Very high levels of up-regulation AHI1 6.80 0.00103 were also noted for COUP-TF1, EMX2, and EMX1 and for other BBS1 2.66 0.04026 genes involved in cortical architecture and layer identity, such as BBS2 TBR1, CTIP2, CUX1, CUX2, POU2F2/BRN2, and TLX (Table 1). 2.56 0.03462 Some genes involved in GABAergic differentiation were also Cortical progenitor proliferation ID4 expressed (including MASH1 and ARX), but the basal telen- 120 0 ASTN1 cephalon differentiation program did not seem to be fully 77.6 0.00153 implemented (Fig. 3A and Fig. S3). This observation is intriguing, FOXG1 29.1 0 given previous findings suggesting that a lineage of human neo- Cortical neuron migration and layer specification cortical GABAergic progenitors expressing MASH1 originates EOMES/TBR2 1,093 0.00001 from the dorsal forebrain (26). RELN 834 0 FEZF2 846 0 Transcription Factors Expression Patterns Suggest Dynamic Acquisition DCX 297 0 of Cortical Layer Identities. Using a panel of antibodies directed to TBR1 73.0 0.04796 TFs that control cortical layer fates during normal cortical de- SEMA3C 61.4 0.00223 velopment (22, 27–29), we next investigated whether the imple- NR2E1 (TLX) 43.6 0 mentation of the neocortical transcriptional program resulted NTSR1 49.9 0.00584 in the specification of cortical excitatory neuron subtypes. We BCL11B/CTIP2 35.5 0.04262 + found that MAP2 neurons coexpressed TFs that specify a wide POU3F2 /BRN2 16.0 0.02681 spectrum of upper and lower cortical layer identity. These included CDK5R1 14.9 0.00077 TBR1, CTIP2, TLE4 (cortical layers V–VI) and BRN2 and RORB 13.0 0.00209 SATB2 (cortical layers II–IV) (Fig. 4). The proportion of neurons ENC1 12.2 0 forlowerlayerswas30–40%, whereas that for upper layers was 10– EFNB2 11.8 0 20% (Fig. 4S). Double-staining experiments showed that TBR1 and SSTR2 9.90 0.00006 TLE4 (layers V–VI) colocalized with CTIP2 (layer V) in many GPRIN1 6.02 0.00005 neurons, but cells displayed predominant expression of either one FYN 3.34 0.00606 – – – factor or another (Fig. 4 C F and J M and Fig. S4 A D). In contrast, PAFAH1B1 4.14 0 cells of lower layers (TBR1, CTIP2, and TLE4) showed more CUX2 3.94 0.00042 segregated pattern of expression from those of upper layers CUX1 – – – 3.80 0.00282 (SATB2 and BRN2) (Fig. 4 G I, N R,andS and Fig. S4 E H and EPHB1 3.62 0.00073 L). In addition to SATB2 and BRN2 , a number of TFs DPF1 3.53 0.03704 required for cortical upper layer formation were up-regulated at the mRNA level: CUX1 and CUX2 (fourfold), BRN2 (14-fold) (Fig. 3A Manually curated list of genes significantly up-regulated (P values ≤0.05) and Table 1), PAX6 (970-fold), and TLX (43-fold) (Table 1). No in day-50 multilayered structures compared with undifferentiated hiPSC in- appreciable differences in neuronal differentiation and sub- volved in mammalian neocortical development and/or expressed in a layer- structural organization were observed between the two iPSC lines specific manner in the human neocortex [see supplementary table 14 in (compare Fig. 4 and Fig. S4) and concentrations of FGF2 (Fig. 4S). Kang et al. (4)]. F-I, fold increase in gene expression.

Correspondence with Patterns of Gene Expression in Developing Human Brain. To verify the conclusion that the mRNA and pro- human cerebral cortex at 4–10 postconceptional weeks (PCW), with tein expression patterns of our hiPSC-derived multilayered a prevalence for 8–10 PCW and frontal regions (Fig. 3B). structures at day 50 are typical of the human dorsal forebrain, we performed a pairwise correlation analysis (details in SI Materials Discussion and Methods) between the global gene expression levels of our We demonstrate that under conditions that mimic neural in- samples (including the hiPSC and the day-50 multilayered duction of early forebrain, aggregates of hiPSCs segregate into structures) and 1,340 postmortem human tissue samples at 15 discrete layers containing radial glia, neuronal progenitors, and different developmental stages, comprising the cerebellar cortex, early neurons that enact a transcriptional program specifying ini- mediodorsal nucleus of the thalamus, striatum, amygdala, hip- tial stages of embryonic human dorsal telencephalic (pallial) pocampus, and 11 areas of the neocortex (4). The day-50 mul- development. The neurons form morphologically identifiable tilayered aggregates from the i03-01#9 line showed the highest synapses and exhibit protein and gene expression profiles typical of correlation coefficients with the human postmortem dataset, and mammalian excitatory cortical neurons encompassing both lower correlated only (as estimated by the 95% confidence interval) with and upper layer fates, as well as GABAergic neurons of presumed

4of6 | www.pnas.org/cgi/doi/10.1073/pnas.1202944109 Mariani et al. Downloaded by guest on September 28, 2021 (LHX2, FOXG1, PAX6, and EMX2) that are fundamentally re- quired for the formation of the six-layered dorsal pallium by suppressing dorsomedial and ventral telencephalic fates (30–32). These 3D structures also express many other TFs (FEZF2, BRN2, TBR1, CTIP2, and CUX1) and extracellular signaling molecules (RELN and EFNB2) known to direct cortical layer formation and identities. At the protein level, TFs known to specify lower-layer identity (i.e., TBR1, CTIP2, and TLE-4) and upper-layer identity (i.e., SATB2 and BRN2) were expressed in a mostly nonoverlapping fashion by developing neurons. Together, our combined morphological, immunocytochemical, and tran- scriptional analyses reveal that it is possible to recapitulate initial stages of development of the human dorsal telencephalon from hiPSCs and that laminar segregation of progenitors may be crucial for directing advanced features of cortical neuron differentiation. Although these data demonstrate dynamic acquisition of correct cortical neuron subtype identities within the 3D struc- tures, cortical layer architecture was not fully recapitulated. For example, upper-layer neurons in the human cortex outnumber those of lower layers, whereas in our 3D structures at day 50 the proportions of lower-layer neurons were higher than those of upper layers, likely owing to the early stage of our preparation. Indeed, at day 50 in vitro the 3D structures displayed a gene expression profile typical of the dorsal telencephalon at an early stage, including FOXG1, which controls the specification of the neocortex from archicortex (33, 34), LHX2, which specifies the dorsal pallium as distinct from the cortical hem (31), and GLI3, which specifies dorsal from ventral telencephalon (35). Other fl highly expressed TFs re ect the acquisition of rostral (frontal) NEUROSCIENCE cortical area identities [i.e., PAX6 (32, 36)], which agrees with the high correlation found between gene expression in our 3D struc- tures and the human frontal cortical primordium (see below). Previous work (37, 38) revealed that neuronal progenitors derived from mouse ES cells seem to be intrinsically able to produce cortical neurons in vitro, recapitulating the normal sequence of cortical layer fates. Differently, dissociated pro- genitors derived from human ESC/hiPSC by at least some of the recent embryoid bodies-based methods (13) do not permit an advanced degree of region-specific gene expression and cell differentiation. Although Shi et al. (39) were able to achieve cortical neuron differentiation from hESC/hiPSC cultures us- ing a dissociated culture system similar to that of Gaspard Fig. 4. Specification of cortical layer fates in hiPSC-derived neurons. (A) et al. (38), their protocol involves the addition of retinoic acid, Scheme of TF directing cortical layer fates. (B–R) Immunostaining for TFs in a molecule that antagonizes forebrain development (40), and it sections from 3D structures at day 50 in vitro. (B) Double immunostaining for will be eventually important to show that global gene expres- MAP2 and CTIP2, showing CTIP2 nuclear localization in MAP2+ cells; (C–F) sion analysis is consistent with the cellular phenotype. Classic TBR1 and CTIP2; (G–I) TBR1 and SATB2; (J–M) TLE4 and CTIP2; (N–P) TLE4 and embryological and genetic work has shown that early forebrain SATB2; and (Q and R) TBR1 and BRN2, showing differences in relative specification occurs via production of Lefty1, Dickkopf, and colocalization of these TFs in differentiating neurons. White arrowheads, Cerberus in the anterior visceral endoderm, which inhibit the cells stained for a single TF; white arrows, cells that colocalize two factors. β β Q BMP, TGF- /activin/nodal, and Wnt/ -catenin pathways at the Broken line in delimits the radial glial layer where BRN2 is also expressed beginning of gastrulation (9). Following the protocol of Eiraku but was excluded from quantification. (S) Stereological quantification of TF expression in 3D structures cultured with different amounts of FGF2 for the et al. (16), we have recapitulated this triple inhibition in vitro, first 18 d. DAPI is in blue. (Scale bars, 200 μminB and G; 100 μminC–F and J; and we have added moderate concentrations of FGF2, a mol- 25 μminH and I;20μminK–P.) ecule that synergizes with the early telencephalic inducers (12, 41). Indeed, we show not only a concerted up-regulation of dorsal telencephalic genes in our preparation but also the dorsal telencephalic identity. Comparison with gene expression absence of molecules that are up-regulated in caudal regions datasets from the human brain revealed that the transcriptome of of the CNS. Global gene expression profiles have not been previously the hiPSC-derived multilayered 3D structures reflects that of the – reported in either mESC- or hESC-derived SFEBq cultures, and human cerebral cortex at 8 10 wk after conception. hence the ability of this 3D culture systems to truly recapitulate The hiPSC aggregates recapitulate in culture the in vivo cyto- telencephalic neural differentiation programs had not been architecture that includes radial glial cells expressing neural pro- deciphered. The global comparison between transcriptomes of genitor proteins, intermediate progenitors, and maturing neurons. all our samples and those obtained from the postmortem human This spontaneous segregation also occurs in SFEBq cultures de- brain at several stages of development revealed a striking cor- rived from mESC and hESC (16) but had not been previously respondence in gene expression between our day-50 3D struc- shown for hiPSC. We demonstrate that the hiPSC-derived radial tures and the cerebral cortical wall at 8–10 wk after conception, glia express, like human radial glia, GFAP; the intermediate pro- particularly the frontal cortical primordium. This stage is char- genitors express TBR2, and the neurons express MAP2 and a acterized by formation of deep-layer cortical neurons and of the variety of TFs specifying cortical neuron identities. The multi- SVZ, as well as axonal extension through the major commissures layered structures at day 50 in vitro expressed the four TFs and descending pathways (4). These processes are reflected in

Mariani et al. PNAS Early Edition | 5of6 Downloaded by guest on September 28, 2021 our 3D structures by significant enrichment in genes with functions Materials and Methods related to cortical neurogenesis, growth and axon guidance; hiPSC and Neuronal Differentiation. Details of the protocol for generating and genes expressed in the SVZ (TBR2, TLX, CUX1/2); as well genes characterizing the hiPSC are listed in SI Materials and Methods. Three-dimensional involved in the establishment of laminar structures such as the cell aggregates known as SFEBq were generated according to Eiraku et al. 2008 dorsal pallium. Thus, the dynamic changes in transcriptional (16), with some modifications as described in SI Materials and Methods. profile that the hiPSC undergo as they differentiate under these conditions reflect those present in the early stages of human Biostatistics. We performed pairwise comparisons between the global gene dorsal pallium development in vivo. expression levels of all our samples (undifferentiated hiPSCs and day-50 We also found ultrastructural, transcriptional, and molecular multilayered structures) and all of the samples of Kang et al. (4) and esti- evidence for early synapse formation, including components of mated the Spearman correlation coefficients for all of the comparisons. We synaptic vesicles, receptors, and transporters, then selected the sample in Kang et al. (4) showing the highest correlation concomitantly with evidence of astroglial differentiation. This coefficient with one of our samples, estimated its 95% confidence interval, may represent either an aspect of neural development that is and included any other sample in the postmortem human tissue dataset fl particularly accelerated in vitro, or it may simply re ect the pe- whose correlation coefficient would fall within this interval (SI Materials culiarities of human cortical development, which is characterized and Methods). by early synapses of cortical neurons with the subplate (42, 43), – which appear as early as 10 11 PCW. ACKNOWLEDGMENTS. We thank Arif Kocabas for expert technical help, Anita In summary, we were able to generate 3D self-organized Huttner for preparation of fibroblasts from the skin biopsy specimen, In-Hyun structures from human iPSCs that recapitulate the program of Park for advice in the characterization of iPSC lines and the gift of the iPSC early dorsal telencephalic development in humans. Using patient- PGP1-1, and Stephen A. Duncan for the gift of the K3 hiPSC line. We are grateful specific iPSCs this model may offer novel insights into the patho- to Nenad Sestan and Feng Cheng for help with the comparison with the human brain datasets and to members of the F.M.V. laboratory and the Program in physiology of a large variety of disorders presenting primarily with Neurodevelopment and Regeneration for insightful discussions. We are cortical dysfunction, including numerous neurodevelopmental and supported by Grants MH087879, MH089176 and DK006850 from the National late-onset neurodegenerative conditions. Institutes of Health, by the State of Connecticut, and by the Simons Foundation.

1. Rakic P (2009) of the neocortex: A perspective from developmental biology. 23. Konopka G, et al. (2012) Modeling the functional genomics of using human Nat Rev Neurosci 10:724–735. neurons. Mol Psychiatry 17:202–214. 2. Geschwind DH (2011) Neurodevelopmental disorders: Hope for a new beginning. Curr 24. Long JE, Cobos I, Potter GB, Rubenstein JL (2009) Dlx1&2 and Mash1 transcription Opin Neurol 24:95–97. factors control MGE and CGE patterning and differentiation through parallel and 3. Johnson MB, et al. (2009) Functional and evolutionary insights into human brain overlapping pathways. Cereb Cortex 19(Suppl 1):i96–i106. development through global transcriptome analysis. Neuron 62:494–509. 25. Batista-Brito R, Machold R, Klein C, Fishell G (2008) Gene expression in cortical in- 4. Kang HJ, et al. (2011) Spatio-temporal transcriptome of the human brain. Nature 478: terneuron precursors is prescient of their mature function. Cereb Cortex 18:2306– 483–489. 2317. 5. Brennand KJ, et al. (2011) Modelling schizophrenia using human induced pluripotent 26. Letinic K, Zoncu R, Rakic P (2002) Origin of GABAergic neurons in the human neo- stem cells. Nature 473:221–225. cortex. Nature 417:645–649. 6. Marchetto MC, et al. (2010) A model for neural development and treatment of Rett 27. Molnár Z, et al. (2006) Comparative aspects of cerebral cortical development. Eur J Neurosci – syndrome using human induced pluripotent stem cells. Cell 143:527–539. 23:921 934. fi 7. Pasxca SP, et al. (2011) Using iPSC-derived neurons to uncover cellular phenotypes 28. Molyneaux BJ, et al. (2009) Novel subtype-speci c genes identify distinct sub- J Neurosci – associated with Timothy syndrome. Nat Med 17:1657–1662. populations of callosal projection neurons. 29:12343 12354. 8. Tam PP (1989) Regionalisation of the mouse embryonic ectoderm: Allocation of 29. Britanova O, et al. (2008) Satb2 is a postmitotic determinant for upper-layer neuron fi Neuron – prospective ectodermal tissues during gastrulation. Development 107:55–67. speci cation in the neocortex. 57:378 392. ’ fi 9. Wilson SW, Rubenstein JL (2000) Induction and dorsoventral patterning of the tel- 30. Chou SJ, Perez-Garcia CG, Kroll TT, O Leary DD (2009) Lhx2 speci es regional fate in Nat Neurosci encephalon. Neuron 28:641–651. Emx1 lineage of telencephalic progenitors generating cerebral cortex. – 10. Rubenstein JLR, Martinez S, Shimamura K, Puelles L (1994) The embryonic vertebrate 12:1381 1389. 31. Mangale VS, et al. (2008) Lhx2 selector activity specifies cortical identity and sup- forebrain: The prosomeric model. Science 266:578–580. presses hippocampal organizer fate. Science 319:304–309. 11. Rhinn M, Picker A, Brand M (2006) Global and local mechanisms of forebrain and 32. Muzio L, et al. (2002) Emx2 and Pax6 control regionalization of the pre-neuronogenic midbrain patterning. Curr Opin Neurobiol 16:5–12. cortical primordium. Cereb Cortex 12:129–139. 12. Levine AJ, Brivanlou AH (2007) Proposal of a model of mammalian neural induction. 33. Hanashima C, Li SC, Shen L, Lai E, Fishell G (2004) Foxg1 suppresses early cortical cell Dev Biol 308:247–256. fate. Science 303:56–59. 13. Kim JE, et al. (2011) Investigating synapse formation and function using human 34. Muzio L, Mallamaci A (2005) Foxg1 confines Cajal-Retzius neuronogenesis and hip- pluripotent stem cell-derived neurons. Proc Natl Acad Sci USA 108:3005–3010. pocampal morphogenesis to the dorsomedial pallium. J Neurosci 25:4435–4441. 14. Hu BY, et al. (2010) Neural differentiation of human induced pluripotent stem cells 35. Rallu M, et al. (2002) Dorsoventral patterning is established in the telencephalon of follows developmental principles but with variable potency. Proc Natl Acad Sci USA mutants lacking both Gli3 and Hedgehog signaling. Development 129:4963–4974. 107:4335–4340. 36. Bishop KM, Rubenstein JL, O’Leary DD (2002) Distinct actions of Emx1, Emx2, and 15. Hansen DV, Rubenstein JL, Kriegstein AR (2011) Deriving excitatory neurons of the Pax6 in regulating the specification of areas in the developing neocortex. J Neurosci Neuron – neocortex from pluripotent stem cells. 70:645 660. 22:7627–7638. 16. Eiraku M, et al. (2008) Self-organized formation of polarized cortical tissues from ESCs 37. Shen Q, et al. (2006) The timing of cortical neurogenesis is encoded within lineages of Cell Stem Cell – and its active manipulation by extrinsic signals. 3:519 532. individual progenitor cells. Nat Neurosci 9:743–751. 17. Ball MP, et al. (2009) Targeted and genome-scale strategies reveal gene-body 38. Gaspard N, et al. (2008) An intrinsic mechanism of corticogenesis from embryonic Nat Biotechnol – methylation signatures in human cells. 27:361 368. stem cells. Nature 455:351–357. fi 18. Takahashi K, et al. (2007) Induction of pluripotent stem cells from adult human - 39. Shi Y, Kirwan P, Smith J, Robinson HP, Livesey FJ (2012) Human cerebral cortex de- fi Cell – broblasts by de ned factors. 131:861 872. velopment from pluripotent stem cells to functional excitatory synapses. Nat Neurosci Nat 19. Müller FJ, et al. (2011) A bioinformatic assay for pluripotency in human cells. 15:477–486, S1. Methods 8:315–317. 40. Simeone A, et al. (1995) Retinoic acid induces stage-specific antero-posterior trans- 20. Englund C, et al. (2005) Pax6, Tbr2, and Tbr1 are expressed sequentially by radial glia, formation of rostral central nervous system. Mech Dev 51:83–98. intermediate progenitor cells, and postmitotic neurons in developing neocortex. 41. Stern CD (2005) Neural induction: Old problem, new findings, yet more questions. J Neurosci 25:247–251. Development 132:2007–2021. 21. Hevner RF, et al. (2001) Tbr1 regulates differentiation of the preplate and layer 6. 42. Allendoerfer KL, Shatz CJ (1994) The subplate, a transient neocortical structure: Its Neuron 29:353–366. role in the development of connections between thalamus and cortex. Annu Rev 22. Hevner RF (2007) Layer-specific markers as probes for neuron type identity in human Neurosci 17:185–218. neocortex and malformations of cortical development. J Neuropathol Exp Neurol 66: 43. Kostovic I, Judas M (2010) The development of the subplate and thalamocortical 101–109. connections in the human foetal brain. Acta Paediatr 99:1119–1127.

6of6 | www.pnas.org/cgi/doi/10.1073/pnas.1202944109 Mariani et al. Downloaded by guest on September 28, 2021