University of Southern California Dissertations and Theses

Generation and long-term culture of human cerebellar organoids to study development and disease.

(USC Thesis Other)

Generation and long-term culture of human cerebellar organoids to study development and disease.

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Generation and long-term culture of human cerebellar organoids to
study development and disease.
By
Alexander Atamian
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(NEUROSCIENCE)
August 2024

ii
Acknowledgements
First and foremost, I would like to thank my mentor, Dr. Giorgia Quadrato for her constant support
and unwavering belief in my potential as a budding scientist. Her drive to advance the field of
brain organoid research fueled my own fire throughout my graduate career.
I would like to thank my family for their undying love and support throughout this process. Also,
I would like to thank my friends and lab mates for their constant emotional and intellectual support.
Ashley and JP, you were both my PhD colleagues and my emotional backbone throughout this
process.
Lastly, I would like to thank my committee members Justin Ichida, Michael Bonaguidi, Ksenia
Gndeva, and Andy McMahon. Their thought provoking questions and guidance helped clarify the
trajectory of my research. Specifically, I would like to thank Andy for putting aside time in his
extremely busy schedule in the summer of 2016 to meet with me and discuss my interest in
pursuing a PhD when I was just an undergraduate student about to embark on my scientific career
at University College London (UCL). Graduating from a PhD program with you on my committee
feels like a full circle moment and I will always appreciate that conversation we had 8 years ago.

iii
Table of Contents
Acknowledgements………………………………………………………………………………..ii
List of Tables……………………………………………………………………..……………….vi
List of Figures…………………………...………………………………………………………viii
Abstract……...…….…………………….………………………………..………………………ix
Chapter 1: Introduction…………………………………………………..………………………..1
1 Cerebellar Development……………………………………..…………………………..1
1.1 The cerebellum: an overview……………………..…………………………...1
1.2 In vivo/vitro signaling cascade toward cerebellar regional identity…………..2
1.3 Cerebellar cell types…………………………………………………………...5
2 Organoids………………………………………………………………………………..8
2.1 Organoids overview…………….…………………………………………..…8
2.2 Brain organoids………….…………………………………………………….9
2.3 Applications in modelling complex brain disorders………….………………13
Chapter 2: Generation and long-term culture of human cerebellar organoids from pluripotent
stem cells…………………………………………………………………………………………15
1 Abstract……………………………………………………………………..………….15
2 Introduction……………………………………………………………………….……17
2.1 Development of the protocol…………………………………………………17
2.2 Establishing cerebellar features………………………………………………17
2.3 Experimental design……………………………………………………….…18
2.4 Overview of the protocol……………………………………………………..19
3 Anticipated Results……………………………………………………………….…….19
4 Discussion……………………………………………………………………..……….21
5 Figures/Tables……………………………………………………………………….....24
6 Materials………………………………………………………………………………..38
6.1 Reagents……………………………………………………………………...38
6.1.1 Cells………………………………………………………………...38
6.1.2 Growth media and supplements……………………………………38
6.1.3 Cell culture reagents………………………………………………..39
6.1.4 Immunostaining and cryosectioning reagents………………………40
6.2 Equipment……………………………………………………………………41
6.3 Reagents Setup……………………………………………………………….43
6.3.1 Feeder-independent hESC and iPSC lines…………………………43
6.3.2 Reconstituting growth factors/morphogens/small molecules………43
6.3.3 Aliquoting Matrigel………………………………………………...44
6.3.4 hiPSC and hESC medium………………………………………..…44
6.3.5 gfCDM+i medium……………………………………………….…44
6.3.6 CerDM1 medium…………………………………………………...45
6.3.7 CerDM2 medium…………………………………………………...45
6.3.8 CerDM3 medium…………………………………………………...46

iv
6.3.9 CerDM4 medium…………………………………………………...46
6.3.10 BrainPhys medium………………………………………………..47
6.3.11 K-gluconate internal solution……………………………………..47
6.3.12 Toxins and drugs preparation……………………………………...48
6.3.13 Papain dissociation preparation…………………………………...48
6.4 Equipment Setup…………………………………………………………..…49
6.4.1 Spinning Bioreactor………………………………………………...49
6.4.2 Orbital Shaker………………………………………………………49
6.4.3 2-Photon calcium imaging………………………………………….49
6.4.4 Patch Rig………………………………………………………...…50
7 Methods…………..………………………………………………………………...…..50
7.1 Making embryoid bodies (EBs): timing 1-2 hrs (Day 0) ……………………50
7.2 Feeding EBs and initiation of germ layer differentiation: timing 4 days
(Day 0-4)……….………………………………………………………………...51
7.3 Induction of isthmic organizing region: timing 11 days (Day 4-16) …………51
7.4 Shaking culture of expanding neuroepithelial buds: timing 13 days
(Day16-29)……………………………………………………………………….52
7.5 Growth and maturation of cerebellar tissue: timing over 1 year
(Day 30 - >1yr)..……….…………………………………………………………53
7.5.1 Option A: Cryosectioning and immunostaining organoids: timing 2-3 d…...53
7.5.2 Option B: Dissociating organoids into single cells: timing 2-3 hr…………..55
7.5.3 Option C: Calcium imaging of whole organoids: timing 7-10 d…………….56
7.5.4 Option D: Patch-clamp recording of whole organoids: timing 7-10 d………58
8 Timing…..……………………………………………………………………………...59
Chapter 3: Human cerebellar organoids with functional Purkinje cells………………………….60
1 Abstract.….….……………………………………………………………….………...61
2 Introduction…………………………………………………………………………….62
3 Results………………………………………………………………………………….63
3.1 Human cerebellar organoids (hCerO) reproducibly generate the
cellular diversity of the human cerebellum……………...………………………..63
3.2 hCerOs display organized laminar layering reminiscent of the EGL and
PCL of the developing cerebellar anlage…………………………………………67
3.3 hCerOs display functionally mature network activity in long-term
cultures, resembling patterns of in vivo cerebellar circuits…………………….....68
3.4 Functionally mature human Purkinje neurons develop within long-term
culture of hCerOs………………………………………………………………...70
4 Discussion……………………………………………………………………………...71
5 Figures………………………………………………………………………………….73
5.1Supplementary Figures/Tables………………………………………………..81
6 Materials and Methods……………………………………………...……………….....97
6.1 Pluripotent stem cell culture………..…………………………….…..………97
6.2 Organoid differentiation……………………………………………………...97
6.3 Quantitative PCR…………………………………………………………......98
6.4 Immunohistochemistry…………………………………………………….....99
6.5 Whole organoid staining……………………………………………………...99

v
6.6 Microscopy and image analysis………………………………………………99
6.7 Calcium imaging……………………………………………………………..99
6.8 Optogenetics and 2-Photon imaging………………………………………...100
6.9 Whole-cell patch-clamp recording………………………………………….101
6.10 Dissociation of cerebellar organoids and single-cell RNA-seq……………102
6.11Single-cell RNA-seq data analysis………………………………………....102
6.12 Detailed explanation of data integration and analysis…………………..…102
6.13 SeqFISH (Spatial Transcriptomics)………………………………………..103
6.14 Single-cell pre-processing (Spatial Transcriptomics)………………..…….103
6.15 Identification of Purkinje Cells (Spatial Transcriptomics)………….…..….103
6.16 Differential expression between Purkinje and non-Purkinje cells
(Spatial Transcriptomics)……………...…………………….………………….104
6.17 Supplementary Text……………………………………………………….104
Chapter 4: Uncovering the function of autism spectrum disorder-associated gene
SYNGAP1 in cerebellar development using human cerebellar organoids (hCerOs)…………...106
1 Abstract……………………………………………………………………………….107
2 Introduction…………………………………………………………………………...107
3 Results………………………………………………………………………………...109
3.1 Generation and characterization of haploinsufficient SYNGAP1
(Patientp.Q503X) and isogenic control (PatientCorrected) cerebellum………………..109
3.2 SYNGAP1 haploinsufficiency increases maturation of granule cells
and Purkinje cells. ……..………………………………………………………..110
4 Discussion…………………………………………………………………………….111
5 Figures/Tables………………………………………………………………………...113
6 Materials and Methods………………………………………………………………..117
6.1 Human iPSC line generation………………………………………………...117
6.1.1 Patientp.Q503Xcell line ………………………………..…………117
6.1.2 PatientCorrected cell line …………………………………………117
6.2 Viral Infection of organoids …………...……………………………………118
6.3 Immunocytochemistry………...…………………………………………….118
6.4 Dissociation of hCerOs and single-cell RNA-seq……...….………………...118
6.5 single-cell RNA-seq data analysis……………....…………………………..119
6.6 Organoid size analysis………...…………………………………………….119
Chapter 5: Future Directions……………………………………………………………………120
1 Introduction…………………………………………………………………………...120
2 Lineage Tracing cerebellar cell types…………………………………………………120
3 Upgrading physiological relevance of hCerOs with additional cell types……………..122
4 Cerebellar disease modelling………………………………………………………….123
References………………………………………………………………………………...…….125

vi
List of Tables
Chapter 2 Tables
Table 1: Troubleshooting table…..……………………………………………………………....29
Table 2: Antibodies for tissue characterization…..…………………………………...………....35
Table 3: Major Checkpoints………….…………………………………………………….…....36
Chapter 3 Supplementary Tables
Table S1: DEGs for the identified clusters of organoid dataset………………………………….93
Table S2: DEGs for organoid cells for SVZ, VZ, and IZ clusters……………………………….93
Table S3: Purkinje cell vs other cell type identification and counts in 6-month-old sample
per ROI. …………………………………………………………………………………..……...93
Table S4: Purkinje cell vs other cell enrichment test statistics…………………………………..94
Table S5: Passive properties of the membrane for PCP2+ neurons recorded by whole-cell
patch clamp recordings. ……………..…………………………………………………………..95
Table S6: qPCR primers…………………………………………………………………………95
Table S7: Primary antibodies used for immunofluorescence……………………………………96
Chapter 4 Tables
Table 1: DEGs associated with ASD and intellectual disability found in GO terms list.……...116
Table 2: Primary antibodies used for immunofluorescence……………………………………116

vii
List of Figures
Chapter 1 Figures
Fig. 1. Proposed model of the signaling pathways controlling anterior-posterior patterning
in pluripotent stem cell-derived neural progenitors in vitro. ………………………………………4
Fig. 2. Timeline of mouse and human cerebellar development. ………………………………...…7
Fig. 3. Development-inspired strategies to recapitulate early human brain regionalization. …………….11
Fig. 4. Timeline of all papers that have used scRNA-seq to characterize hPSC derived brain
organoids. …………………………………………..……………………………………………12
Chapter 2 Figures
Fig. 1. hCerO differentiation timeline…………………………………………………………….24
Fig. 2. Morphological features of early hCerOs up to Day 16………………………………….....25
Fig. 3. Morphological and immunohistochemical features of Day 30+ hCerOs………………….26
Fig. 4. Morphological and immunohistochemical features of suboptimal outcomes……………..28
Chapter 3 Figures
Graphical Abstract……………………………………………………………………………….61
Fig. 1. Human cerebellar organoids (hCerOs) reproducibly generate the cellular diversity of
the human cerebellum…………………………………………………………..………………..73
Fig. 2. hCerOs display organized laminar layering reminiscent of the external granule cell
layer (EGL) and the Purkinje Cell layer (PCL). ……………………………………………..….75
Fig. 3. hCerOs display functionally mature network activity in long-term cultures,
resembling patterns of in vivo cerebellar circuits. ……………………………………………….77
Fig. 4. Functionally mature, human Purkinje neurons develop within long-term
culture of hCerOs……………………………………………………………………...…………79
Chapter 3 Supplementary Figures
fig. S1. Human cerebellar organoid protocol optimization….……………………………………81
fig. S2. Reproducible generation of cerebellar cells across cell lines……………………………83
fig. S3. Single cell RNA-seq analysis of 2-month-old organoids…………………………………84

viii
fig. S4. Top 100 Purkinje cell DEGs from human fetal dataset compared to
2-month-old hCerO….…………………………………………..……………………………….85
fig. S5. Identification of human specific rhombpic lip ventricular and subventricular subsets…...86
fig. S6. hCerO cell type enrichment in various diseases…………………………………………87
fig. S7. Reproducibility of functional hCerO across cell lines…………………………………....89
fig. S8. GCaMP parameters across cell lines and between organoids…………………………….91
fig. S9. Spatial transcriptomics of 6-month-old organoids………………………………………92
Chapter 4 Figures
Fig. 1. SYNGAP1 haploinsufficient hCerOs can generate inhibitory and excitatory
progenitors of the developing cerebellum…………………………………………….…………113
Fig. 2. SYNGAP1 haploinsufficient organoids exhibit accelerated maturation of
cerebellar neurons….……………………………...……………………………………………115

ix
Abstract
In this thesis, I will discuss a new protocol to obtain reproducible cerebellar organoids
derived from human pluripotent stem cells (hCerOs). When I first joined Giorgia’s laboratory as a
technician, I began my work trying to optimize a protocol that would allow me to generate
cerebellar organoid with the least amount of forebrain contamination. Long-term culture of this
optimized protocol yielded organoids that had a high degree of reproducibility in terms of its
cellular composition determined through single cell RNA sequencing and immunohistochemistry
(IHC) as well as reproducible functional outputs determined by neuronal calcium dynamics
(GCaMP recordings). Moreover, we were able to identify the human specific rhombic lip
subventricular zone from our 2-month-old single cell RNA seq data, which has recently been
identified as a human specific feature. Additionally, through whole cell patch clamping and spatial
transcriptomics, we determined that these organoids could generate Purkinje cells that are
transcriptomically and functionally similar to their in vivo counterparts after 6-months in culture.
This model system was therefore an ideal model to probe the cerebellum’s phenotype within a
SYNGAP1 happloinsufficient (SYNGAP1+/-) context. Through scRNAseq, we see a trend toward
a delayed maturation phenotype within our hCerOs opposite to what our lab has previously shown
with cortical projection neurons in cortical organoids.
In summary, with great help and support from my mentor Dr. Giorgia Quadrato and my coauthors, we have created a model system to study the human cerebellar ontogenesis. This model
can now be used to probe the emergence of cerebellar cell types and study various human
cerebellar disorders such as ataxia and medulloblastoma. The use of these hCerOs to study
SYNGAP1+/- also highlights the importance of having a system that can explore how
neurodevelopmental disorders present themselves in various brain regions to gain a wholistic
picture of disease pathogenesis, which in turn will aid in the emergence of better therapeutics.

1
CHAPTER 1
Background
1 CEREBELLAR DEVELOPMENT
1.1 THE CEREBELLUM: AN OVERVIEW
The development of the human cerebellum is quite more complex and protracted than most
mammals. Evolutionarily, the expansion of the cerebellum began in parallel lineages of apes, but
only rapidly increased in the great ape clade1
. This region of the brain, which is often endearingly
referred to as the “little brain” or “mini brain” contains more than half of all the neurons of the
entire cerebral cortex, which is a testament to its immense computational power.
The cerebellum has historically been attributed to controlling motor function, which is a concept
that arose in the early 19th century after observations of cerebellar damage in animal models 2
.
Additionally, the emphasis of the cerebellum’s function in motor control was originally identified
from foundational studies in cats and monkeys3,4. The first of these studies mapped the
cerebellum’s motor topography by stimulating motor areas of the cerebral cortex and analyzed the
downstream cerebellar discharge3
. Following these studies, neuroimaging and motor movement
studies conducted in humans confirmed these findings, which solidified the cerebellum’s role in
governing motor functions 5–7
. Cerebellar damage in patients is typically characterized by
impairments in movements of limbs such as ataxia of stance, gait irregularities, and distorted
rhythm8
.
In addition to its role in motor functions, emerging evidence over the years points toward additional
roles in other higher order cognitive functions and emotion in humans9
. One of the first instances
to show this experimentally was during the emergence of neuroimaging techniques such as
positron emission tomography (PET) in the 1980s in which the participants’ right lateral
cerebellum elicited a robust response to tasks that did not require any sensory or motor
computation10. Following that study, other neuroimaging studies have shown a similar “cognitive”
response in the cerebellum using PET and fMRI11–13. This additional role of the cerebellum was
only recently discovered due to the lack of monosynaptic connections between the cerebellum and

2
cerebral cortex, which renders conventional anterograde and retrograde tracing methods useless
due to their inability to cross the synapse.
Specific anterograde tracing technology allowed researchers to identify cortical projections that
terminate in the pons. The presence of these pontine-labelled neurons indicated that inputs to the
cerebellum deriving from the prefrontal cortex exist. However, this system does not specify where
in the cerebellar cortex these projections terminate14–16. Therefore, to resolve this issue and show
the projections of prefrontal cortical cognitive networks to specific regions of the cerebellum,
Strick and colleagues developed and incorporation transneuronal tracing techniques17,18. This
technique allowed for the virus to map polysynaptic connections (jump synapses) and directly
show neuronal initiation and termination sites within the brain thereby overcoming traditional viral
tracing techniques.
The mounting evidence over the past 40 years has forced the field of cerebellar biology to reframe
their approach when studying the cerebellum in the context of other more complex disorders
involving neurodevelopment and neurodegeneration. Therefore, to understand how the cerebellum
develops its complex circuitry leading to the acquisition of other functions besides motor control
in human, there must also be a deeper understanding of cerebellar ontogenesis and development.
1.2 IN VIVO/VITRO SIGNALING CASCADE TOWARD
CEREBELLAR REGIONAL IDENTITY
In the mammalian system, the pluripotent epiblast-derived neural ectoderm is responsible for the
development of the central nervous system (CNS)19. These cells originate from the inner cell mass
of the mammalian blastocyst. Initially, the most rostral region of the epiblast gives rise to both
neuroectodermal (NE) and non-neuroectodermal cells. Subsequently, neural cells undergo rostralcaudal patterning, leading to the formation of distinct regions in the developing CNS, including
forebrain, midbrain, hindbrain, and spinal cord.
The "activation-transformation model" suggests that the ectoderm, in vivo, receives an "activation"
signal, which neuralizes it and triggers the initiation of forebrain specification20,21. This is
mimicked in embryonic and induced pluripotent stem cell (ESC/iPSC) cultures through dual
SMAD inhibition (inhibition of TGFB and BMP pathways)22. Without this neuralization signal, in
vitro differentiation of stem cells may potentially result in the acquisition of a different germ layer

3
fate when exposed to various concentrations of BMP, FGF, and WNT agonists at Day 0. Although
this same neuralization signal may be achieved in vitro simply through TGFB inhibition, a “triple
inhibition” condition (inhibition of WNT, TGFB, and BMP pathways) is superior in inducing a
robust cortical identity in 3D human pluripotent stem cell aggregates23.
Following the “activation” signal, a "transformation" signal induces cascading signals that drive
the cells towards caudalization, resulting in the formation of various neuroectodermal regional
identities20,21. After the "activation" signal has been triggered, the primary transforming signal
recognized for inducing the caudalizing effect on the neural tube is commonly attributed to the
secretion of the wingless WNT family of ligands and fibroblast growth factors (FGFs) (Figure 1).
It is after the coordinated, dose-dependent signaling cascades that the isthmic organizer emerges,
which is characterized by a high level of FGF8 and flanked with OTX2+ mesencephalon and
GBX2+ rhombencephalon expressing cells at the rostral and caudal ends, respectively.

4
Figure 1. Proposed model of the signaling pathways controlling anterior-posterior
patterning in pluripotent stem cell-derived neural progenitors in vitro. Neural progenitors
induced by inhibition of TGFβ signaling in primed pluripotent cells (EpiSCs or hESCs) can be
steered to distinct anterior-posterior fates, marked by specific transcription factors, in different
culture conditions. Inhibition of both TGFβ and Wnt/β-catenin signaling allows specification of
rostral forebrain fates, marked by Foxg1 and Six3 expression. Activation of BMP, Wnt/β-catenin
and FGF signaling with appropriate timing and doses promotes caudal forebrain and/or midbrain
fates, marked by Irx3, En1/2, and Gbx2 expression. Posteriorization to hindbrain/spinal cord fates
(expressing Gbx2 and Hoxb1-9) can be achieved by high levels of Wnt/β-catenin, FGF2, and/or
RA signaling. The drawings show the main anterior-posterior subdivisions of the developing
mammalian nervous system in vivo. Fb forebrain, Mb midbrain, Hb hindbrain, sc spinal cord.
Adapted from Lupo et al 2014.24

5
1.3 CEREBELLAR CELL TYPES
In the developing neural tube, the rhombencephalon is divided into the rostral metencephalon and
the caudal myelencephalon. The metencephalon gives rise to the cerebellum and pons while the
myelencephalon gives rise to the spinal cord. It is the most rostral portion of the metencephalon,
the portion adjacent to the isthmic organizer also known as the first rhombomere (r1) from which
both the inhibitory and excitatory cerebellar cell types emerge. In mammals, the cerebellum
derives from the alar lamina (dorsal) portion of r1. The most dorsal portion of r1(ventral to the
roof plate) gives rise to the ATOH1+ Rhombic Lip (RL) while the ventrally located regions gives
rise to the PTF1A+ ventricular zone (VZ). Loss of either ATOH1 or PTF1A has been shown to
cause a complete loss of either glutamatergic or GABAergic neurons respectively25–27.
Additionally, misexpression of ATOH1 or PTF1A in the developing neural tube causes ectopic
generation of glutamatergic or GABAergic neurons, respectively28. The VZ produces GABAergic
neurons, including Purkinje Cells, molecular layer interneurons (MLI), and deep cerebellar nuclei
(iCN), while the more dorsally located RL gives rise to glutamatergic neurons, including granule
cells (GC), unipolar brush cells (UBC), and large deep cerebellar nuclei projection neurons
(eCN)25–27,29–31 (Figure 2).
In recent years, the use of single-cell RNA sequencing has allowed for characterization of
cerebellar cell types in the developing mouse cerebellum32–34. Pseudotime-based lineage analysis
in these studies has allowed for the discovery of novel, putative transcriptional regulators after
developmental trajectory reconstruction of both geminal zones of the developing cerebellum (VZ
and RL). Carter et al.32 also created Cell Seek, which is a web interface that allows for an
interactive way of analyzing mice cerebellar cell types between E10 – P10. These studies not only
characterize the transcriptional profile of various cell types, but also identify subtypes of Purkinje
cells previously unknown. Additionally, the use of single nucleus ATAC-sequencing
(snATACseq) has allowed for a deeper understanding of dynamics of the regulators of
transcriptional fates (CREs such as enhancer and promoter) at a cell type specific resolution35.
Advancements in spatial transcriptomics has also aided in the understanding of the spatial gene
expression patterns of cellular subtypes of the mouse cerebellar36.

6
Since these seminal studies using multi-omics approach to study mouse cerebellar development,
there has been an emergence of studies that have aimed to characterized the developing human
cerebellum37–39. The first of these series of publications that conduct longitudinal characterization
of cerebellar cell types using single nucleus RNA-sequencing (snRNAseq) was a study from the
Millen lab, which captured the molecular organization of the cerebellar anlage as well as
distinguishing its various cytoarchitectural regions include those that are unique to humans such
as the RLsvz37. A similar study incorporated the use of scRNAseq and spatial transcriptomics to
compare cerebellar development across various species. Interestingly, this study identified that the
development of the cerebellum in terms of the sequential generation of cell types is conserved
across species as seen in Figure 2 when comparing mice to human cerebellar development.
However, this paper identified an expansion of early-born Purkinje cell subtypes within the human
lineage when compared with the Purkinje cell’s developmental trajectory of other species38.

7
Figure 2. Timeline of mouse and human cerebellar development. Cerebellar neurogenesis in
the mouse and human is driven by three progenitor zones: two primary zones, namely the VZ and
RL, and one secondary zone, the EGL. Development of each cell type is indicated by a unique
colored line. In mice, cerebellar development takes place over a period of approximately 30–35
days, with a significant portion occurring in the postnatal period. In humans, cerebellar
development is protracted, taking place over 2–3 years, beginning approximately 30 days post-

8
conception, with all major developmental events taking place in utero. Peak progenitor
proliferation is indicated by thickened dashed lines. In both species, VZ- and RL-driven growth
precedes an EGL-driven increase in cerebellar volume and foliation. However, unlike in the
mouse, where RL presence is transient, in humans, the RL is spatiotemporally expanded and
promotes growth and maintenance of the posterior lobe throughout gestation. Abbreviations: e,
embryonic day; EGL, external granule layer; P, postnatal day; PC, Purkinje cell; pcw,
postconceptional week; PN, postnatal; RL, rhombic lip; SVZ, subventricular zone; VZ, ventricular
zone. Adapted from Haldipur et al. 2022.40
2 ORGANOIDS
2.1 ORGANOIDS OVERVIEW
Organoids are an in vitro, 3D, cell-based model system that replicates aspects of the intricate
architecture and complex functions of their corresponding in vivo tissues41. This system can be
dissected to understand steps in early human development that has not been easily accessible
previously. In addition to studying early human development, organoids can be manipulated as
well to probe human specific diseases that are not possible to study in other animal model systems.
A common feature of all organoids is that they can be derived from pluripotent or embryonic stem
cells (PSC/ESC). Therefore, studying the formation of organoids can provide valuable insight into
early organ generation, highlighting the value of the system to understand the mechanism of early
development in addition to their applications. However, organoids come with their own challenges
as they are still in their infancy compared to other more established animal and 2D model systems.
The key limitations in using organoids as a model system include the accumulation of debris within
the organoid/presence of a necrotic core, limited reproducibility within and between batches of
organoids, and inability to properly recapitulate the complex in vivo microenvironment composed
of diverse cell types of multiple lineages. As organoids grow and mature over time, a natural
process of replacing older cells with newer cells is an issue especially within a system that is highly
proliferative such as intestinal organoids42. This results in a buildup of debris within the organoid
and may cause other downstream issues due to their lingering presence. In other organoid systems
such as brain organoids, a major concern is the presence of a necrotic core due to nutrient and
oxygen perfusion limits based on size. As organoids reach certain limits in size, some researchers

9
have adopted cutting the organoids or applying more refined methods such as bioengineering
innervating tubules or vasculature to circumvent issues deriving from a necrotic core43,44.
Additionally, the variability that occurs within and between batches of organoids can potentially
cause a misinterpretation of observable phenotypes. For example, differences in the ratios of
inhibitory to excitatory neurons within a batch of brain organoids may lead to incorrect conclusions
about a biological readout. This has been slowly addressed over the years with the advent of singlecell transcriptomic and epigenomic profiling, which has allowed for a more standardized way of
determining the quality and validity of the organoid model45. Moreover, the lack of a complex
microenvironment as is seen within animal models may cause the system to be less biologically
relevant depending on the questions that is being asked about the system. To combat this, coculture methods are being implemented (such as the integration of microglia or endothelial cells
into brain organoids) to address the lack of complexity46.
Despite these shortcomings that are being addressed in more recent iterations of this model system,
the key advantages have shown to be crucial in pushing the boundaries of our understanding of
human specific development and disease. The major advantages of the human organoid system
include the ability to probe early human development in a more complex environment compared
to 2D cultures, to genetically engineering organoids to model human disease, and to create more
rapid, personalized in vitro system geared toward the generation of patient specific therapeutics.
The increased maturation within organoids compared to 2D model systems is mainly because
organoids can be cultured for longer periods of time without the physical limitations of a culture
plate. Additionally, the advent of CRISPR system has allowed unprecedented levels of genetic
manipulation to better understand human disease within a complex organoid system, which can
model aspects of human specific disease.
2.2 BRAIN ORGANOIDS
The development of the human brain is governed by very specific and highly regulated morphogen
gradients in addition to cell-cell and cell-extracellular matrix (ECM) interactions that determine
brain regionalization and cell fate. Over the course of gestation, these interactions regulated by
positive and negative feedback loops determine the organization of the brain and provide guidance
cues for cells to properly build the various subregions and interconnected network of the human
brain. To mimic this intricate interplay and spatio-temporal complexity of in vivo developmental

10
dynamics, a combination of different inhibitors of key signaling pathways have been used to direct
the differentiation of human pluripotent stem cells (hPSC) into human brain organoids.
Human brain organoids are small 3D, self-organizing, aggregates of human pluripotent stem cells
that resemble aspects of early brain development and have been used in recent year as a powerful
tool to model brain development and study the emergence and progression of disease in a human
system. Brain organoids can be generated either through a “self-pattered” or a “patterned”
approach. Both self-patterned and patterned organoids are initiated when dissociated PSCs are
aggregated into 3D spheres or embryoid bodies (EBs) to generate self-organizing 3D
neuroectodermal structures. However, given the high level of heterogeneity between and among
organoid batches using the self-patterned approach, many labs have turned to generating brain
region specific organoids through a patterned approach that takes inspiration from in vivo signaling
pathways46. This method generates more robust and reproducible brain region specific organoids
to study development and disease.
Self-patterned protocols utilize basic fibroblast growth factor (bFGF) and retinoic acid during
hPSC reaggregation to generate whole-brain organoids. These protocols can generate all regions
of the brain with high levels of variability. These protocols leverage the default bias of early
ectoderm to generate dorsal forebrain fates47. This has been achieved thanks to pioneering work
that led to the identification of factors, which efficiently induce progenitors toward a telencephalic
identity from PSCs22,48–50. Since then, different combinations of these factors have been used in
protocols for generating cortical identity46.
Patterned protocols have been shown to generate more precise regional identity and more
reproducible cellular diversity and has emerged as a widely used alternative to self-patterned
protocols. A common strategy employed at the beginning of a patterned protocols is the inhibition
of the SMAD pathway22. The SMAD family of proteins act as signal transducers downstream of
TGF-β superfamily ligand binding to promote the generation of epidermis over neural ectoderm.
Additionally, WNT inhibition during early stages of cortical organoid generation is often
employed to repress the mesodermal lineage and, in turn, promote the production of anterior
neuroectoderm51 while a WNT agonist can generate caudal neuroectoderm52 (Figure 3). Despite
patterning and culturing differences, brain organoids across protocols can develop neural-tube-like
neuroepithelial structures that follow a similar temporal developmental trajectory as in vivo. These

11
structures give rise to a diverse range of cell types as seen in the endogenous human tissue. More
specifically, cortical organoids are widely used because of their relevance to human
neurodevelopmental disorders, as defects in higher-order cognitive function have historically been
studied within this brain region.
Figure 3. Development-inspired strategies to recapitulate early human brain regionalization.
Morphogen gradients within the neural tube are critical in determining the rostro-caudal and dorsal-ventral
axis of the developing brain and contribute to the formation of discrete regionalization. Several key
signaling centers present in the neural tube including the Roof Plate (RP) and Floor Plate (FP) are a
prominent source of BMP (bone morphogenic protein), WNT, SHH (sonic hedgehog), RA (retinoic acid)
and FGFs (fibroblast growth factors). Patterning strategies have allowed for the establishment of protocols
resembling distinct brain regions including the dorsal and ventral forebrain, hippocampus, hypothalamus,
anterior pituitary, thalamus, midbrain, cerebellum, midbrain, spinal cord, and choroid plexus. Adapted from
Del Dosso et al. 202046
Up until the last few years, bulk sequencing has been the primary approach to obtain a lowresolution, average molecular RNA profile of a given sample. However, the advent of single cell
sequencing has revolutionized the study of the transcriptome by allowing researchers to analyze
the composition of a sample at a single cell resolution. This single cell approach has given

12
researchers insight to the diverse cellular heterogeneity that exists with an unprecedented level of
detail. Single-cell RNA sequencing (scRNAseq) is the current gold standard to characterize brain
region specific organoids across protocols by comparing it to the fetal transcriptomic databases as
a benchmark to uncover the similarities with the native tissue53. Since the first scRNAseq
experiment conducted on cortical organoids in 201554, there have been a host of alternative
approaches used to capture and characterize region-specific, brain organoids at a single cell
resolution with the 10X Genomics platform being the most used method45 (Figure 4). With this
approach, researchers were finally able to uncover not only the cellular composition and
heterogeneity within organoids, but also deconvolute the inferred lineage dynamics as cells
transition from a progenitor to a fully differentiated state through pseudotime ordering.
Figure 4. Timeline of papers that have used scRNA-seq to characterize hPSC derived brain
organoids from 2015-2020. Each colored bubble contains author, year of pub- lication, single cell
platform used, and number of cells characterized after passing quality control. SMART-seq (red),
DropSeq (blue), BD Resolve/Rhapsody microwell (purple), SMART-seq2 (orange), Chromium by 10X
Genomics (green), Fluidigm C1 (pink). Human Cortical Spheres (hCS); human Striatal Spheres (hSS);
human Cortical Organoids (hCOs); human Medial Ganglionic Emience Organoids (hMGEOs);
vascularized human Cortical Organoids (vhCOs).

13
2.3 APPLICATIONS IN MODELLING COMPLEX BRAIN
DISORDERS
The ability to create an all-human system to model human specific, complex brain disorders has
been a difficult task to a achieve for modern medicine. Disorders such as autism spectrum disorder
(ASD) are complex human disorders that are difficult to model in animals due to species
differences in structure and cellular composition. ASD has historically been characterized by
higher order cognitive function deficits with the primary focus on the cortical dysregulation. In
patients with ASD, comorbidities such as Intellectual Disability (ID), Developmental Delay (DD),
and Epilepsy is common, and symptoms can arise either early at birth or in the few years shortly
after birth and can have dramatic effects on a patients’ life. Although this disease is quite prevalent
in the world (~4% of the world’s population), there are very little interventions that exist due to
the lack of a suitable human model system. Human brain organoids offer an avenue to study this
complex brain disorder by providing a more complex system compared to the conventional 2D
culture systems and a platform to study human specific disease instead of the current gold standard
animal model systems.
To date, patterned human brain organoids are primarily used to study disorders affecting one
region of the brain such as the cerebral cortex. These reductionist models of a specific region of
the brain recapitulates key developmental stages in addition to generating the diversity of cell types
found in that region of the brain, which are susceptible to dysregulation in the context of ASD.
Additionally, brain organoids can be made patient specific while also being amenable to genetic
engineering, which makes personalized medicine within reach. This allows for the investigation
of human and patient specific disease mechanism and interrogate potential avenues for
understanding the phathophysiology of the mutation.
Despite the immense knowledge that can be gleaned from brain region specific organoids,
investigating local network connectivity requires the use of multiple organoids of various brain
regions. This will be necessary to study more complex brain disorders such as psychiatric disorders
or connectopathies. Recent studies have started to show the importance of connecting various brain
regions as organoids in the form of assembloids55. As microfluidics device improves the ability to
modify these culture methods to model long-range connections between organoids become more
physiologically relevant56. This is a unique approach as mix-and-match method of modeling brain

14
region connections is difficult to achieve within 2D culture systems compounded on top of the low
availability of post-mortem human tissue.

15
CHAPTER 2
Generation and long-term culture of human cerebellar organoids from
pluripotent stem cells
Alexander Atamian1,2, Marcella Birtele1,2, Negar Hosseini1,2, Giorgia Quadrato1,2*
Affiliations:
1
Department of Stem Cell Biology and Regenerative Medicine, Keck School of Medicine
University of Southern California; Los Angeles, CA, USA.
2
Eli and Edythe Broad CIRM Center for Regenerative Medicine and Stem Cell Research at
USC, Keck School of Medicine, University of Southern California; Los Angeles, CA, USA.
Author contributions:
A.A., M.B., and G.Q. conceived the experiments. A.A., M.B., and N.H. generated, cultured,
and characterize all organoids used in this study; G.Q. supervised all aspects of the project;
A.A., N.H., M.B., and G.Q. wrote the manuscript.
Under review in Nature Protocols
1 ABSTRACT
The advancement of research on human cerebellar development and diseases has been hindered
by the lack of a cell-based system that mirrors the cellular diversity and functional characteristics
of the human cerebellum. In this study, we introduce a human pluripotent stem cell (hPSC) derived
human cerebellar organoid (hCerO) model, which successfully replicates the cellular diversity of
the fetal cerebellum along with some of its distinct cytoarchitectural features. Our approach
involves patterning of hPSCs, resulting in the generation of both cerebellar excitatory and
inhibitory progenitor populations – specifically, the rhombic lip and ventricular zone progenitors,
respectively. This patterning strategy leads to the reproducible differentiation of the major neurons
of the cerebellum such as granule cells and Purkinje cells within just one month of culture. hCerOs
serve as platforms for molecular, cellular, and functional assays, including single-cell
transcriptomics, immunohistochemistry and investigations into calcium dynamics and
electrophysiological properties. Remarkably, the cultivation of hCerOs for up to 8 months enables
the healthy survival and maturation of Purkinje cells, which exhibit molecular and
electrophysiological features akin to their in vivo counterparts. Overall, our protocol generates and

16
allows for the long-term culture of all major cell types within the cerebellum. Consequently, this
significant advancement provides the developmental neurobiology field with a robust platform for
exploring both cerebellar development and diseases within an all-human system. This protocol can
be easily implemented by a technician with cell culture experience and take 1-2 months to
complete with an option for extended maturation over the course of several months.
2 INTRODUCTION
The emergence of human organoid model systems for studying brain development and disease has
ushered in a new era of understanding previously inaccessible aspects of early human brain
development. Given the limited availability of human tissue during these critical stages, organoids
provide a robust platform for delving into the complex genetics underlying numerous
neurodevelopmental disorders characterized by human-specific phenotypes and polygenic
etiology.
While organoids capable of simulating a forebrain-like regional identity have been extensively
established, characterized, and benchmarked against their endogenous counterparts44,57–65, offering
invaluable insights into forebrain development and disease66, similar efforts targeting the more
caudal regions of the brain have been relatively scarce. This disparity stems mainly from the
intrinsic bias of ectodermal aggregates towards generating telencephalic structures67, which
naturally renders telencephalic organoids more reproducible within and across hPSC lines.
However, optimization and thorough characterization of other types of organoids, including
cerebellar organoids, is highly needed, especially in light of mounting evidence highlighting the
pivotal role of the cerebellum in higher-order cognitive functions68,69. Interestingly, the substantial
evolutionary expansion of this brain region within the great ape clade suggests that the cerebellum
might play a role in the acquisition of human-specific traits70. Notably, it has been shown that the
human rhombic lip possesses a human-specific subventricular zone, which is not found in mice
and other primates37,39, and more recently, it has been shown that human Purkinje cells exhibit
unique developmental dynamics, marked by an expansion of early-born subtypes compared to
other species38. Therefore, hCerOs hold promise for enhancing our understanding of the unique
contributions of the human cerebellum to brain development, functionality, and disease.

17
In this article, we present a detailed method for generating hCerOs52, capable of recapitulating the
complex cellular diversity of the fetal cerebellum, organizing into laminar layers, and exhibiting
molecular and electrophysiological features characteristic of their in vivo counterparts.
2.1 DEVELOPMENT OF THE PROTOCOL
The methods below have been developed based on previous work identifying cues for the
specification of the isthmic organizer in vivo71–73 and in vitro74,75, as well as strategies to improve
long-term culture techniques using brain organoid protocols76. Our protocol involves three main
steps: (i) patterning hPSCs to resemble an isthmic organizer-like environment, (ii) expanding
inhibitory and excitatory cerebellar progenitor pools, and (iii) maturing cerebellar neural cell types.
2.2 ESTABLISHING CEREBELLAR FEATURES
Cerebellar development is initiated by a series of patterning cues that delineate the alar lamina at
the midbrain/hindbrain boundary, from which both inhibitory and excitatory populations of the
cerebellum arise.
To model the midbrain/hindbrain boundary in human pluripotent stem cell (hPSC) cultures, the
first step involves directing the cells towards neuroectoderm and subsequently caudalizing them.
Previous studies have demonstrated the rapid and efficient generation of neuroectoderm through
dual SMAD inhibition, which inhibits the TGFB and BMP pathways22. However, without
caudalization factors, hPSCs tend to adopt a forebrain-like identity, as observed in foundational
developmental biology studies67. Thus, it was necessary to expose the cells to various morphogens
to drive caudalization towards a hindbrain fate.
Critical morphogen gradients within the developing neural tube, regulating brain region
specification, include molecules governing the BMP, RA, WNT, and FGF pathways. Among
these, the WNT family of proteins has been identified as a principal dose-dependent caudalization
factor within the neural tube77,78. Hence, we apply a canonical WNT agonist (CHIR-99021) to
promote caudalization of neural progenitors. Additionally, to model the signaling cues from the
adjacent isthmic organizer at the midbrain/hindbrain boundary, critical for generating the most
rostral region of the hindbrain (rhombomere 1), we introduce FGF8 into our protocol.

18
The isthmic organizer, one of the “organizing centers," of the developing neural tube possesses a
defined genetic profile that specifies neighboring neuroectodermal identity along the dorsalventral and rostro-caudal axes via morphogenic regulation. FGF8, highly expressed at the
midbrain/hindbrain boundary, in conjunction with other signaling molecules such as WNT1,
delineates this boundary73,79. Loss-of-function experiments of FGF8 resulted in a complete loss of
the cerebellum and tectum80, underscoring its significance. Our protocol involved the specific use
of FGF8b, known to convert midbrain-fated cells into hindbrain identity71,72.
Unexpectedly, we have found that alterations in the formulations of the basal media without
modifying the small molecule and patterning factor cocktail can modulate the regional identity.
Indeed, without the switch to CerDM1 (cerebellar differentiation media 1) on Day 6 of
differentiation, organoids cultured in gfCDM+i (growth factor reduced media + insulin) for 16
days generated only excitatory neurons of the cerebellum.
After exposure to isthmic organizer-like conditions for 16 days, organoids are transferred to 10 cm
shaking culture conditions for further neuroepithelium expansion and maturation. By day 30, these
organoids began generating both inhibitory and excitatory neurons of the developing cerebellum,
derived from the established isthmic-like organizing center in vitro.
Finally, though not essential for generating the complete diversity of cerebellar cell types, we
found that the addition of the chemoattractant SDF1a, acting on the CXCR4 receptor between days
30-60 in culture, induce migration of rhombic lip progenitors towards the organoid periphery,
creating a transient laminar layered structure and overall improving cytoarchitecture within
hCerOs52.
2.3 EXPERIMENTAL DESIGN
We have differentiated hESC and hiPSC lines in feeder-independent conditions. hPSCs are
initially dissociated to single-cells and reaggregated in a low attachment V-bottom plate (Steps 1-
8); Fig. 1a. Although these plates are optimal for aggregating hPSCs, other plates such as the low
attachment U-bottom plates or AggreWells are also commercially available and have been used
for culture of other 3D aggregates but not tested with our protocol. The EBs are simultaneously
exposed to various morphogens that select for neuroectodermal specification and followed by
FGF8b to specify the midbrain-hindbrain boundary (Step 9-15). At this stage, the media
composition affects the cell types generated in these cultures. (Steps 11-15). Following the static,

19
suspension culture of aggregates in 96-well V-bottom plates, the EBs are then transferred to orbital
shaking condition or spinning bioreactors (Steps 16-20) to aid in oxygen and nutrient exchange
within the organoids for long term growth and maturation.
2.4 OVERVIEW OF THE PROTOCOL
The protocol initiates with the maintenance of pluripotent stem cells in feeder-independent
conditions (step 1). Following stem cell maintenance, the cells are detached from the plates,
dissociated, and aggregated in into small spheres (steps 2-8) and cultured in various media
conditions to (1) generate the isthmic organizer (steps 9-15), (2) expand the progenitors (steps 16-
18), and (3) mature the neuronal cell types (steps 19,20). Once generating hCerOs, downstream
assays including IHC, single cell dissociation, calcium imaging, and Patch-Clamp recordings are
performed for an in-depth molecular and functional characterization (Step 20, Option A-D).
3 ANTICIPATED RESULTS
This protocol aims to generate hCerOs that contain the cellular diversity of the fetal
cerebellum and can be cultured long-term to achieve functional maturation of inhibitory and
excitatory neuronal populations. This method is easy to implement and reproducible across various
hPSC lines.
Typically, within the first 10 days organoids grow consistently in shape and size to
show signs of neuroectodermal differentiation with the appearance of rosette-like structures within
the organoid around day 6 of differentiation (Fig. 2a). If the cells do not aggregate properly (Fig.
4a), the differentiation will yield suboptimal results. Once transferred to 10cm dishes at day 16,
the organoids may display slightly different morphologies by either appearing slightly smaller and
more translucent, large/smooth and more opaque, or bulbs of growing neuroepithelium around the
organoid giving it a bumpy shape (Fig. 2b). These morphologies are potentially all typical of
successful hCerOs deriving from various cell lines, however, it is essential to culture these
organoids out to day 30 in order to confirm the presence of progenitors that give rise to both
excitatory and inhibitory neurons.
At day 30, the presence of translucent rosette-like neuroepithelial structure may appear,
which is indicative of a successful differentiation (Fig. 3a). However, modifying the protocol by
maintaining the organoids in gfCDM instead of switching them to CerDM1 at day 6 changes the

20
way these organoids develop in terms of their continuous, translucent neuroepithelial morphology
(Fig. 3b) and their cell type composition (Fig. 3c,d). It is important to note that the typical
morphology of an optically translucent organoid to indicate a successful protocol may not apply
to all hCerOs. This is because some cell lines may not be as successful in forming rosette-like
structures given that consistent WNT administration has been shown to disrupt ventricle formation.
If the organoids seem large and opaque, it is important to first section and stain them to determine
if there are smaller rosettes forming within the organoid prior to terminating the experiment.
Additionally, the presence of translucent rosette structures alone will not be a good indicator of a
successful cerebellar differentiation as these tighter and more uniform translucent rosette structures
are more likely cortical organoids derived from suboptimal patterning (Fig. 4b,c)
The original protocol, which involves switching to CerDM1 at day 6 of differentiation
generates both progenitor cell types of the developing cerebellum (ATOH1 and KIRREL2) as well
as its respective neuronal progenitors (BARHL1 and SKOR2) (Fig. 3c). Maintaining the organoids
in gfCDM for the duration of patterning (day 0-16) biases the cells to only generate the more dorsal
granule cell progenitors of the developing cerebellar anlage that are ATOH1+ and BARHL1+
(Fig.3d). As these organoids grow to day 60, it is possible to start noticing that these large,
smoothe, and round structures also have a textured portion to it, which is often an indicator of an
emerging choroid plexus-like structure indicated by black arrows in Fig. 3e. This, however, may
vary from cell line to cell line as the development of this region in culture is not specifically
targeted. It is important to note that these choroid plexus-like structures should not make up more
than 25% of the organoid. Choroid plexus exceeding this percentage might indicate a failed
differentiation (Fig. 4d).
As a benchmark of a successful differentiation at day 60 in culture, it is important to
confirm that common markers used to identify excitatory cerebellar progenitors such as PAX6 and
TBR1 are not instead labeling forebrain cells (identified by the expression of FOXG1 in Fig. 4eg). One clear indication of the presence of forebrain within the hCerOs are regions that are
PAX6+TBR1+BARHL1- (Fig. 4e-g). Cerebellar cells must have PAX6 and BARHL1 in close
proximity or colocalizing within a region of the organoid.
Another indication of a successful differentiation at the day 60 time point includes the
identification of maturing Purkinje cells expressing CALB1 and exhibiting a distinctive dendritic
morphology with single dendritic outgrowth (Fig 3f). As the Purkinje cells mature approaching 6

21
months inculture, they exhibit an increase in dendritic arborization and complexity (Fig. 3g).
Functional maturation of Purkinje Cells in hCerOs can be further confirmed by functional assays
including the study of calcium dynamics, GCaMP analysis and Patch Clamp recordings of whole
organoids.
4 DISCUSSION
The protocol described here applies to a wide range of studies focusing on human cerebellar
development and disease. This system allows us to gain insight into human-specific traits that
cannot be studied in other model systems. For example, hCerOs provide a platform for exploring
the abnormal development of the human rhombic lip, which has been associated with various
disorders including Dandy-Walker malformations and cerebellar vermis hypoplasia37,40. The
presence of human-specific rhombic lip subventricular zone progenitors within our hCerOs offers
an opportunity to model medulloblastoma, the most common childhood brain tumor, which has
recently been shown to originate from this specific region81,82. Additionally, the potential to induce
laminar layering within hCerOs allows for tracking the migration of the rhombic lip derivatives as
they move toward the edge of the organoid, replicating the tangential migration of granule cell
progenitors during cerebellar development. This can provide deeper insight into the migration
dynamics of these excitatory progenitors, similar to the extensive studies conducted on migrating
interneurons from the MGE, LGE and CGE in the ventral forebrain83–90. Furthermore, this system
has the potential to better mimic human-specific disease phenotypes, such as the loss of Purkinje
cells in patients with ataxia telangiectasia, a phenotype not observed in murine models of the
disease91,92. In addition, by leveraging these capabilities, hCerOs can serve as a more robust
platform for conducting drug screenings to address these phenotypes and aid in developing more
precise and targeted therapies.
However, these organoids are not without limitations, which are commonly encountered in human
brain organoid model systems. One such limitation pertains to nutrient and oxygen delivery
inadequacies, particularly affecting the core of the organoid, thereby restricting its growth.
Moreover, the absence of surrounding embryonic tissue impairs cellular organization. To mitigate
stress on the organoid core, we implement shaking/bioreactor conditions during culturing. Unlike
forebrain organoids, which undergo significant size increase due to rapid expansion of the radial

22
glial population, hCerOs tend to remain relatively smaller. Consequently, this results in less
extensive necrotic core formation observed in cortical organoids.
To address the absence of exogenous guidance cues, we introduce SDF1a to facilitate
cytoarchitecture organization. However, the maintenance of laminar layering within hCerOs
necessitates consistent addition of SDF1a to the culture media. Nevertheless, further research is
needed to overcome the deficiency of various supporting cell types and guidance cues present in
vivo.
A distinctive challenge in hCerOs lies in the absence of climbing fibers and mossy fibers, crucial
cell types originating from sources other than inhibitory or excitatory progenitor populations of
the developing cerebellum. In mice, climbing fibers originate from the inferior olive, while mossy
fibers originate from the pontine nuclei/brainstem/spinal cord and innervate the cerebellum around
E17 and E13, respectively93. Given that climbing fibers cannot be generated within this system,
alternative co-culture methods are necessitated to introduce them.
Cell line to line variability is a well-known hurdle for iPSC derived culture systems and therefore
should be accounted for within this protocol as well. This variability can arise from factors such
as the cell line, passage, and batch-to-batch variability of drugs. To mitigate this limitation, it is
recommended to minimize variability across drug batches. Additionally, when utilizing patientderived lines, the use of isogenic cell lines is advised to limit cell line variability that may change
the ratio of different cell types found in the organoids.
Since the groundbreaking publication from the Sasai lab outlining a method to generate hCerOs94,
subsequent studies have adapted and refined this protocol, leading to the generation of cerebellar
tissue in 3D culture that can survive longer in culture95–98. Each of these methods utilizes the ability
of hPSCs to self-organize into the progenitor zones of the cerebellum after TGFbeta inhibition for
neuroectoderm specification and caudalization through high concentrations of FGF2+insulin99.
Unlike these methods, our protocol relies on dual SMAD inhibitors (SB43152/Noggin) and CHIR,
a WNT pathway activator, to promote neuralization and caudalization, respectively. Crucially, we
specify the midbrain-hindbrain boundary using FGF8b on Day 4, recreating the isthmic organizing
center as seen in vivo71–73,80,100. Our optimized basal media composition during the initial 16 days
facilitates the development of both inhibitory and excitatory neurons. Among the differences
between our protocols and previous ones, there is not only the unique formulation of our basal

23
medium for neural identity specification but also differences in the differentiation medium that
include distinct combination of pro-neuronal survival components and shaking culture conditions
to enhance oxygen and nutrient exchange. After 30 days, our organoids are transferred to serumfree media containing lipids, heparin, and a combination of B27, N2, BDNF, and T3 for prolonged
growth and maturation. By this stage, our organoids reproducibly exhibit spatially segregated
excitatory and inhibitory progenitor populations, that will give consistently rise to the main cell
types of the developing cerebellum.
Although a careful side-by-side comparison of these protocols has not been made, from published
data, it appears that patterning relying solely on FGF2+insulin requires murine cells to achieve
maturation of Purkinje cells with a distinctive firing profile of their in vivo counterpart94. However,
our protocol enables the maturation of Purkinje cells with functional features and canonical
markers of endogenous Purkinje cells in an all-human system.
Finally, while patterned, forebrain organoids have achieved a higher level of complexity by
induction or integration of other cell types present in the human brain including oligodendrocyte,
microglia, and endothelial cells101–107, these approaches have not been attempted in hCerOs
limiting their potential for studying development and disease.

24
5 FIGURES/TABLES
Fig. 1: hCerO differentiation timeline. (A) Timeline of Day 0 dissociation and platting of stem
cells to initiate hCerO differentiation. (B) Timeline of cerebellar induction and patterning protocol
for generating hCerOs along with key developmental stages, cellular identity, and culture
conditions. NE: neuroepithelium, PC: Purkinje cell, GC: granule cell, PIP: PAX2+ interneuron
progenitors, UBC: unipolar brush Cells, iCN/eCN: inhibitory and excitatory cerebellar nuclei,
ROCKi: Y-27632-ROCK inhibitor, SB: SB431542-TGFB inhibitor, Noggin: BMP inhibitor,
CHIR: CHIR99021-GSK3 inhibitor, FGF8b: fibroblast growth factor 8b, gfCDM+i: growth
factor-free chemically defined media + insulin, CerDM1/2/3: cerebellar differentiation media
1/2/3, T3: triiodothyronine, BDNF: bone derived neurotrophic factor, SDF1a: stromal cell-derived
factor 1a.

25
Fig. 2: Morphological features of early hCerOs up to Day 16. (A) Brightfield images of hCerOs
of various pluripotent stem cell lines. Scale bar: 200um. (B) Day 16 organoids from 3 different
cell lines. Scale bar: 1mm. (C) Brightfield images of hCerOs derived from PGP1 cell line
comparing the original protocol (switch to CerDM1 on day 6 up to day 16) vs remaining in
gfCDM+i all the way through the patterning phase of the differentiation (day 0-16). Scale bar:
200um.

26
Fig. 3: Morphological and immunohistochemical features of Day 30+ hCerOs. (A) Brightfield
image of hCerO (gfCDM+i from d0-6, CerDM1 from d6-16) at Day 30. Scale bar: 200um. (B)
Immunostaining of key cerebellar markers including BARHL1 (granule cell progenitors) and
SKOR2 (post-mitotic Purkinje Cell progenitor) and their respective progenitor pools (ATOH1
(rhombic lip progenitor cells) and KIRREL2 (ventricular zone progenitor cells)). SOX2 stains

27
neuroepithelial progenitors. Scale bar: 200um. (C) Brightfield image of modified hCerO
(gfCDM+i from day 0-16) at Day 30. Scale bar: 200um. (D) Immunostaining of key cerebellar
markers including BARHL1 (granule cell progenitors) and SKOR2 (post-mitotic Purkinje Cell
progenitor) and their respective progenitor pools (ATOH1 (rhombic lip progenitor cells) and
KIRREL2 (ventricular zone progenitor cells)). SOX2 stains neuroepithelial progenitors and
endogenous tdTOMATO signal from FOXG1 reporter cell line. Scale bar: 200um. (E) Brightfield
image of day 60 organoids. Arrows indicate the presence of choroid plexus-like cells. (F) 2-month
old hCerOs stained with CALB1. Arrows indicate the presence of dendrites. Scale bar: 50um. (G)
6-month old hCerOs stained with CALB1. Scale bar: 50um

28
Fig. 4: Morphological and immunohistochemical features of suboptimal outcomes. (A)
Brightfield image of an unsuccessful seeding with significant cell death surrounding the central
ball in a V-bottom plate. Scale bar: 200um. (B) Brightfield image of an improperly patterned
organoid leading to a forebrain identity with many small rosettes within the organoid. Scale bar:
200um. (C) Lower magnification of unsuccessful pattered organoid batch leading to some
organoid with forebrain-like identity indicated by black arrows. (D) Brightfield image of day 60
organoids with a high amount of choroid plexus-like cells taking over the organoid. Scale bar:
200um. (E-G) Immunofluorescence images of 20um thick, consecutive sections of day 60

29
organoid stained with BARHL1, PAX6, MAP2, SOX2, TBR1, and endogenous tdTOMATO
signal deriving from a FOXG1 reporter cell line. Scale bar, 200um.
Table 1: Troubleshooting Table
Step Problem Possible Reason Solution
Step 8,9 Significant cell death
surrounding central
ball
Old media or incorrect
addition of morphogens
-Make sure to use fresh
media and include all
necessary components
including
Steps 9-15 Small EBs, no
expansion
Seeded too little cells
Unhealthy hPSCs
Density of hPSCs was too
high prior to dissociation
-Check accuracy of cell
counting.
-Robust stem cells should
still be alive in Trypan blue
solution after 30min. If
most/all cells are blue, then
perform dissociation faster
and more gently.
-Check pluripotency by
morphology of stem cells
and stain for pluripotent
markers.
Steps 9-15 Disintegrating
organoids in 96-well
V-bottom plates and
media is very yellow.
Too many cells plated in one
well
-Check accuracy of cell
counting.

30
Steps 9-15 Outgrowths and buds
forming from the
organoid
Generation of other germ
layers
-Check pluripotency by
morphology of stem cells
and stain for pluripotent
markers.
-Make sure dual SMAD
inhibitors have been stored
properly, not undergone too
many freeze thaws, and
added at the correct
concentration.
Steps 16-18 Disintegrating
organoids after
transfer
Unhealthy hPSCs
Rough transfer step/serrated
pipette tip
-Check pluripotency by
morphology of stem cells
and stain for pluripotent
markers.
-Make sure CerDM2 media
is fresh.
-Make sure the tip cut off of
the pipette tip to create a
wide bore is smooth and
without jagged edges.
Steps 19 Media yellow/acidic
prior to next feeding
Too many organoids in one
plate
-Feed the plates with 12-15
ml of media between Day
16-30
-Feed the organoids in the
bioreactor with a minimum
of 50mL
If not adjusted for, the
organoids will be more

31
sensitive overtime and
disintegrate easily.
Step 20,
Option B
Not enough cells
recovered
Organoid is too small
Dissociation was too gentle
-Collect more organoids
when doing the dissociation
by pooling organoids
together when doing the
Papain dissociation.
-Increase the time the
organoids remain in the
Papain solution to up to 1
hour during the initial
incubation stage.
Too many “blue cells”
(dead cells) after
Trypan blue staining
in Step 12
Too harsh dissociation
Unhealthy/Very old organoid
-Ensure the dissociation is
not too harsh by limiting the
use of a P1000 tip to prevent
excessive sheering of cells.
-Reduce the amount of time
within the Papain solution
and keep within
recommended time.
-Conduct single nuclei
isolation for RNA
sequencing on
older/unhealthy organoids as
an alternative if the initial
intent was to conduct whole
cell single cell RNA

32
sequencing.
Step 20,
Option C
Low/no
GcAMP/opsins viral
labeling after
transfection
Low viral titer
Not enough incubation time
Disintegrating organoids
Diffuse fluorescent signal
-Make sure viral titers are
sufficient, if not then add
more of the virus to the
initial incubation.
-Wait at least 24 hours prior
to conducting full media
change after initial
inoculation of organoids
with virus
-Examine under brightfield
microscopy to ensure that
organoids exhibit a healthy
morphology with smooth
edges. Any organoid
displaying disintegrating
edges or cell death
following viral transfection
should be excluded.
-The GFP and RFP signals
should manifest as sharp and
contained within the cell
soma. While a diffuse signal
may be acceptable within
the initial days postinfection, if no cell-specific
fluorescent signal is

33
observed by the second
week after infection, it is
necessary to infect new
organoids.
Step 20,
Option D
Equipment setup
Patch Clamp
Noise signals in the
recordings
Difficulty in creating a stable
patch
-The volume of media in the
recording chamber depends
on the type and size of
chamber used. Adjust
volume so that is enough for
the organoid and the
reference electrode to be
submerged while avoiding
over floating and spillage of
media as this can cause
noise signals in the
recordings and damaging
equipment.
-The amount of internal
solution should be adjusted
depending on the length and
shape of patch pipettes. It is
important to have enough
internal solution to reach the
recording electrode,
however large volumes of
internal solution should be
avoided for ensuring optimal
seal.

34
Wrong resistance on Patch
Pipettes
-Find the proper resistance
of patch pipette before
beginning the recordings. If
resistance is not within 3-5
MOhm check under
brightfield image that the
pipette has a clear internal
solution without debris. If
the debris or small particles
are visible, filter the solution
with a 0.22-μm PES Filter
System. If no debris is
present, change parameters
on puller (according to
provider) to adjust up to the
right values.
Patch-clamp
recordings
Cells are difficult to patch
Difficulty in reaching a seal
-Select GFP+ cells on the
edge of the organoid with a
clear visible soma coming
out from the edge. Internal
cells will not be accessible
for patch clamp recordings.
-Leave the patch in attached
mode for up to 10 minutes
and apply negative pressure
for few times to help
braking in the membrane

35
Table 2: Antibodies for Tissue Characterization
Cell Type/tissue Antigen Supplier Catalog
Number
Dilution Temporal
Expression
Rhombic Lip
progenitors
Rabbit monoclonal
anti-ATOH1
Abclonal A11477 1:200 From Day 30
Rhombic Lip
derived Excitatory
neuronal
precursors
Rabbit polyclonal
anti-BARHL1
Atlas
Antibodies
HPA004809 1:500 From Day 30
Ventricular Zone
progenitors
Rabbit polyclonal
anti-KIRREL2
Proteintech 10890-1-AP 1:100 From Day 30
Ventricular Zone
derived postmitotic
Purkinje cell
precursors
Rabbit polyclonal
anti-SKOR2
Atlas
Antibodies
HPA046206 1:500 From Day 30
Granule Cells Rabbit polyclonal
anti-PAX6
Biolegend 901301 1:200 From Day 30
Deep Layer
Projection Neurons
of the Forebrain
Rabbit polyclonal
anti-TBR1
Abcam AB31940 1:200 From Day 60
Purkinje Cells Mouse monoclonal
anti-CALB1
SigmaAldrich
C9848-
100UL
1:300 From Day 60
Radial Glia Goat polyclonal
anti-SOX2
RD
systems
AF2018 1:500 From Day 30

36
Table 3: Major Checkpoints
Checkpoint
day
Question Continue Terminate IHC markers
Day 2 Have the hPSCs
aggregated into a tight
EB with a defined
boundary and only
have a small amount
of cell death
surrounding the central
EB?
Yes No
Make sure you
terminate only if
there is a significant
number of cells
surrounding the
central ball. (See
Figure 2)
N/A
Day 19 Are the organoids
intact after Day 16
transfer?
Yes No
(Unhealthy
organoids often
start to disintegrate
and no longer have
distinct borders.)
N/A
Day 30 Is the aggregate
forming a clear
neuroepithelial layer
with both progenitor
populations?
Yes No
(Make sure there
aren’t smaller
rosette structures
within the organoid
prior to terminating
by IHC of
cryosectioned
organoids.)
SOX2,
ATOH1,
KIRREL2
Do the organoid
express both excitatory
Yes No
(Not having one
BARHL1,
SKOR2

37
and inhibitory
neuronal markers?
cell type may not
compromise the
health of the
organoid, but may
affect the
maturation rate.)
Day 60 Do the organoids
express more mature
neuronal markers
Yes No CALB1,
PAX6

38
6 MATERIALS
6.1 REAGENTS
6.1.1 Cells
hPSCs or iPSCs. We have successfully the PGP1 (Personal Genome Project 1) human iPS cell line
was from the laboratory of G. Church; PGP1-FOXG1 human iPS cell lines were obtained from
Harvard Stem Cell Institute; the 11a human iPS cell line was from the laboratory of Kevin Eggan,
and the D2 human iPSC cell line was from the laboratory of Marcelo Coba. We have also used H9
human ESCs (hESCs), feeder-independent (Wisconsin International Institute, WA09 cells),
however they have yield less reproducible outcome52.
!CAUTION
The use of human tissues and human stem cells must adhere to institutional and funding body
regulations, as well as to relevant ethical guidelines.
6.1.2 Growth media and supplements
● AmphoB (Gibco, cat. no. 15290018)
● Apo-transferrin (Sigma, cat. no. T1147-100MG)
● BDNF (R&D Systems, cat. no. 248-BDB-050)
● BrainPhys Neuronal Medium (STEMCELL Technologies, cat. no. 05790)
● BrainPhys Optimized Medium (STEMCELL Technologies, cat. no. 05796)
● Bovine Serum Albumin (BSA) (fatty acid free and globulin free) (Sigma, cat. no. A0281)
● B-27+vitA supplement (Gibco, cat. no. 17504001)
● Chemically Defined Lipid Concentrate (CDLC) (Gibco, cat. no. 11905031)
● CHIR-99021 (Reprocell, cat. no. 04-0004)
● DMEM/F12 1:1 medium with L-glutamine, HEPES (Cytiva, cat. no. SH30023.FS)
● FGF8b (Peprotech, cat. no. 100-25)
● Glutamax (Gibco, cat. no. 35050061)
● Growth-factor reduced matrigel (Corning, cat. no. 354230)
● Ham’s F-12 medium (Gibco, cat. no. 11-765-054)
● Heparin (Sigma, cat. no. H3149-25KU)

39
● IMDM medium (Gibco, cat. no. 12440053)
● Insulin (Sigma, cat. no. I9278-5ML)
● Knockout Serum Replacement (KSR) (Gibco, cat. no. 10828028)
● Mono-Thioglycerol (Sigma, M1753)
● mTeSR1 Complete Kit (Basal Medium and 5X supplement) (STEMCELL Technologies,
cat. no. 85850)
● Penicillin/Streptomycin (Cytiva, cat. No. SV30010)
● Neurobasal medium (Gibco, cat. no. 21103049)
● BrainPhys NeuroCult SM1 Neuronal Supplement (STEMCELL Technologies, cat. no.
05711)
● Noggin (Peprotech, cat. no. 120-10C)
● N2 supplement (Gibco, cat. no. 17502001)
● ROCK inhibitor (Y-27632) (STEMCELL Technologies, cat. no. 72302)
● TGFB inhibitor (SB31542) (Selleckchem, cat. no. S1067)
● SDF1a (CXCL12) (Peprotech, cat. no. 300-28A)
● Triiodothyronine (T3) (Sigma, T2877-250MG)
● 2-Mercaptoethanol (BME) (Gibco, 21985023)
6.1.3 Cell Culture reagents
● Accutase (STEMCELL Technologies, cat. no. 07922)
● Ammonia Solution (4M in methanol) (Sigma, cat. no. 779423)
● DMSO (Sigma, D2650)
● Ethyl Alcohol (LabChem, cat. no. LC222105)
● Geltrex (Thermo Fisher Scientific, cat. no. A1413301)
● Papain Dissociation Kit (Worthington BioChem, cat. No. PDS/LK003150
● Phosphate Buffer Saline (PBS 1X) sterile (Cytiva, cat. no. SH30256.01)
● Trypan Blue Stain (0.4%) (Gibco, cat. no. 15250061)
● Geltrex (ThermoFisherScientific, cat.no. A1413301)
● Bleach (Pure Bright Kik Bleach)
● Potassium D-gluconate,≥99% (Sigma Aldrich, cat.no. G4500)
● Potassium chloride,BioXtra, ≥99.0% (Sigma Aldrich, cat.no. P9333)

40
● Ethylene glycol-bis(2-aminoethylether)-N,N,N′,N′-tetraacetic acid (Sigma Aldrich, cat.no.
E3889)
● HEPES, Crystalline Powder, ≥99.5% (Sigma Aldrich, cat.no.H3375)
● Sodium chloride, ReagentPlus®, ≥99% (Sigma Aldrich, cat.no.S9625)
● Adenosine 5′-triphosphate magnesium salt (Sigma Aldrich, cat.no.A9187)
● Guanosine 5′-triphosphate sodium salt hydrate (Sigma Aldrich, cat.no.G8877)
● Potassium hydroxide (Sigma Aldrich, cat.no.P5958)
● NBQX disodium salt (Tocris, cat.no. 1044)
● D-AP5 (Tocris, cat.no. 0106)
● NMDA (Tocris, cat.no. 0114)
● Picrotoxin (HelloBio, cat.no. HB0506)
● Tetrodotoxin (Tocris, cat.no. 1078)
● L7-eOPN3 Lentivirus (From Dr. Carlos Lois at Caltech)
● AAV8.L7-6.eGFP.WPRE.hBG (Addgene, cat.no. 126462)
● pAAV-CAG-SomaGCaMP6f263 (Addgene, cat.no. 158757)
● Borosilicate Glass Pipette (Sutter Company, cat.no. B150-86-10)
6.1.4 Immunostaining and cryosectioning reagents
● BSA (for non-sterile use) (GeminiBio, cat. no. 700-105P)
● DAPI
!CAUTION
Given DAPI’s photosensitivity, it should be exposed to light for as little time possible)
● Fluoromount (Electron Microscopy Science (EMS), cat. no. 50-259-73)
● 16% Paraformaldehyde (16% PFA) (Electron Microscopy Science (EMS), cat. no. 15710)
● PBS (non-sterile) (VWR, cat. no. 0780)
● Primary Antibodies (See Table 2)
● Sucrose (Sigma, cat. No. S0389)
● Tissue-Tek O.C.T. compound (Sakura, cat. no. 62550)
● TritonX-100 (Sigma, cat. no. T9284)
● Tween20 (Sigma, cat. no. P9416)

41
6.2 EQUIPMENT
● CO2 incubators (Thermo Scientific, cat. no. 370)
● Biological safety cabinet (Thermo Scientific, cat. no. 13-261-304)
● Eppendorf Research Plus - Mechanical Pipette (P1000, P200, P10, P2, and P200-8-
channel)
● Sterile pipette tips (1 ml, 200,10, and 2 μl; (CELLTREAT, cat. no. 229047; Genesee
Scientific cat. no. 24-150RL; and VWR cat. no. 47745-164)
● Sterile microcentrifuge tubes (1.7-ml size; VWR cat. no.87003-294)
● 0.22-μm PES Filter System (1 L, 500, and 150 ml; CELLTREAT, cat. nos. 229708,
229707, and 229705)
● Falcon 5mL round bottom polystyrene test tube with cell strainer cap (“filter basket”)
(Corning, cat. no. 352235)
● PrimeSurface 3D Culture Spheroid plates: Ultra-low attachment, 96 well, V-bottom plates
(S-Bio, cat. no. MS-9096VZ)
● Conical tubes, 15 ml (MTC bio, cat. no. C2601)
● Conical tubes, 50 mL (CELLTREAT, cat. no. 229421)
● Tissue culture dish, 60 mm (CELLTREAT, cat. no. 229660)
● Costar® 24-well Clear Flat Bottom Ultra-Low Attachment Multiple Well Plates (Corning,
cat.no. 3473)
● Costar® 6-well Clear Flat Bottom Ultra-Low Attachment Multiple Well Plates (Corning,
cat.no. 3471)
● Glass Bottom Microwell Dishes, 35 mm (MATTEK, cat.no. P35G-0-14-C)
● Velp Scientifica™ Glass Bottle for Waste Collection (TermoFisherScientific, cat.no. 11-
392-021)
● Spinner flask, 125 ml (Corning, cat. no. 3152)
● Dura-Mag 9-position, digital stirrer (for spinner flasks) (Chemglass Life Sciences, cat. no.
CLS-4100-09)
● Orbital shaker (ThermoFisher Scientific, cat. no. 88881101)
● Orbital shaker accessory (rubber mat) (ThermoFisher Scientific, cat. no. 88881123)

42
● Serological pipettes, 5, 10, 25, and 50ml VWR, cat. nos. 89130-896, 89130-898, 89130-
900, 89130-900, respectively)
● Sterilized scissors
● Water bath, 37 °C (Fisher Scientific, Isotemp water bath)
● Stereomicroscope (Zeiss, Stemi 2000)
● Inverted contrasting tissue culture microscope (Leica, DMIL)
● Levy Counting Chamber, Hausser Scientific (“Hemocytometer”) (VWR, cat. no. 15170-
208)
● Benchtop centrifuge (Thermo Scientific, Heraeus Multifuge 1s)
● Weighing dishes (Sigma-Aldrich, cat. no. Z154873)
● Dry ice
● Low-temperature thermometer (Sigma-Aldrich, cat. no. Z257400)
● Laboratory spatula (Sigma-Aldrich, cat. no. S3897)
● Glass slides (Globe Scientific, cat. no. 1354W)
● Coverslips 24x60mm (VWR, cat. no. 48393-106)
● Cryostat (Leica)
● Cryomold: biopsy (small), intermediate (medium), and standard (large) (VWR, cat. no.
25608-922, 25608-924, and 25608-916, respectively)
● SP-8X microscope with a multiphoton laser (Leica) and DH-40iL Culture Dish
Microincubator (Warner Instruments corporation division of Harvard Apparatus)
connected to a TC-344C Dual Channel Temperature Controller (Warner
InstrumentsWarner Instruments corporation division of Harvard Apparatus). Fiber coupled
LEDs for optogenetic stimulation (ThorLabs, cat.no. M530F2)
● Glass capillary puller, Sutter company, P-1000
● Multiclamp 700B amplifier, Molecular Devices
● Axon Digidata 1550B
● Scientifica SliceScope Pro 3000
● Clampfit software, Molecular Devices
● Multiclamp software, Molecular Devices
● Milli-Q® IQ 7000 Ultrapure Water Purification System (Sigma-Aldrich, cat.no.
ZIQ700T0C)

43
6.3 REAGENTS SETUP
6.3.1 Feeder-independent hESC and hiPSC lines
The hESC and hiPSCs are cultured with mTeSR1 medium using standard practice in 5% CO2
incubators set to 37C. Cells are thawed into 60mm dishes that are coated with geltrex (1:100
dilution dissolved in DMEM-F12). Thawed cells are passaged with 0.5mM EDTA in sterile 1xPBS
without calcium or magnesium. from one 60mm dish at 80% confluency into 3, 60mm dishes.
6.3.2 Reconstituting growth factors, morphogens, and small molecules
Initially, prepare a 10% (wt/vol) BSA/1xPBS solution by dissolving 1g of BSA into 10ml of sterile
1xPBS, which can be stored in -80°C for 1 year.
Reconstitute 2mg of CHIR in 2.52mL of DMSO.
Reconstitute 100mg of apo-transferrin in 10mL of sterile water to obtain a 10mg/mL solution.
Reconstitute 100ug of Noggin in 1mL of sterile 1xPBS + 0.1% (wt/vol) final concentration BSA
to obtain a 100ug/mL solution.
Reconstitute 100ug of FGF8b in 1mL of sterile PBS + 0.1% (wt/vol) final concentration BSA to
obtain a 100ug/mL solution.
Reconstitute 10mg of SB431542 in 2.602mL of DMSO (TGFBi) to obtain a 10mM solution.
Reconstitute 10mg of ROCKi in 3.122mL sterile water to obtain a 10mM solution.
Reconstitute 100ug of SDF1a in 1mL of sterile PBS + 0.1% (wt/vol) final concentration BSA to
obtain a 100ug/mL solution.
Reconstitute 1mg of BDNF in 10mL of sterile 1xPBS + 0.1% (wt/vol) final concentration BSA to
obtain a 100ug/mL solution.
Reconstitute 50mg of Heparin in 5mL of sterile water to obtain a 10mg/mL solution.
Reconstitute 250mg of T3 into 12.8mL of DMSO to get a 1:2000 dilution. Dissolve each of the
200uL aliquots of DMSO in 200uL of ammonia solution (4M in methanol) when ready to use at
1:1000 dilution.
After thawing growth factors, morphogens, and small molecules, store them at 4°C for up to 1
week.
Make 5ml aliquots of N2, 10ml aliquots of B27, 50ml aliquots of KSR, supplements and store it
at -20°C for up to 1 year.

44
!CRITICAL
Make sure to store reconstituted solutions at -20°C for up to 6 months. Limit freeze thawing of all
reagents. Once thawed, store at 4°C for up to 1 week.
6.3.3 Aliquoting Matrigel
Thaw Matrigel on ice at 4°C overnight. Pre-cool a sterile, 1mL pipette tip box and ten
microcentrifuge tubes at −20°C for 15 min. In a sterile hood, quickly transfer 1 ml of Matrigel into
each pre-cooled microcentrifuge tube. Store aliquots at −20°C for up to 1 year. Avoid repeated
freezing and thawing. Once thawed, place aliquot at 4°C and use as needed.
!CRITICAL
Matrigel polymerizes at room temperature, so it's crucial to ensure that all materials coming into
contact with it are chilled. Additionally, aliquoting should be performed swiftly to minimize
polymerization.
6.3.4 hiPSC and hESC medium
Add 100ml of 5X supplement to 400ml of mTeSR1 medium to form a complete medium that is
used to conduct full media change on stem cells daily.
6.3.5 gfCDM+i medium
First, weigh out 0.25g of BSA (FA free + globulin free) and add it to a combined 1:1 ratio of 25ml
IMDM with 25ml Ham’s F-12 leading to a final concentration of 5mg/ml BSA. Let that dissolved
in the 4°C for about 1 hour prior to adding apo-transferrin (final concentration of 15ug/mL), insulin
(final concentration of 7ug/mL), 1% (vol/vol) CDLC, mono-thioglycerol (final concentration of
0.1mM), and 1% (vol/vol) Pen/Strep. Once all components are added, filter sterilize the media
through 0.22um filter and store it at 4°C and use within 1 month.
Patterning Media (50ml) (day 0-6)
Reagents Volume Target
IMDM 25ml n/a
F-12 25ml 20%
BSA 0.25g 5mg/mL
apo-transferrin 75ul 15ug/mL

45
Insulin 35ul 7ug/mL
CDLC 500ul 2mM
1-thioglycerol 1.95ul 0.1mM
Pen/Strep 500ul
Total Volume 51.11ml
Filter sterile media
6.3.6 CerDM1 medium
Combine DMEM-F12 with 20% (vol/vol) KSR, 1% (vol/vol) GlutaMAX supplement, apotransferrin (final concentration of 15ug/mL), insulin (final concentration of 7ug/mL), BME (final
concentration of 0.1mM), 1% (vol/vol) Pen/Strep. Once all components are added, filter sterilize
the media through 0.22um filter and store it at 4°C and use within 1 month.
Patterning Media (50ml) (day 6-16)
Reagents Volume Target
DMEM/F12 40ml n/a
20% KSR 10ml 20%
apo-transferrin 75ul 15ug/mL
Insulin 35ul 7ug/mL
Glutamax 500ul 2mM
0.1mM BME 90.9ul 0.1mM
Pen/Strep 500 1%
Total Volume 51.2ml
Filter sterile media
6.3.7 CerDM2 medium
Combine DMEM-F12 with 1% (vol/vol) N2, 2% (vol/vol) B27+vitA, 1% (vol/vol) GlutaMAX,
1% (vol/vol) P/S. Once all components are added, filter sterilize the media through 0.22um filter
and store it at 4°C and use within 1 month.
Induction Media (50mL) (Day 16-30)
Reagents Volume Target
DMEM/F12 50ml n/a

46
1% N-2 500ul 1%
2% B-27 1ml 2%
Glutamax 500ul 2mM
Pen/Strep 500ul 1%
Total Volume 52.5ml
Filter sterile media
6.3.8 CerDM3 medium
Combine 1:1 ratio of DMEM-F12 and Neurobasal with 1% (vol/vol) N2, 2% (vol/vol) B27+vitA,
1% (vol/vol) GlutaMAX, 1% (vol/vol) CDLC, Heparin (final concentration of 5ug/mL), 1%
(vol/vol) Pen/Strep. Once all components are added, filter sterilize the media through 0.22um filter
and store it at 4°C and use within 1 month.
Maintenance Media (50mL) (Day 30 - 60)
Reagents Volume Target Comment
DMEM/F12 25ml n/a
Neurobasal 25ml n/a
1% N-2 500ul 1%
2% B-27 1ml 2%
Glutamax 500ul 2mM
Heparin 25ul 5ug/ml
1% CDLC 500ul 1%
Pen/Strep 500ul 1%
T3 50ul 0.5ng/mL add right before use
Matrigel 500uL 1% add right before use (optional)
SDF1a 50ul 100ng/ml add right before use (optional)
Total Volume 53.13ml
Filter sterile media
6.3.9 CerDM4 medium
Once all components are added, filter sterilize the media through 0.22um filter and store it at 4°C
and use within 1 month.

47
Maintenance Media (50mL) (Day 60+)
Reagents Volume Target Comment
DMEM/F12 25ml n/a
Neurobasal 25ml n/a
1% N-2 500ul 1%
2% B-27 1ml 2%
Glutamax 500ul 2mM
Heparin 25ul 5ug/ml
1% CDLC 500ul 1%
Pen/Strep 500ul 1%
T3 50ul 0.5ng/mL add right before use
BDNF 7ul 14ng/ml add right before use
Total Volume 53.13ml
Filter sterile media
!CRITICAL
Store gfCDM+i, CerDM1, CerDM2, CerDM3 for up to 2 weeks at 4°C for up to 2 weeks after making it.
6.3.10 BrainPhys medium
Maintenance Media prior to Ephys analysis (50mL)
Reagents Volume Target Comment
BrainPhys 500ml
Neuro Supplement 10ml
BDNF 7ul 14ng/ml add right before use
T3 50ul 0.5ng/mL add right before use
6.3.11 K-gluconate internal solution
Patch pipette internal solution(100mL)
Reagents mg x 100 ml Milli-Q water
K-D-gluconate 2873
KCL 93.19
KOH-EGTA 7.6

48
KOH-Hepes acid 238.3
NaCl 46.7
Mg2ATP 101.44
Na3GTP 15.69
KOH adjustment of pH
!CRITICAL
Agitate the mixture as you proceed with the preparation, adhering to the specified order. Mg2ATP
and NaGTP should be stored at -20 °C. Quickly introduce them into your solution and
subsequently keep the solution on ice. Add water, but avoid reaching the final volume, as pH
adjustment to 7.3 is needed. Typically, the initial pH is acidic; utilize 1 or 2M KOH to achieve the
desired pH. Once the pH is satisfactory, if needed, add water to reach the intended volume.
Subsequently, aliquot your solution into Sterile microcentrifuge tubes and store them at -20 °C.
6.3.12 Toxins and drugs preparation
Reconstitute 1mg of NBQX in 496 ul of Milli-Q water to obtain a 5 mM solution. Use at final
concentration of 5 uM.
Reconstitute 50mg of D-AP5 in 101.45 ul of Milli-Q water to obtain a 50 mM solution. Use at
final concentration of 50 uM.
Reconstitute 10mg of NMDA in 6.8 mL of Milli-Q water to obtain a 10 mM solution. Use at final
concentration of 3-100 uM.
Reconstitute 301.295mg of Picrotoxin in 5 mL of DMSO to obtain a 100 mM solution. Use at final
concentration of 100 uM.
Reconstitute 1mg of Tetrodotoxin in 3.13 mLl of Milli-Q water to obtain a 1 mM solution. Use at
final concentration of 1 uM.
6.3.13 Papain dissociation preparation
Prepare a 10% (wt/vol) BSA/1xPBS solution by dissolving 1g of BSA into 10 ml of sterile 1xPBS,
which was stored in -80°C for 1 year. Use this to further dilute it to a 0.04%(wt/vol) BSA/1xPBS
solution in order to create final cell suspension after dissociation of organoids.
Dissolve one bottle of lyophilized papain with 5mL of Earle’s media.
Dissolve one bottle of DNase with 500uL of Earle’s media.

49
Dissolve one bottle of Inhibitor with 32mL of Earle’s media.
6.4 EQUIPMENT SETUP
6.4.1 Spinning Bioreactors
Place and install a CO2 resistant low-speed stir plate into the tissue culture incubator by sterilizing
it with 70% ethanol. Place the shelves of the incubator in a way that allows easy placement and
removal of the bioreactors on the stir plate. Exit the power cord out the back of the incubator and
through a stopper to prevent excess CO2 leakage out of the incubator. Tape the chords to hold
them in place.
6.4.2 Orbital Shaker
Sterilize the rubber mat platform with 70% ethanol prior to installing it onto the orbital shaker.
Place and install a CO2 resistant orbital shaker into the tissue culture incubator by sterilizing it
with 70% ethanol. Move the shelves to allow ample vertical space to stack the 10cm dishes as well
as to easily place/remove the 10cm dishes. Place the control box on the side of the incubator and
exit the power cord and control cord out the back of the incubator and through a stopper to prevent
excess CO2 leakage out of the incubator. Tape the chords to hold them in place.
!CRITICAL
Make sure that the 10cm dishes have ridges on the bottom of the plate that allow them to lock into
one another or else stacking plates is not possible.
!CRITICAL
A standard tissue culture incubator will allow for the placement of either an orbital shaker or a
spinning bioreactor, but not both.
6.4.3 2-Photon Calcium Imaging
Set wavelength of 2-photon to 880 nm with emission spectrum between 400-550 nm.
Set time lapse image acquisition for 10 minutes at acquisition rate 10 Hz. Set image size to
512x512 pixels.
Before the start of recordings, set the stage to 37.5°C.
Place pipettes and tips at the microscope station. Safely position bleach in a glass bottle for the
disposal of tips contaminated with drugs.

50
6.4.4 Patch Rig
Thaw the internal solution and keep it on ice for the remainder of the experiment. Add 2 mL of
BrainPhys Optimized Medium into the recording chamber. Prepare 3-5 MOhm patch pipettes from
glass capillaries. Add 3 ul of internal solution in each pipette.
!TROUBLESHOOTING
Test the resistance by mounting the pipette on the electrophysiology rig holder and submerge into
the recording chamber medium.
!TROUBLESHOOTING
Before the start of recordings, set the stage to 37.5°C.
7 METHODS
7.1 MAKING EMBRYOID BODIES (EBs): TIMING: 1 – 2HRS
(DAY 0)
1. Grow the hPSCs in one 60mm dish coated with Geltex until the plate is 70-80% confluent.
This whole plate will be used to start the differentiation. Typically, one confluent plate can
be used to seed ~3 96-well plates.
!CRITICAL
The proper morphology of the stem cell colonies is essential for a successful differentiation.
Therefore, make sure that the colonies do not have any signs of differentiation.
2. Take feeder free hPSCs being cultured and feed the plate 1 hour prior to starting
differentiation with fresh mTeSR media.
3. Wash the plate of hPSCs by removing mTeSR media and add 2ml of 1X PBS.
4. After swirling the plate a few times, remove the 1xPBS and add 2 ml of accutase to the
plate to initiate dissociation of the cells and place plate in the incubator set to 37°C for 3-5
minutes depending on the confluency of the plate and the density of the cell colonies.
!CRITICAL
Make sure to not keep the accutase on the cells for longer than 7 minutes as this causes damage to
the cells and does not yield a good starting EB for differentiation.

51
5. Quench the accutase with at least 4mL of DMEM/F12 and transfer to a 15ml conical tube
to triturate the cell clumps ~10 times to break them up into a single cell suspension.
6. Using 10ul of Trypan blue at a 1:1 ratio with 10ul of the cells in DMEM/F12 suspension,
count the live cells on a hemocytometer while the cells spin in a centrifuge at 200g for 5
minutes at room temperature.
7. Remove the supernatant from the cell pellet and resuspend the pellet in 1ml of DMEM/F12.
Prepare to seed one 96-well V-bottom plate with 6,000 cell per well by transferring 600,000
cells from the resuspended pellet into a 15mL conical tube containing 10mL of gfCDM+i
media containing 10uM TFGBi, 1.7uM CHIR, 50ng/mL of Noggin, and 20uM of ROCKi.
8. Pour this cell suspension into a reservoir and add 100ul to each well of one 96-well Vbottom plate using a multichannel P200 pipette. 24 hours later you should observe a small
EB with defined borders under a microscope. Maintain these organoids within a tissue
culture incubator set at 37°C and 5% CO2.
!CRITICAL
There can be a few dead cells around the EB, but if there is excessive cell death, the differentiation
must be terminated.`
7.2 FEEDING EBs AND INITIATION OF GERM LAYER
DIFFERENTIATION: TIMING 4 DAYS (DAY 0-4)
9. Day 2: prepare gfCDM+i containing 10uM TGFBi, 1.7uM CHIR, 150ug/mL Noggin, and
20uM ROCKi. Remove 40uL of media from each well without disturbing the cells at the
bottom of the well and add 50ul of the freshly prepared gfCDM+i.
10. Day 4: prepare gfCDM+i containing 10uM TGFBi, 1.7uM CHIR, 100ug/mL Noggin,
200ng/mL FGF8b. Remove 40ul of media from each well and add 50ul of this freshly
prepared gfCDM+i.
7.3 INDUCTION OF ISTHMIC ORGANIZING REGION: TIMING
11 DAYS (DAY 4-16)
11. Day 6: prepare CerDM1 containing 10uM TGFBi, 1.7uM CHIR, 100ug/mL Noggin, and
100ug/mL FGF8b. Remove 40ul of media from each well and add 50uL of this freshly
prepared CerDM1.

52
12. Day 8: prepare double the volume of CerDM1 containing 10uM TGFBi, 1.7uM CHIR,
100ng/mL Noggin, and 300ng/mL FGF8b. Remove 40ul of media from each well and add
100uL of this freshly prepared CerDM1, which brings the total volume up to 150uL.
!CRITICAL
Be sure to increase the concentration of FGF8b from 100ng/mL to 300ng/mL from Day 6 to Day
8.
13. Day 10: repeat Step 12, but this time remove 80uL of media and add 100uL of freshly
prepared CerDM1.
14. Day 12: simply conduct a half media change by removing 70uL of media and replacing it
with CerDM1 that does not contain any drugs or morphogens.
15. Day 14: repeat half media change as was done on Day 12 in Step 14. By this day, you may
notice a bit of cell debris around the edge of the organoid in which case you may conduct
a “wash step”. Do this by disturbing the pellet by pipetting the media in the well up and
down prior to removing it and replacing it with fresh CerDM1 to remove as much debris
as possible.
7.4 SHAKING CULTURE OF EXPANDING NEUROEPITHELIAL
BUDS: TIMING 13 DAYS (DAY 16-29)
16. On Day 16 of differentiation, transfer the organoids from the 96-well V-bottom plates
equally into two 10cm dishes.
a. Prepare sterile scissors by spraying the open scissors with 70% ethanol and leaving
it open to dry in the culture hood. Then cut the edge of a 200uL pipette tip to create
a wide-bore pipette tip.
b. Transfer the organoids into a 10cm plate containing ⅕ old media (3mL) from the
96-well V-bottom plate and ⅘ (12mL) CerDM2 media (bringing the total to 15mL
in the 10cm plate).
CRITICAL!
Do not add more than half a 96 well plate full of organoids (48 organoids) into one 10cm dish as
this may affect the overall health of the organoids over the long term.
17. Culture the organoids on a CO2 resistant orbital shaker placed inside of an incubator set at
37°C and 5% CO2. Set the shaker for 70 r.p.m.

53
CRITICAL!
Make sure to use a low speed, CO2 resistant shaker intended for tissue culture use inside of an
incubator. A standard orbital shaker may overheat and be damaged by humidity.
18. Conduct a full media change by removing the media within the 10cm plate and replacing
it with fresh 15mL of CerDM2 every 3 days.
7.5 GROWTH AND MATURATION OF CEREBELLAR TISSUE
(DAY 30 - >1 YEAR): TIMING OVER 1 YEAR
19. Conduct a full media change by removing the media within the 10cm plate and replacing
it with fresh 15mL of CerDM3 starting at Day 30 with 0.5ng/mL of T3 added directly to
the media prior to feeding and add 14ng/mL of BDNF after Day 60.
a. If you intend to create a stratified layered organization within the organoids, add
100ng/mL of SDF1a and a 1:100 dilution (1%) of growth factor reduced matrigel
dissolved directly into the media from Day 30-60. The SDF1a and growth factor
reduced matrigel are not necessary to obtain the cell types of interest, Add these
components to the 10cm plate every 3 days in addition to the T3 when conducting
the full media change until Day 60 to obtain the organized layering.
!CRITICAL
Add Matrigel, T3, SDF1a, or BDNF into CerDM3 on the same day that you intend to use the
media. The base CerDM3 media can be stored for up to 2 weeks in 4C. gfCDM1+i, CerDM1, and
CerDM2 can also be stored at 4°C for up to 4 weeks after preparing.
20. Starting from Day 30 onward, the organoids can be subject to various procedures to be
analyzed at various time points. These may include cryosectioning and immunostaining
the organoids (option A), dissociation into single cell suspension for further analysis at a
single cell resolution (option B), calcium imaging (option C), patch-clamp recordings on
whole organoids (option D).
7.5.1 Option A: Cryosectioning and immunostaining organoids: Timing 2-3 d
1. Transfer organoids into a 1.5mL conical tube using a cut 200ul tip (or 1000ul tip organoids
are over 60 days old).

54
!CRITICAL
Make sure to use sterile scissors to cut the tip.
2. Remove as much of the media as possible without damaging the organoids and completely
submerge the organoids in at least 500uL of 4%PFA (wt/vol) and let it stay at room
temperature for 30 min. Gently remove the PFA, dispose of it properly, and replace it with
1mL of 1X PBS. Let it sit at room temp for 5 min and repeat this 1xPBS wash twice more.
3. Remove the last 1X PBS wash and replace it with 500uL of 30% sucrose (wt/vol) and let
it sit overnight at 4°C until the organoids submerge to the bottom of the tube.
!CRITICAL
Make sure the organoid is floating on the surface when placed in sucrose. If the organoid is still
floating after overnight incubation at 4C, attempt to push the organoid down with a 200uL tip to
see if it starts to sink. If the organoids sink immediately after sucrose is added to the tube or does
not sink after overnight incubation, terminate the experiment.
4. The next day, transfer the organoids into a cryomold and remove all of the sucrose by
pipetting it off.
5. As the organoids sit on the bottom of the cryomold, add Tissue-Tek O.C.T. into the mold
and cover the organoids completely.
!CRITICAL
Do not add too many organoids as this makes sectioning and collecting intact samples difficult.
Make sure the organoids have some space between one another and are away from the walls of the
mold.
6. Place the embedded organoids on dry ice directly until the O.C.T. turns completely white.
Pause Point! You may store these molds at -80°C for over 1 year until they are ready to
section.
7. Transfer the organoid blocks to a cryostat and let it equilibrate to the chamber temperature
of -20°C for 15 min prior to sectioning.
8. Section the organoids at 14-20um and collect them onto glass slides.
9. Stain organoids using a standard immuno technique. See previous publication for details52.
Antibodies used are found in Table 2.

55
10. If you wish to perform whole organoid staining, increase the incubation times of primary
and secondary antibodies to 48 hours at 4°C on a nutator. See previous publication for
details52.
7.5.2 Option B: Dissociating organoids into single cells: Timing 2-3 hr
1. Transfer one organoid from the spinner flask or 10cm dish into a 60 mm cell culture dish
containing 5 mL of prewarmed Worthington papain solution 500 μL of Worthington
DNase solution from the Papain Dissociation Kit.
2. Mince the organoid into small pieces with a razor blade.
3. Place the 60 mm dish that contains the tissue onto an orbital shaker in a cell culture
incubator, and incubate the chopped up organoid on the shaker with a speed set at 70 rpm
for 30 minutes at 37°C.
4. Triturate gently using a 1mL pipette tip several times to break up the minced pieces.
5. Incubate for an additional 5-10min at 37°C.
6. Triturate with 10mL serological pipette ~10 times. Allow any pieces of undissociated tissue
remaining after trituration to settle to the bottom of the tube.
7. Transfer cell suspension to a 15mL conical tube and let the debris and undissociated tissue
settle to the bottom of the tube after 1-3min.
8. Transfer cell suspension (not containing settled debris) to a new 15mL conical tube already
containing 5mL of Earle’s medium + 3mL of Inhibitor Solution.
9. Pellet the cells by centrifugation for 7min at 300g.
10. Remove supernatant and resuspend the cells in 500uL to 1mL of 0.04% BSA/1xPBS.
11. Filter cells through a 0.22um filter basket.
12. Count cells using trypan blue.
13. Resuspend cells to a target concentration of 1000 cells/uL
14. Place cells on ice until ready for further experimentation.
!CRITICAL
It is important to note that organoids over the age of 3 months old are more sensitive to single cell
Papain dissociation. Specifically, Purkinje Cells often are too sensitive to survive the dissociation.

56
Therefore, it is recommended organoids be dissociated with the intent to obtain single nuclei in
lieu of whole-cells when aiming to conduct RNA-sequencing at a single cell resolution.
7.5.3 Option C: Calcium imaging of whole organoids: Timing 7-10 d
1. Place 3 organoids in one well of a 24 low attachment well plate with 150uL of CerDM3
media.
!CRITICAL
Add in the well enough media to cover the organoids. No excess media should be added to the
organoid as this will impact the success of the viral transfection.
2. Add 1.5uL of pAAV-CAG-SomaGCaMP6f2 by tilting the plate, letting the organoids sink
on the bottom of the well and releasing the volume directly on top of the organoids without
contacting them with the pipette tip.
If a calcium imaging experiment is performed together with optogenetic stimulation, add
1.5uL of L7-eOPN3 Lentivirus immediately after SomaGCaMP6f2 infection.
!CRITICAL
Carefully rotate the dish to prevent organoids from coming into contact with each other and to
avoid the fusion of organoids. Place the plate in the incubator on static conditions.
3. After 24hrs perform a full media change with 500uL of the same CerDM4 media used for
viral infection.
4. After 48hrs perform half media change with BrainPhys Medium.
5. After 72hr transfer the organoid to a 6-well low attachment plate with 2mL of BrainPhys
Medium. Perform half-media change with BrainPhys Medium every 2 days.
6. After 5 days from viral transfection, routinely check the appearance of
GreenFluorescentProtein (GFP) for SomaGCaMP6f2.
!TROUBLESHOOTING
7. Prepare organoids for calcium imaging by selecting the ones with the appearance of a
strong cell specific fluorescent signal. Add 5uL of Culturex to a Glass Bottom Microwell
Dish Promptly positioning the organoid over it with a 50 ul media droplet. Place in a static
condition in the incubator.
8. After 30 minutes, add 2mL of BrainPhys Optimized Medium and bring the organoid to the
2-photon microscope.

57
!CRITICAL
Transfer the organoid with minimal motion, using a styrofoam box, to prevent fluctuations in room
temperature.
9. Perform time-lapse recording of the organoids.
!CRITICAL
Be aware that the laser generates heat. Therefore, closely monitor the stage temperature to ensure
it does not exceed 37.5 degrees Celsius.
10. For optogenetic stimulation of L7-eOPN3, perform a 525nm LED laser stimulation for
500ms at maximal intensity. After stimulation, time-lapse record calcium responded as in
step 9.
11. Drugs can be added to the petri dish to modulate the organoid activity. Specifically, the
AMPA antagonist NBQX, the NMDA antagonist D-AP5, the GABA antagonist
Picrotoxin, the NMDA agonist NMDA and the Na+ channel blocker Tetrodotoxin (See
Atamian et al. for details on drugs combinations that can be used52)
!CAUTION
Pause the recording to avoid eye exposure to laser and carefully add the drug of interest in the
media by using gloves and tossing the used pipettes in decontaminating bleach. Be sure to follow
the institute guideline for specific rules on drug disposal. Wait 60 seconds to allow drug diffusion
and start recording again.
BOX 1 OPTOGENETIC MODULATION OF PURKINJE CELLS IN hCerOs
Intro (Timing 7-10 d)
Transducing hCerO with L7-eOPN3 provides a direct and efficient means to sparsely label
and induce expression of the inhibitory opsin (eOPN3) in Purkinje cells. This method
enables the modulation of their activity. When coupled with SomaGCaMP6f2 labeling and
subsequent calcium analysis, it elucidates the integration of Purkinje cells within the
network and assesses their level of maturation.
Additional reagents and materials
- L7-eOPN3 Lentivirus
Additional equipment

58
- 525 nm LED laser
Procedure
1. Perform lentiviral infection with L7-eOPN3 Lentivirus and pAAV-CAGSomaGCaMP6f2 as in step 2-6 in “Option C”. Start recordings upon the appearance of
strong Red Fluorescent Protein (RFP) for opsins expression.
2. Position the organoids in the imaging chamber as performed in step 7 and 8.
3. Place the LED laser in a position where it directs towards the organoid, covering the
exact area where imaging will take place.
4. Perform a 525 nm LED laser stimulation for 500ms. Perform time-lapse recording of the
organoids.
CRITICAL! Be aware that the laser generates heat. Therefore, closely monitor the stage
temperature to ensure it does not exceed 37.5 degrees Celsius.
7.5.4 Option D: Patch-clamp recording of whole organoids: Timing 7-10 d
1. Perform a viral infection with AAV8.L7-6.eGFP.WPRE.hBG to label Purkinje Neurons.
Follow step 1 to 6 of “Option C” for viral infection procedure.
2. Prepare organoids for Patch-clamp recordings by choosing those exhibiting a robust, cellspecific fluorescent signal.
3. Transfer a single organoid in the recording chamber and let the organoid sink at the bottom
of the recording chamber.
4. Focus by brightfield on the edge of the organoid and use the green channel for selecting
the cell to patch.
!TROUBLESHOOTING
5. Reach the seal with the patch pipette and perform current clamp recording after break-in.
(See our original publication for different types of patch-clamp recordings52).
6. Drugs can be added to the recording chamber to modulate the organoid activity (See our
previous publication for details on drugs that can be used52)
!CAUTION

59
Pause the recording to avoid noise signals and carefully add the drug of interest in the media by
using gloves and tossing the used pipettes in decontaminating bleach. Be sure to follow the institute
guideline for specific rules on drug disposal. Wait 60 seconds to allow drug diffusion and start
recording again.
8 TIMING
Step 1-8, Making EBs: 1-2 hr
Step 9-10, Feeding EBs and Initiation of germ layer differentiation: 4 d
Step 11-15, Induction of Isthmic organizing region: 11 d
Step 16-18, Shaking culture of expanding neuroepithelium: 13 d
Step 19-20, Growth and maturation of cerebellar tissue: >1 y
Step 20, Option A: Cryosectioning and immunostaining organoids: 2-3 d
Step 20, Option B: Dissociating organoids into single cells: 2-3 hr
Step 20, Option C: Calcium imaging of whole organoids: 7-10 d
Step 20, Option D: Patch-clamp recording of whole organoids: 7-10 d
Box 1, Optogenetic stimulation

60
CHAPTER 3
Human cerebellar organoids with functional Purkinje cells
Alexander Atamian1,2, Marcella Birtele1,2, Negar Hosseini1,2, Tuan Nguyen1,2, Anoothi Seth1,2,
Ashley Del Dosso1,2, Sandeep Paul3
, Neil Tedeschi3
, Ryan Taylor3
, Marcelo P. Coba4,5,6, Ranmal
Samarasinghe7
, Carlos Lois8
, Giorgia Quadrato1,2*
Affiliations:
1
Department of Stem Cell Biology and Regenerative Medicine, Keck School of Medicine,
University of Southern California, Los Angeles, CA 90033, USA
2
Eli and Edythe Broad CIRM Center for Regenerative Medicine and Stem Cell Research at
USC, Keck School of Medicine, University of Southern California, Los Angeles, CA 90033,
USA
3
Spatial Genomics, 145 Vista Avenue Suite 111, Pasadena, CA, 91107, USA
4
Department of Psychiatry and Behavioral Sciences, Keck School of Medicine, University of
Southern California, Los Angeles, CA 90033, USA
5
Zilkha Neurogenetic Institute, Keck School of Medicine, University of Southern California;
Los Angeles, CA 90033, USA
6
Department of Physiology and Neuroscience, Keck School of Medicine, University of
Southern California, 1501 San Pablo Street, Los Angeles, CA 90033, USA
7
Department of Clinical Neurophysiology and Neurology, University of California, Los
Angeles; Los Angeles, CA 90095, USA
8
Division of Biological and Biological Engineering, California Institute of Technology,
Pasadena, CA 91125, USA
Author contributions:
A.A., M.B., and G.Q. conceived the experiments. A.A., N.H., A.S., A.D. generated,
cultured, and characterized all organoids used in this study; A.A., T.N. performed scRNAseq experiments; T.N. performed scRNA-seq analysis and worked on cell type assignments
and data analysis with assistance from A.A.; S.P. performed sample preparation for spatial
transcriptomics while N.T. and R.T. performed analysis of the spatial transcriptomics;
M.B. N.H, and R.S. performed the calcium imaging and analysis; M.B. conducted whole-

61
cell patch clamp recordings and analysis; M.P.C. generated the D2 iPSC line; C.L.
designed and generated the constructs to optogenetically silence Purkinje neurons, and
produced lentiviral particles to infect organoids, G.Q. supervised all aspects of the project;
A.A., M.B., and G.Q. wrote the manuscript with contributions from all authors.
Published in Cell Stem Cell:
Atamian, Alexander, et al. Cell Stem Cell 31, 39-51.e6 (2024):
https://doi.org/10.1016/j.stem.2023.11.013
Graphical Abstract:
1 ABSTRACT
Research on human cerebellar development and disease has been hampered by the need for a
human cell-based system that recapitulates the human cerebellum's cellular diversity and
functional features. Here, we report a human organoid model (hCerOs) capable of developing the
complex cellular diversity of the fetal cerebellum, including human-specific rhombic lip
progenitor populations that have never been generated in vitro prior to this study. Two months old
hCerOs form distinct cytoarchitectural features, including laminar organized layering and create

62
functional connections between inhibitory and excitatory neurons that display coordinated network
activity. Long-term culture of hCerOs allows for healthy survival and maturation of Purkinje cells
that display molecular and electrophysiological hallmarks of their in vivo counterparts, addressing
a long-standing challenge in the field. This study therefore provides a physiologically relevant, allhuman model system to elucidate the cell type specific mechanisms governing cerebellar
development and disease.
2 INTRODUCTION
The cerebellum has been gaining substantial attention, given emerging evidence of its role in
cognitive functions, including language, spatial processing, working memory, executive functions,
and emotional processing, in addition to its well-described role in motor behaviors 9,108,109. In all
mammals, the cerebellum is initially derived from the alar lamina (dorsal) rhombomere 1 (r1)
region caudal to the isthmic organizer at the midbrain-hindbrain boundary. The r1 comprises two
distinct progenitor zones including the PTF1A+ ventricular zone (VZ) and the ATOH1+ Rhombic
Lip (RL). The VZ produces GABAergic neurons, including Purkinje Cells, molecular layer
interneurons (MLI), and deep cerebellar nuclei (iCN), while the more dorsally located RL gives
rise to glutamatergic neurons, including granule cells (GC), unipolar brush cells (UBC), and large
deep cerebellar nuclei projection neurons (eCN) 25–27,29–31.
Despite the conservation of early developmental stages of cerebellar ontogenesis across species,
human-specific traits, including altered neuronal subtype ratios and increased folial complexity,
have been identified when compared to non-human mammalian counterparts 9,38. In addition,
comparison between human and non-human primates has identified the presence of a humanspecific population of neural progenitors derived from the rhombic lip 39. This is important, as
abnormal development of the human rhombic lip is associated with multiple cerebellar disorders
including Dandy-Walker malformation, cerebellar vermis hypoplasia, and medulloblastoma, the
most common metastatic childhood brain tumor 110. Other cerebellar disorders including ASD and
ataxia have been associated with the degeneration of Purkinje cells, the main output neurons of the
cerebellar cortex 111–113. These disorders have mainly been studied in mouse models, which cannot
fully recapitulate human disease phenotypes. An example of this includes the inability of mice to
recapitulate the Purkinje cell loss phenotype observed in patients with ataxia telangiectasia 91,92.

63
Thus, there is a pressing need to develop an all-human cell-based culture system capable of
generating functional Purkinje cells.
Here we report an organoid model that allows for the first time healthy long-term survival and
maturation of functional cerebellar cells, including Purkinje neurons, in an all-human 3D context
addressing a longstanding challenge of the field. We show that hCerOs reliably generate spatially
segregated ventricular and rhombic lip progenitor zones, which give rise to the cellular diversity
of the human cerebellum within and across multiple cell lines. At two months in culture, we
observed the emergence of an organized laminar layering with inhibitory and excitatory cerebellar
neuronal subtypes, which are functionally connected and display coordinated activity, which
increases over a period of six months in culture. Finally, long-term culture of cerebellar organoids
allows the maturation of Purkinje cells that display the transcriptomic profile and
electrophysiological features of functional neurons.
3 RESULTS
3.1 HUMAN CEREBELLAR ORGANOIDS (hCerO)
REPRODUCIBLY GENERATE THE CELLULAR DIVERSITY OF
THE HUMAN CEREBELLUM
The cerebellum develops from the most rostral region of the rhombomere 1, an area of the
developing neural tube just caudal to the isthmic organizing center, which defines the midbrainhindbrain boundary. Modulation of key morphogen pathways has allowed for establishing
protocols for the differentiation of 3D cultures of human pluripotent stem cells into organoids
specific to particular brain regions including the cerebellum. However, it has been shown that
caudalization relying on only FGF2+insulin, commonly used to generate hCerOs 94,95,97,114 is
inferior to other patterning strategies in generating a hindbrain fate 115. This is not surprising, as
foundational neurodevelopmental biology studies have shown that the combinatorial activity of
several caudalizing factors is required in vivo to overcome the preference of neural tissue to
develop an anterior character 67. To develop a reproducible in vitro model of the cerebellum we
established an organoid protocol based on the modulation of the signaling molecules required for
the development of the isthmic organizer in vivo. Previous robust in vitro generation of
midbrain/hindbrain fates has incorporated the activation of WNT and FGF8 signaling 74,75,116,117.

64
Therefore, to promote neuroectodermal differentiation toward a cerebellar fate, cell aggregates
were simultaneously treated with the dual SMAD inhibitors Noggin and SB431542 22 and the
canonical WNT pathway activator CHIR–99021. On day 4, FGF8b was then added to specify the
midbrain-hindbrain boundary and suppress midbrain fate, re-creating the isthmic organizing center
seen in vivo (Fig. 1A,B). 71,72,100,118.
The use of this patterning strategy along with the use of insulin and growth factor reduced media
(gfCDM) 94,119 led to the successful elimination of forebrain progenitors, while the commonly
used FGF2+insulin patterning strategy, in our hands, yielded high levels of forebrain as observed
through the use of a forebrain reporter cell line (PGP::tdTOMATO::FOXG1) (fig. S1A-F, H, I,
K). Our patterning strategy with insulin and gfCDM produced a "rhombic lip" organoid that only
expressed markers of the excitatory lineage (ATOH1 and BARHL1). This organoid lacked
ventricular zone progenitors marked by the inhibitory lineage marker KIRREL2+ and postmitotic
Purkinje cell precursor derivatives marked by SKOR2 (fig. S1K). The ability to generate
BARHL1+ cells was also seen in the previously published FGF2+insulin strategy in gfCDM,
however, a significant level of forebrain was also detected (fig. S1I). It is known that base media
can affect regional patterning 119. We thus created a cerebellar induction strategy based on an
optimized combination of base media to lock in the midbrain/hindbrain boundary identity and to
generate a pure compendium of cerebellar cell types (fig. S1J, L, M-R). With this protocol, the
hCerOs properly caudalized with increased expression of caudal markers such as IRX3 and GBX2
(Fig. 1D; fig S1G). By day 30, the hCerOs reliably generated spatially segregated ventricular
(KIRREL2) and rhombic lip (ATOH1) progenitor zones (Fig. 1C; fig. S1S, T), as well as their
derivatives: excitatory BARHL1+ granule cell progenitors and newborn SKOR2+ inhibitory
Purkinje neurons (fig. S1L). Importantly, the ability to generate these two populations of neuronal
progenitors was identified across three hiPSC lines including PGP1, 11a, and D2, and one hESC
line, H9 (fig. S2A-D). Despite the H9 line having a tendency to favor dorsal forebrain development
when instructed to become ectoderm, it was still able to generate SKOR2+ Purkinje cell
progenitors, although it was not consistent in producing the more dorsal granule cell population
(fig. S2E). These data indicated the potential of our protocol to generate all the cell types of the
developing cerebellum across multiple hPSC lines.
We conducted single-cell RNA sequencing on 14,947 cells from 3 individual organoids that were
2 months old. Through VoxHunt, a tool used to assess cell type composition and developmental

65
stage of neural organoids 120, we mapped the 2-month-old hCerO’s scRNA-seq data onto the
BrainSpan human transcriptomic dataset and found that our dataset the highest scaled correlation
with cerebellum (CB) when compared to primary brain samples between 12-24 PCW (Fig. 1E;
fig. S3A). These findings have also been recently corroborated by an independent analysis,
contributing to the construction of a human brain organoid atlas 121, which was benchmarked
against the reference dataset from developing human brains 122. In order to understand the diversity
of cells present in our hCerOs, we systematically categorized the clusters by comparing signatures
of differentially expressed genes to a pre-existing single cell dataset of endogenous cell types of
the human developing cerebellum 37 (Fig. 1F, G; fig. S3B-J). Uniform Manifold Approximation
and Projection (UMAP) dimensionality reduction identified 16 major clusters, including the two
main progenitor types, the KIRREL2+, PTF1A+, and VIM+ ventricular zone progenitors, and
ATOH1+ and BARHL1+ rhombic lip progenitors, which derive the inhibitory and excitatory
neuronal populations. In vivo, the ventricular zone also gives rise to Bergmann glia, a scaffolding
cell type necessary to aid in the migration of maturing granule cells into their proper orientation
123. Indeed, within our datasets, we were also able to identify a cluster of cells that expressed key
Bergmann glial markers including PTPRZ1, EDNRB, GFAP, HOPX, SLCA4A4, and EDNRB.
Among the neuronal clusters, we identified cells expressing in vivo canonical markers of Purkinje
cells, including SKOR2, RORA, FOXP2, and CALB1. In addition, we classified three other
GABAergic (STMN2+, GAD2+) clusters: interneuron precursors expressing PAX2, molecular
layer interneurons expressing SOX14 and DMBX1, and inhibitory deep cerebellar nuclei with
markers MEIS2, LHX9, and IRX3. Additionally, the Purkinje cells and cerebellar nuclei had
corresponding “immature” clusters, which had a similar transcriptomic profile as their more
mature counterparts but lacked the expression of genes associated with later stages of maturation
in vivo. These include CALB1, FOXP2, and RORA for Purkinje cells, and NEUROD1/2 for
cerebellar nuclei. We also identified three types of glutamatergic neurons (STMN2+, SLC17A7+)
expressing Unipolar Brush Cell (UBC) marker EOMES, Granule Cell progenitor (GCP) markers
ATOH1 and BARHL1, and mature Granule Cell (GC) expressing higher levels of more mature
neuronal markers including NEUROD1 and NHLH1. We conducted a detailed analysis of the
expression level of the top 100 genes that characterize human Purkinje cells (37) and compared it
with the Purkinje cells present in our 2-month-old hCerOs. Our findings revealed that the Purkinje

66
cells in our hCerOs expressed a considerable number of these markers that define Purkinje cells,
which are observed between 9-20 PCW of the developing human cerebellum (fig. S4).
Previously reported work has revealed that the human rhombic lip is divided into a ventricular and
subventricular zone (RLvz and RLsvz), a feature that is not shared by any other vertebrates 39,40.
Because of the human-specific nature of our system, we then sought to identify these uniquely
human rhombic lip progenitor subsets. To determine if the cell types specific to the RLvz and
RLsvz are present in our system, we subclustered the RL and dividing VZ clusters, which were
then integrated with 1018 cells identified to have RL identity in the human developing cerebellum
dataset 37 (fig. S5A). We observed that our cells correlated well with the RLvz and RLsvz clusters
of the human dataset, indicating that we have both populations of progenitors present in our
organoids (fig. S5B-D). We then took the DEGs from the human dataset used to characterize the
RLvz, RLsvz, and the IZ. This comparison confirmed the transcriptomic similarity of hCerOderived cells to distinct sub-clusters found in the human dataset (fig. S5E-G). We also saw the
expression of classic RL markers, including MKI67, PAX6, and LMX1A, throughout our dataset,
as was seen in the human dataset. Additionally, the expression of WLS, SOX2, and CRAYAB was
predominantly expressed in the RLvz cluster, while higher levels of expression for EOMES,
DPYD, GPC6, OTX2, and BARHL1 were seen in the RLsvz. (fig. S5H).
In addition to the cell types deriving directly from the two progenitor zones of the cerebellum, we
also observed the ability of hCerOs to co-develop other cell types also found in the human fetal
dataset known to be important contributors to cerebellar development in vivo 124,125. These cell
types include TTR+ Choroid Plexus (TOP2A+ dividing choroid plexus), LUM+DCN+ Meninges,
and GDF7+ Roof Plate cells.
Then, we utilized this single-cell data set to look at the expression of risk genes linked with a
variety of disorders associated with cerebellar dysfunction, including intellectual disability (ID),
autism spectrum disorder (ASD), spinocerebellar ataxia (SCA), Joubert Syndrome, Alzheimer’s
disease, and cerebellar malformations to assess cell type enrichment in 2-month-old hCerOs (fig.
S6, Sup. Text) and found an overall enrichment in neuronal cell types.
Finally, to assess organoid-to-organoid reproducibility in cell type composition, we performed
scRNA-seq on three 2-month-old hCerOs (org1: 4,817 cells, org2: 6,768 cells, org3: 3,362 cells)

67
and found that these organoids can generate all the main cell types of the developing cerebellum.
(fig. S3I).
Altogether, our data show that our hCerO protocol reproducibly generates all the major cell types
of the developing cerebellum, including human-specific progenitor subtypes of the fetal rhombic
lip and Purkinje cells that express the transcriptional profile of their in vivo counterparts. Thus,
this model comprehensively allows for the examination of cell type-specific mechanisms
governing cerebellar development and disease.
3.2 hCerOs DISPLAY ORGANIZED LAMINAR LAYRING
REMNISCENT OF THE EGL AND PCL OF THE DEVELOPING
CEREBELLAR ANLAGE
While analysis of the cell type composition of hCerOs demonstrated that the generation of
cerebellar cellular diversity is an intrinsic process that does not require the in vivo cues of the
developing embryo, we next investigated whether distinctive cytoarchitectural features could also
organically emerge from this system. During the second trimester of human cerebellar ontogenesis,
the excitatory granule cell precursors undergo directed tangential migration along the pial surface
to form the external granule layer (EGL) (Fig. 2A) 126, meanwhile the Purkinje cells migrate
radially toward the pial surface, settle below the EGL, and form the multicell Purkinje layer (PCL).
Despite the presence of these neuronal populations, we found that this mechanism is not
intrinsically regulated in 2-month-old hCerOs, a stage roughly equivalent to the second trimester
of human fetal development. Thus, the 2-month-old hCerOs are unable to organize as seen in vivo.
Therefore, we decided to instruct their migration by using external cues.
During fetal development, SDF1a (also known as CXCL12), a chemoattractant released by
meningeal cells, guides the tangential migration of granule cell progenitors along the pial surface
through the activation of the corresponding CXCR4 receptor 127. In vivo CXCR4 is expressed in
granule cell progenitors 128, and as these cells mature to granule cells, they lose expression of the
CXCR4 receptor and consequently sensitivity to SDF1 and migrate radially away from the pial
surface 129. In agreement with in vivo expression patterns, we found that BARHL1+ granule cell
progenitors express CXCR4 in our 2-month-old hCerOs (Fig. 2B, C). It worthwhile to mention
that CXCR4 is also found in neurogenic progenitors and cerebellar ventricular radial glia in vivo

68
130–132 as is seen with the RNA expression of CXCR4 in our single cell data further corroborating
the similarity between the expression pattern of CXCR4 in vivo and in hCerOs. To assess the
response of hCerOs to SDF1, we added 100ng/ml of recombinant SDF1a into our maintenance
media every 3 days (CerDM3) for 1 month (between day 30-60). After treatment, we observed the
BARHL1+ and PAX6+ cells aligned adjacent to the edge of the organoid, while the SKOR2+ cells
appeared beneath that layer (Fig. 2D-I). When co-staining for the Purkinje cell zone (CALB1+)
and the RL-derivative zone (BARHL1+), we noticed that there appear to be two distinct layers in
the +SDF1a treated hCerOs (Fig. 2J-N), reminiscent of the apical-basal polarity seen in the early
stages of murine cerebellar development (~E15). Interestingly, a previous attempt to modify the
structural organization of hCerOs using 300ng/mL SDF1a for 7 days following FGF19 treatment,
yielded a reversed organization of the cerebellar anlage 94, suggesting the importance of the fine
tuning the duration and concentration of chemoattractants to build specific cytoarchitectrual
features within organoid models. It is important to note that the effect of SDF1 on the laminar
layering gradually started to disappear as soon as SDF1 was discontinued at day 60 and by the
time organoids were 6 months old we observed that granule cells and Purkinje cells were
intermingled (Fig. 2O). Altogether, we found that while laminar layering is not intrinsically
encoded in developing hCerOs, administration of development-inspired external cues at the right
concentration and timing, can aid in improving the ability of organoids to recapitulate features of
the anatomical organization found in the developing human brain.
3.3 hCerOs DISPLAY FUNCTIONALLY MATURE NETWORK
ACTIVITY IN LONG-TERM CULTURES, RESEMBLING
PATTERNS OF IN VIVO CEREBELLAR CIRCUITS
Following the characterization of the cellular diversity and cytoarchitectural features of hCerOs,
we interrogated whether spontaneously active networks were present. We performed live twophoton microscopy-based imaging of intact 2-month-old hCerOs, following infection with AAV8
GCAMP6f2 virus to record intracellular calcium dynamics (Fig. 3A). By performing an unbiased
analysis and categorization of single-cell calcium dynamics, we identified individual Tetrodotoxin
(TTX, a sodium channel blocker) sensitive Ca2+ neuronal calcium signals within hCerOs,
indicative of spontaneous neuronal activity (Fig. 3B-F) 133,134. Importantly, we found that levels
of spontaneous activity, as shown by the average activation per frame, amplitude, and duration of

69
calcium events, were similar across organoids derived from 3 hiPSC lines: D2, PGP1, and 11a
(fig. S7A-F, fig. S8). This further indicates that hCerOs display a high degree of organoid-toorganoid reproducibility, not only in terms of cellular composition (fig. S3J), but also at a
functional level.
In addition, the single cell calcium signals were then used to characterize brain organoid
physiological activity at network levels. We performed clustering analysis and found the presence
of multiple functional clusters spatially distributed across the entire organoid, suggesting that as
early as 2 months, neurons in hCerOs display coordinated firing indicative of the emergence of
spontaneous network activity (Fig. 3G, H).
We then investigated whether long-term culture of hCerOs would lead to changes in network
activity. Comparison of 2- and 6-month-old hCerO neuronal spiking profiles showed that 6-monthold organoids had an overall increase in average activation per frame and decrease in interpeak
time per neuron, indicating an increase in overall calcium transient activity (Fig. 3I, J) and more
neurons contributing to the activation of organoids. Additionally, we see an increase in the average
correlation coefficient (Fig. 3K) and increase in burstiness within clusters (Fig. 3L), suggesting a
transition from single mature neuronal events into spontaneous coordinated activity across
multiple neurons.
As hCerOs contain a mixture of glutamatergic granule cells/eCN/UBC and GABAergic Purkinje
cells/MLI/iCN/PIP cells, we sought to assess their distinct functional contributions through bath
application of NMDA and PTX to 2- and 6-month-old organoids. We found that combined
treatment of 6-month-old hCerOs with NMDA+PTX greatly increased burstiness in each cluster,
compared to treatment with NMDA alone. This indicates a contribution of both glutamatergic and
GABAergic neurons to the overall activity within clusters. The level of activity post-treatment
with NMDA+PTX is significantly higher than 2-month-old hCerOs, indicating an increase in the
level of functional connectivity in long-term cultured hCerOs (Fig. 3M-O). In addition, validation
through immunostaining confirmed the presence of neurons expressing Calbindin and GABA,
indicating that Purkinje neurons can release the GABA neurotransmitter (fig. S7H). We sought to
assess the active involvement of Purkinje cells in the network by introducing a lentiviral construct
targeting L7-PCP2 in 6-month-old hCerO. PCP2 is selectively expressed in Purkinje cells and has
been used previously to identify Purkinje cells in live human culture systems 74. We also introduced

70
the opsin eOPN3 135. Remarkably, LED stimulation inhibited Purkinje cells, leading to elevated
calcium activity levels in the organoid (Fig. 3P-R; Fig. S7I-K). These results strongly suggest that
Purkinje cells actively contribute to the network's function.
Overall, this data shows the emergence of functional network activity in hCerOs after as little as 2
months in culture. Long-term culture increases the level of functional connectivity between
excitatory and inhibitory neurons within hCerOs. The data provide, for the first time, a foundation
for the use of hCerOs to model functional impairment associated with dysfunction in the
development of cerebellar microcircuits.
3.4 FUNCTIONALLY MATURE HUMAN PURKINJE NEURONS
DEVELOP WITHIN LONG-TERM CULTURE OF hCerOs
To our knowledge, functional maturation of Purkinje cells has never been obtained in an all-human
culture system. Therefore, we focused our investigation on the functionality of these neurons,
which are dysfunctional in an array of neurological disorders. We identified Purkinje cells in 6-
month-old hCerOs through the expression of classical in vivo marker genes, such as PCP2, DAB1,
RORA, and FOXP2, (Fig. 4A-C; fig. S9A; table S3,4), via spatial transcriptomics, which also
gave us access to the spatial localization of these neurons (Fig. 4D; fig. S9B). Having identified
these neurons on the outer edge of the organoids, we decided to conduct whole cell patch clamp
recordings on intact hCerOs to prevent the disruption of the complex neuronal connections.
Following the transduction of hCerOs with a virus that fluorescently reports for the expression of
PCP2/L7, we performed whole-cell patch-clamp recordings of PCP2/L7+ neurons labeled with an
AAV8.L7-6.eGFP.WPRE.hBG (Fig. 4E), and found that PCP2/L7+ cells (n=19) at days 135–220
presented an average resting membrane potential (RMP) of -53.88 mV (table S5). 84.2% of the
recorded cells showed the ability to fire multiple mature induced action potentials with an average
amplitude of 68.80 mV (Fig. 4F,H). 42.10% of the recorded neurons presented the ability to fire
spontaneous TTX-sensitive action potentials with an average frequency of 14.63 Hz (Fig. 4G, I;
fig. S9C), highlighting the mature profile of the recorded neurons. Importantly, as observed in
cerebellar Purkinje neurons in vivo, we were able to detect the presence of Ih current (Fig. 4J,K)
and repetitive spontaneous firing (Fig. 4L), a profile distinctive of Purkinje cells 136,137, in our
PCP2/L7+ cells.

71
We conducted imaging of the Calbindin marker on whole hCerO organoids that were 6 months
old to determine the morphology of the Purkinje neurons. Our findings, which were in line with
calcium imaging analysis and patch-clamp data, showed the existence of Calbindin-labeled
Purkinje neurons and significant arborization and diverse morphology at the 6-month mark that is
similar to a human Purkinje cell in fetal tissue between 22-28PCW 138 (Fig. 4M). Postnatal mice
glia and granule cells have been previously shown to be necessary for the functional maturation of
hiPSC-derived Purkinje cells 74,94. Our data show, for the first time in an all-human system, the
development of functional Purkinje cells that express classical in vivo Purkinje cell markers and
display the distinct electrophysiological profile of their in vivo counterparts.
4 DISCUSSION
Our study establishes an organoid model to investigate the cellular interactions that orchestrate
development, homeostasis, and diseases of the human cerebellum. This is enabled by the
establishment of a protocol that reproducibly generates all the major cell and supporting cell types
that aid in overall cerebellar development, including choroid plexus and roof plate cells. This
demonstrates the existence of an intrinsic cellular program that regulates the acquisition of the
molecular identity and functional characteristics of the diverse repertoire of human cerebellar
neurons. This diverse compendium of cell types includes a human-specific progenitor subtype of
the Rhombic Lip (RLsvz) population that has never been generated in vitro prior to this study.
The generation of RLsvz in our study signifies that hCerOs can serve as a valuable tool for
advancing our understanding of cerebellar evo-devo biology and for exploring diseases associated
with disruptions in rhombic lip development. These diseases include conditions such as DandyWalker Syndrome 139 and pediatric cancers 81,82.
Furthermore, MLIs play a crucial role in shaping the spike activity of Purkinje cells within
cerebellar circuitry. Consequently, the presence of these cells within hCerOs, coupled with the
consistent and high degree of cellular diversity in our experimental protocol, could establish an
environment conducive to the functional maturation of Purkinje cells. Interestingly, our 6-monthold organoid-derived Purkinje cells resemble 22-28 weeks post-conception Purkinje cells from the
fetal human cerebellum 138, suggesting a similar developmental trajectory in vitro as observed in

72
vivo. This significant milestone, previously attainable only through the co-culture of mouse glia
and granule cells 74,75,94, has been challenging to achieve.
Although Purkinje cells are primarily affected in numerous neurodegenerative disorders, it is
increasingly clear that interactions with other cell types, including, but not limited to, glia cells,
may have a central role in disease pathogenesis 140. This further underscores the importance of
developing an all-human system.
As the link between cerebellar dysfunction and neuropsychiatric connectopathies, including ASD
and ID, becomes more evident 112,141, the generation of multiregional organoids that can model
long-range connections within the patients’ brains becomes a priority. In this scenario, hCerOs
with functional Purkinje cells, the main neuronal output of the cerebellum, will enable the
recapitulation of cerebellar connectivity with other brain regions and accelerate therapeutic
discovery.

73
5 FIGURES/TABLES
Fig. 1. Human cerebellar organoids (hCerOs) reproducibly generate the cellular diversity of
the human cerebellum. (A) Protocol schematic for generating hCerOs; ROCKi: ROCK
inhibition, SB: TGFB inhibition, Noggin: BMP inhibitor, CHIR (CHIR-99021 - GSK3 inhibitor),
FGF8b: Fibroblast Growth Factors 8b, T3: triiodothyronine. (B) Schematic of the developing
human cerebellar plate (sagittal view) EGL: external granular layer, isth: isthmic organizer, mb:
midbrain, NTZ: nuclear transitory zone, r1: rhombomere 1, RL: rhombic lip, SVZ: sub ventricular

74
zone, VZ: ventricular zone. (C) Immunofluorescence staining of KIRREL2+ ventricular zone
progenitors and ATOH1+ rhombic lip progenitors on 1 month old hCerO. SOX2 defines early
neuroepithelium. (D) Normalized mean qRT-PCR expression for region-specific markers among
various brain region specific organoids at Day 16. (E) VoxHunt analysis on neuronal subtypes
(GCP, GC, eCN/UBC, iCN, PIP, immature PC, PC, MLI) for PCW 12-24. STR: striatum, NCx:
neocortex, HIP: hippocampus, DTH: dorsal thalamus, CB: cerebellum, AMY: amygdala. (F)
Uniform Manifold Approximation and Projection (UMAP) plot of scRNA-seq data from 2-month
organoids after canonical correlation analysis (CCA) batch correction and alignment (D2: one
batch; n = 3 individual organoids combined). Left UMAP: combined organoids, colored by cell
types. Right UMAP: human dataset. Ast: astrocyte, BG: Bergmann glia, BS: brainstem, CHR:
choroid plexus, Div-: dividing, eCN/Unibrush: excitatory cerebellar nuclei and unipolar brush
cells, GC: granule cells, GCP: granule cell progenitors, iCN: inhibitory cerebellar nuclei, MEN:
meninges, MLI: molecular layer interneurons, OPC: oligodendrocyte progenitor cells, PC:
Purkinje cells, PGC: progenitor cells, PIP: pax2+ interneuron progenitors, RG: radial glia, RL:
rhombic lip, VZ: ventricular zone. (G) Violin Plot of key markers that identify major cell types of
the developing cerebellum in organoids vs human fetal tissue.

75
Fig. 2. hCerOs display organized laminar layering reminiscent of the external granule cell
layer (EGL) and the Purkinje Cell layer (PCL). (A) Schematic representation of in vivo human
development around Carnegie Stage 23 (CS23) [56 days post conception]. (B) Feature plot of
CXCR4 (receptor for SDF1a) within 2-month-old hCerOs and human dataset. (C)
Immunofluorescence of hCerOs BARHL1 (granule cell progenitors) and CXCR4 (SDF1a
receptor). scale bar: 200um. (D-I) Immunofluorescence of hCerOs BARHL1, SKOR2, and PAX6
(excitatory granule cell progenitor and marker) D-F (-SDF1a) and G-I (+SDF1a) are consecutive
sections. (J-M) Immunofluorescence of hCerOs BARHL1 (granule cell progenitors) and CALB1
(Purkinje cells) (N) Binning quantification of BARHL1+ cells. Student’s t-test was performed on

76
total of 12 regions for each condition from 3 independent experiments. (O) 6-month-old hCerO no
longer treated with SDF1a. ***p-value<0.001, ****p-value<0.0001. scale bar: 200um

77
Fig. 3. hCerOs display functionally mature network activity in long-term cultures,
resembling patterns of in vivo cerebellar circuits. (A) Schematic showing functional
characterization of 2- and 6-month-old hCerOs. (B) Representative 2-month-old 11a organoid
transduced with SomaGCaMP6f2. (C) Spontaneous calcium signal traces as ΔF/F in 2-month-old
11a-derived hCerOs. (D) Spontaneous calcium signal traces as pseudocolor heatmap. (E)
Representative heatmap as ΔF/F after bath application of 1μM TTX of a 2-months-old 11a
organoid. (F) Representative pseudocolor heatmap after bath application of 1μM TTX of a 2-
months-old 11a organoid. (G) Representative clustering identification on a 2-month-old hCerO.
(H) Quantification of number of clusters in a 2-month-old hCerO. Dots represent average value
for each recorded field; bars: standard deviation. (I) Quantification of average activation for frames

78
for each 2- and 6-month-old hCerOs. Student unpaired t test of ****p-value<0.0001. Dots
represent max average value for each recorded field; bars: standard deviation. (J) Quantification
of mean interpeak times for 2- and 6-month-old hCerOs. Student unpaired t test **p-value=0.0022.
Dots represent average value for each recorded field; bars: standard deviation. (K) Quantification
of correlation coefficient of calcium events in 2 and 6-month-old hCerO. Student unpaired t test
**p-value=0.0020. Dots represent average value for each recorded field; bars: standard deviation.
(L) Quantification of burstiness in each cluster for each 2 and 6-month-old hCerO. Student
unpaired t test ***p-value<0.0001. Dots represent average value for each recorded field; bars:
standard deviation. (M) Representative heatmap after bath application of 3μM NMDA of a 2-
months-old 11a organoid. (N) Representative heatmap after bath application of 3μM NMDA and
3μM PTX of a 2-months-old 11a organoid. (O) One way ANOVA comparison of burstiness in
each cluster for 2- and 6-month-old hCerOs treated with NMDA and NMDA+PTX. Dots represent
average value for each recorded organoid; bars: standard deviation. ***p-value=0.005, **pvalue=0.0065. (P) Schematic representation of 6-months-old hCerO transduction with AAV8-
SomaGCaMP6f and Lentivurus L7-eOPN3. Optogenetic stimulation was performed with 520 nm
LED. Calcium imaging was performed with 2-photon microscopy. (Q) Representative images of
maximum projection of stacks acquired during calcium imaging recording. (R) Analysis of
average intensity per frame from calcium imaging analysis in eOPN3- and eOPN3+ infected
organoids. Student paired t-test.***p-value=0.0001.

79

80
Fig. 4. Functionally mature, human Purkinje neurons develop within long-term culture of
hCerOs. (A) Schematic representation of spatial transcriptomics conducted on hCerOs. (B)
Differential expression of genes between cells identified as Purkinje and non-Purkinje cells in 6-
month-old hCerOs. Significant changes (p-value<0.05) are marked with solid circle. (C) UMAP
feature plots showing cell level expression of Purkinje cell marker genes identified with
overlapping expression of high CA8 levels. (D) Spatial localization of upregulated Purkinje cell
genes within a section of 6-month-old hCerO sample #2. (E) Upper panel: schematic of patch
clamp recording of PCP2+ neurons from intact hCerOs. Lower panel: representative PCP2+
neuron. Imaged with SP8-8X microscope. (F) Representative whole-cell patch-clamp trace of
induced AP from PCP2+ neuron from intact hCeOs. (G) Representative whole-cell patch-clamp
trace of spontaneous firing from PCP2+ neuron from intact hCerOs. (H) Quantification of peak
amplitude for AP elicited from each recorded cell. Mean + SD. Each dot represents a single AP
from an individual neuron. (I) Quantification of AP frequency from each spontaneously firing cell.
Mean + SD. Each dot represents a recorded neuron. (J) Quantification of Ih current for each Inward
Sodium/Outward Potassium Rectified currents. Mean + SD. Each dot represents a recorded
neuron. (K) Representative whole-cell patch-clamp trace of Ih current from PCP2+ neuron from
intact hCeOs. (L) Representative whole-cell patch-clamp trace of repetitive spontaneous firing
from PCP2+ neuron from intact hCerOs. (M) Drawings of human Purkinje cells at various
developmental stages (REF) compared to 6-month-old hCerO stained for CALB1. Imaged with
Leica Model TL LED Thunder widefield scope.

81
5.1 SUPPLEMENTARY FIGURES/TALBES

82
fig. S1. Human cerebellar organoid protocol optimization
(A-E) PGPFOXG1 derived organoids expressing various levels of FOXG1 depending on
organoid protocol: (A) Forebrain control (B) gfCDM+insulin (gfCDM+i) +FGF2 (C)
CerDM1+CHIR+FGF8b (D) gfCDM+i+CHIR+FGF8b (E) Day 6 switch (see Supp Figure 1M).
(F) Quantification of levels of FOXG1 red signal in Day 16 organoids of various protocols.
Forebrain control: n=18, gfCDM+i+FGF2: n=39, CerDM1+CHIR+FGF8b: n=26,
gfCDM+i+CHIR+FGF8b: n=26, Day 6 switch: n=26. (G) Day 16 qPCR of regional identity
markers across various brain region specific organoids normalized to forebrain organoid levels.
n=16 qPCR fold-change values for various brain region specific organoids. (H-L)
Immunofluorescence stain for BARHL1 and SKOR2 on various protocols (Supp Figure 1A-E).
(M) Schematic representation of gradual exposure to CerDM1 media after initial seeding in
gfCDM+i media. (N) Quantification on levels of FOXG1 red signal in Day 16 organoids using
PGPFOXG1 reporter cell line to derive organoids from Supp Figure 1M. (O-R) Day 16 qPCR of
markers identifying forebrain (FOXG1), midbrain/hindbrain boundary (FGF8), hindbrain
(GBX2), and midbrain (OTX2). (S) Representative organoid staining of consecutive sections for
KIRREL2 and ATOH1. (T) Quantification of superimposed images of germinal zones: ATOH1
and KIRREL2 stained in consecutive sections n=3. *p-value<0.05, **p-value<0.01, ****pvalue<0.0001

83
fig S2. Reproducible generation of cerebellar cells across cell lines
(A-D) The inhibitory and excitatory populations of the developing cerebellum are generated
reproducibly across cell lines (11a, D2, H9, PGP1). BARHL1: excitatory granule cell
progenitors; SKOR2: inhibitory post-mitotic newborn Purkinje cells. (E) quantification of
percentage of BARHL1 and SKOR2 cells across various cell lines from 3 organoids from 3
different batches.

84
fig. S3. Single cell RNA-seq analysis of 2-month-old organoids
(A) VoxHunt similarity analysis on 2-month-old hCerOs by cell type and age (pcw). (B)
Unbiased clustering of D2 and fetal datasets generating 25 clusters. (C) The major cell cluster
groupings of D2 organoid cells (15,304 cells total). (D) Feature plots of main markers which

85
identify major cell types including granule cells, purkinje cells, rhombic lip, ventricular zone,
pax2+ interneuron precursors, and cerebellar nuclei. (E-I) Excluded organoid cells that coclustering with fetal tissue cell types because they did not express the key cell types that identify
that cluster of cells. (357 cells total). (J) Ratios of each defined clusters colored by population,
which shows representation of cell types between 3 separate hCerOs (org1: 4,817 cells, org2:
6,768 cells, org3: 3,362 cells).
fig. S4. Top 100 Purkinje Cell DEGs from human fetal dataset compared to 2-month-old
hCerO.

86
fig. S5. Identification of human specific rhombic lip ventricular and subventricular subsets.
(A) UMAP unbiased clustering and DEG-based annotation of the RL and dividing-VZ subcluster
(n=1629 cells; n=788 for SVZ; n=645 for VZ, n=196 for IZ). IZ, intermediate zone; SVZ,
subventricular zone; VZ, ventricular zone. (B) UMAP visualization of cells grouped by sample
(n=611 for D2 and n=1018 for fetal) (D2 vs fetal). (C,D) UMAP visualization of cells split by
sample (D2 vs fetal). (E-G) Heatmap showing the organoid cells’ expression of human single
cell DEGs from Aldinger et al 2022 used to identify IZ, SVZ, and VZ. (H) Dot plot showing the
expression of selected marker genes in each subcluster used in Aldinger et al 2022.

87

88
fig. S6. hCerO cell type enrichment in various diseases.
(A-F) Heatmaps of mean expression per organoid cerebellar cell type for genes associated with
various diseases (A, intellectual disability; B, autism spectrum disorder; C, spinocerebellar ataxia;
D, Joubert syndrome; E, Alzheimer’s disease; F, cerebellar malformations). Color scheme is based
on z-score distribution. In the heatmaps, each row represents one gene, and each column represents
a single cell type. Gene expression was clustered by row. Enrichment p-values (-log10P value) for
each cell type are shown in the bottom bar plots. Significance was determined by one-sample ztest, two-tailed p-value. The horizontal line in the bar plots represents the 0.05 significance
threshold.

89
fig. S7. Reproducibility of functional hCerO across cell lines.
(A) 2-month-old organoids from two different cell lines (D2 and PGP1) transduced with
SomaGCaMP6f2. Scale bar: 200 μm. (B) Spontaneous calcium signal traces for each example
cell as ΔF/F. Representative of calcium signal per cell line (D2, PGP1). (C) Spontaneous calcium
signal traces as heatmap. Representative of calcium signal per cell line (D2, PGP1). (D)
Quantification of average activation for frames for each 2-month-old hCerO. One way ANOVA
comparison of 9 orgs (11a), 9 orgs (D2), 9 orgs (PGP1). Dots represent average value for each
recorded organoid; bars: standard deviation. (E) Quantification of average amplitude for each 2-
month-old hCerO. One way ANOVA comparison of 9 orgs (11a), 9 orgs (D2), 9 orgs

90
(PGP1). Dots represent average value for each recorded organoid; bars: standard deviation. (F)
Quantification of average event length for each 2-month-old hCerO. One way ANOVA
comparison of 9 orgs (11a), 9 orgs (D2), 9 orgs (PGP1). Dots represent average value for each
recorded organoid; bars: standard deviation. (G)Analysis of number (%) of active ROIs in
baseline and TTX condition of 2-months-old hCerO. (H) Immunohistochemistry of 6 months-old
hCerO for Calbindin (CALB1) and GABA. (I) Average intensity per frame analysis of 2-
months-old hCerO in baseline, NMDA, NMDA+AP5, and NMDA+AP5+PTX. (J) Schematic
representation of 6-months-old hCerO transduction with AAV8-SomaGCaMP6f only without
Lentivirus L7-eOPN3. Optogenetic stimulation was performed with 520 nm LED. Calcium
imaging was performed with 2-photon microscopy. (K) Representative images of maximum
projection of stacks acquired during calcium imaging recording before and after LED
stimulation.

91
fig. S8. GCaMP parameters across cell lines and between organoids
(A) 2-month-old organoids from PGP, 11a, and D2 cell lines (B) 6-month-old organoids from
PGP, 11a, D2 cell lines.

92
fig. S9. Spatial Transcriptomics of 6-month-old organoids
(A) UMAP of significantly enriched genes in the Purkinje cell cluster of 6-month-old hCerOs.
n=2 organoids. (B) Spatial localization of significantly enriched genes in PCs in one 10uM
section of 6-month-old organoid. n=2 organoids. (C) Spontaneous action potential without (left
panel) and with 1uM TTX (right panel).

93
Table S1.
DEGs for the identified clusters of organoid dataset. See original manuscript.
Table S2.
DEG for organoid cells for SVZ, VZ, and IZ clusters. See original manuscript.
Celltype Sample ROI Cell Count
PC 1 ROI2 21
PC 1 ROI3 20
PC 1 ROI4 9
PC 1 ROI5 0
PC 2 ROI1 55
PC 2 ROI2 39
PC 2 ROI3 51
PC 2 ROI4 33
PC 2 ROI5 23
non-PC 1 ROI2 3183
non-PC 1 ROI3 1839
non-PC 1 ROI4 856
non-PC 1 ROI5 155
non-PC 2 ROI1 2613
non-PC 2 ROI2 3820
non-PC 2 ROI3 3758
non-PC 2 ROI4 2489
non-PC 2 ROI5 1121
ambiguous 1 ROI2 1016
ambiguous 1 ROI3 894
ambiguous 1 ROI4 315
ambiguous 1 ROI5 50
ambiguous 2 ROI1 226
ambiguous 2 ROI2 398
ambiguous 2 ROI3 379
ambiguous 2 ROI4 268
ambiguous 2 ROI5 126
Table S3.
Purkinje cell vs other cell type identification and counts in 6-month-old sample per ROI.

94
Table S4.
Purkinje Cell vs other cell enrichment test statistics.

95
Table S5.
Passive properties of the membrane for PCP2+ neurons recorded by whole-cell patch-clamp
recordings.
Table S6.
qPCR primers
Gene Name Forward Primer (5’-3’) Reverse Primer (5’-3’)
EN1 GAGCGCAGGGCACCAAATA CGAGTCAGTTTTGACCACGG
EN2 CCGGCGTGGGTCTACTGTA GGCCGCTTGTCCTCTTTGTT
FOXG1 CCTGCCCTGTGAGTCTTTAAG GTTCACTTACAGTCTGGTCCC
GBX2 GACGAGTCAAAGGTGGAAGA
C
GATTGTCATCCGAGCTGTAGTC
HOXA2 CGTCGCTCGCTGAGTGCCTG TGTCGAGTGTGAAAGCGTCGA
GG
HOXA5 TCATAGTTCCGTGAGCGAGC ATCCATGCCATTGTAGCCGT
IRX3 GGCTTGCGCCCCGTAGAAAT
GT
AGGAGCCAGGTCAGGTCCGAA
C
OTX2 AGAGGACGACGTTCACTCG TCGGGCAAGTTGATTTTCAGT
PAX8 ATAGCTGCCGACTAAGCATT
GA
ATCCGTGCGAAGGTGCTTT

96
SIX3 ACCGGCCTCACTCCCACACA CGCTCGGTCCAATGGCCTGG
Table S7.
Primary antibodies used for immunofluorescence
Primary
Antibody
Host Manufacturer Catalog Number Dilution
BARHL1 Rabbit Proteintech HPA004809 1:500
SKOR2 Rabbit Proteintech HPA046206 1:500
KIRREL2 Rabbit Proteintech 10890-1-AP 1:100
SOX2 Goat RD Systems AF2018 1:500
ATOH1 Rabbit Abclonal A11477 1:200
CALB1 Mouse Sigma-Aldrich C9848-100UL 1:300
PAX6 Rabbit Biolegend 901301 1:200
Movie S1.
Representative 2-month-old hCerO spontaneous calcium activity recording.
Movie S2.
Representative 6-month-old hCerO spontaneous calcium activity recording.
Movie S3.
Representative 6-month-old hCerO spontaneous baseline calcium activity recording without
Opsin activation.

97
Movie S4.
Representative 6-month-old hCerO spontaneous calcium activity recording after Opsin
activation.
6 MATERIALS AND METHODS
6.1 PLURIPOTENT STEM CELL CUTLTURE
The PGP1 (Personal Genome Project 1) human iPS cell line was from the laboratory of G. Church;
PGP1-FOXG1 human iPS cell lines were obtained from Harvard Stem Cell Institute; the H9
human ES cell line was purchased from WiCell, the 11a human iPS cell line was from the
laboratory of Kevin Eggan, and the D2 human iPSC cell line was from the laboratory of Marcelo
Coba. All stem cell lines were cultured on Geltrex (Gibco)-coated tissue culture plates, using
mTeSR1medium (Stem Cell Technologies) with 100 U/ml penicillin and 100 µg/ml streptomycin
(Corning) at 37ºC in 5% CO2. All human stem cells were maintained below passage 50 and
periodically karyotyped via the G-banding Karyotype Service at Children’s Hospital Los Angeles
and were negative for mycoplasma (MycoAlert Plus Mycoplasma Detection Kits).
6.2 ORGANOID DIFFERENTIATION
For differentiation, feeder-free cultured human stem cells, 80% confluent, were dissociated to
single cells using room temperature Accutase (Sigma) and were reaggregated in ultra-lowattachment, V-bottomed, 96-well plates (Sbio) at 6,000 cells per 100 ul per well, in a growth factor
reduced chemically defined medium (gfCDM), based on the Wataya119, consisting of 1:1 IMDM
(Gibco) and F-12 (Gibco), 5mg/ml >99% Bovine Serum Albumin (BSA) (Sigma), 1% chemically
defined lipid concentrate (CDLC) (Gibco), 15ug/ml Apo-transferrin, and 450uM MonoThioglycerol (Sigma) with 7ug/ml insulin (Sigma), 10 µM of SB431542 (Selleckchem), 50 ng/ml
Noggin (Peprotech), 1.7 µM CHIR (Reprocell), and 20 µM Y-27632 ROCK inhibitor (Stem Cell
Technologies). On Day 2, the concentration of Noggin was increased to 100 ng/mL by removing
40uL of the media and adding 50uL containing 150 ng/ml Noggin in addition to the same
concentration of the previous morphogens. To initiate the specification of the midbrain/hindbrain
boundary, 40ul of media was removed and 50ul of 200ng/mL FGF8b was added to each well to
bring the final concentration in the well to 100ng/mL at Day 4. Although SB431542, CHIR, and
Noggin were still added at the same concentration on Day 4, ROCKi was no longer added to

98
promote cell proliferation and aggregate growth. To promote the cerebellar cell fate, on Day 6,
40ul of media was removed and 50ul of fresh cerebellar differentiation medium (CerDM) I,
containing DMEM/F-12 (Corning/Cytiva), 20% Knockout Serum Replacement (KSR) (Gibco),
15 µg/ml Apo-Transferrin (Sigma), 7 µg/ml Insulin (Sigma), 2 mM Glutamax (Gibco), 0.1 mM 2-
Mercaptoethanol (Gibco), 100 U/ml penicillin and 100µg/ml streptomycin (Corning) was added
to each well while supplementing the media with the same final concentration of morphogens as
Day 4. On Day 8, the final concentration of FGF8b was increased from 100ng/ml to 300ng/ml to
push further toward the midbrain/hindbrain specification and 100ul of media was added to each
well after the removal of 40ul taking the final volume to 150ul. On Day 10, 80ul of media was
removed and supplemented with 100ul of media with the same final concentration of the drugs
from Day 8. On Day 12 and 14, half media change was conducted without any drugs by removing
70ul of media and adding 75ul of fresh CerDM1.
From Day 16 to Day 30, organoids were cultured in 10-cm culture plates (Falcon) with orbital
agitation (10 r.p.m) in CerDM II, containing DMEM/F12 medium (Corning), 1% N2 (Gibco), 1%
B27 (Gibco), 2 mM Glutamax (Gibco), 100 U/ml penicillin and 100 µg/ml streptomycin (Corning)
and full media change was conducted every 3 days. From Day 30 to Day 60, organoids were
cultured in CerDM III, consisting of 1:1 DMEM/F12 (Corning) and Neurobasal medium (Gibco),
1% N2, 1% B27, 2 mM Glutamax, 5 µg/ml Heparin (Sigma), 1% Chemically Defined Lipid
Concentrate (Gibco), 100 U/ml penicillin, 100µg/ml streptomycin (Corning), 0.25 μg/ml AmphoB
(Gibco), 0.5 ng/ml T3 (Sigma) and 1% Matrigel (Corning). From day 60 onwards, organoids were
cultured in CerDM IV, consisting of CerDM III without Matrigel and supplemented with 14 ng/ml
Brain Derived Neurotrophic Factor (BDNF) (R&D Systems). Full media change was conducted
once a week on organoids after day 30.
The differentiation protocols for forebrain, midbrain, and spinal cord organoids used in regional
identity qPCR experiments were adapted from previous publications57,142.
6.1 QUANTITATIVE PCR
Organoids were lysed and the RNA was collected using the RNeasy Mini kit (Qiagen, #74004).
Template cDNA was prepared by reverse transcription using the RevertAid First Strand cDNA
Synthesis Kit (Thermo Fisher Scientific, #K1621). qPCR was performed using the SYBR Green

99
PCR Master Mix (Thermo Fisher Scientific, #4309155) on a ViiA7 Real-Time PCR System
(Thermo Fisher Scientific). Primers used in this study are listed in Table 6.
6.4 IMMUNOHISTOCHEMISTRY
Organoids were fixed in 4% paraformaldehyde for 30 min at room temperature before overnight
incubation at 4C in 30% sucrose solution. Organoids were then embedded in Tissue-Tek O.C.T.
compound (Sakura, #62550) and cryosectioned at 20μm thickness onto glass slides (Globe
Scientific, #1354W). Slides were washed 3x with a 0.1% Tween20 (Sigma #P9416) solution
before a 1-hour incubation in 0.3%TritonX-100 (Sigma, #T9284) and 6% bovine serum albumin
(Sigma, #AA0281) solution. Slides were incubated for 1-hour at room temperature in a primary
antibody solution followed by a 1-hour room temperature incubation in a secondary antibody
solution, both consisting of 0.1% TritonX-100 and 2.5%BSA with 3 washes before and after
secondary antibody incubation. Slides were coverslipped using Fluoromount G (EMS, #50-259-
73). Primary antibodies and dilutions used are specified in Table 7.
6.5 WHOLE ORGANOID STAINING
Organoids were fixed in 4% paraformaldehyde for 30 min at room temperature before a 1-hour
incubation in 0.3%TritonX-100 (Sigma, #T9284) and 6% bovine serum albumin (BSA) (Sigma,
#AA0281) solution at 4C with gentle movement. Organoids were then incubated in primary
antibody solution for 48 hours at 4C with gentle movement followed by a 48-hour incubation in a
secondary antibody solution for 48 hours at 4C with gentle movement, both consisting of
0.1%TritonX-100 and 2.5%BSA with several rounds of washing over a period of 8 hours in a
0.1%Tween20 (Sigma #P9416) solution before and after secondary antibody incubation.
6.6 MICROSCOPY AND IMAGING ANALYSIS
Organoids in culture were imaged using a Leica DMIL LED microscope (Leica).
Immunofluorescence images were taken with a Leica Model TL LED Thunder widefield scope
(Leica) and analysed using ImageJ software and MATLAB software. Whole organoids were
imaged using a SP8-8X microscope and processed with Leica LASX software.
6.7 CALICUM IMAGING
Organoids were transduced with pAAV-CAG-SomaGCaMP6f2 (Addgene, #158757) using 1.5 ul
of virus in 150 ul of CerDM3 medium overnight on static condition. One day after infection, a full
media change was performed using CerDM3 media. A full media change to BrainPhys Medium

100
(STEMCELL Technologies, #05792) was performed on day three after transduction. Organoids
were kept for at least one week in culture before the imaging session. Organoids were randomly
selected and transferred to a recording chamber kept at 37 °C using a heating platform and a
controller (TC-324C, Warner Instruments) in BrainPhys Optimized Medium (STEMCELL
Technologies, #05796). Imaging was performed using a SP-8X microscope with a multiphoton
laser. Time-lapse images were acquired at 1 frame for 860 ms, using a 25x 0.95 NA water objective
(2.5 mm WD) and resulting in a view of 200 x 200 µm2
. All imaging conditions including
excitation light intensity, camera sensor gain, and exposure time were identical for all calcium
imaging experiments.
Basal activity was recorded for 10 mins of the imaged organoid. Pharmacological treatment was
performed with a bath application of Tetrodotoxin, TTX (Tocris, #1078/1), at a final concentration
of 1 µM, 3µM NMDA (Tocris, #0114) and 3µM PTX (Hello Bio, #HB0506). Raw tiff calcium
imaging files were analyzed using the CNMF and CalmAN package as previously described
(CaImAn an open-source tool for scalable calcium imaging data133 to identify fluorescent
transients and spike estimation in MATLAB (MathWorks). Calcium traces were plotted as relative
scaled height in function of time. Hierarchical clustering and pairwise correlation was performed
as previously143 with Correlation, linkage parameter was set to 1.5.
6.8 OPTOGENETICS AND 2-PHOTON IMAGING
A fragment of the L7 promoter (854 bp) was cloned inside of the FUGW lentiviral backbone. The
silencing opsin eOPN3 fused to gfp144 was cloned downstream from the L7 promoter in the
lentiviral vector, named L7 eOPN3. Viral particles were produced and concentrated by
ultracentrifugation essentially as described in Lois 2002, Development144. 6-month-old hCerO
were transduced with L7-eOPN3 Lentivirus and pAAV-CAG-SomaGCaMP6f using 1.5 µl of each
virus into 150 µl of CerDM3 medium overnight in static conditions. Control organoids from the
same differentiations were transduced at the same time using only pAAV-CAG-SomaGCaMP6f.
Media changes and organoid selection were performed as previously described for SomaGCaMP6f
transduction. To activate eOPN3 we used a 525 nm LED stimulation for 500ms at maximal
intensity (as in Mahn 2021, Neurons). As previously described eOPN3 activation lasts for 5
minutes after stimulation, to monitor calcium activity we performed 2P-imaging recordings at 880
nm for 5 minutes immediately after eOPN3 activation. It is important to note that the light intensity
used to image gcamp under the 2P microscope is not sufficient to activate ePON3. ePON3 is only

101
activated with a much stronger light intensity that cannot be provided with a 2P microscope. Thus,
to activate the sufficient light intensity required to activate eOPN3 we used an LED at 525nm
(10mW/mm2
).
6.9 WHOLE-CELL PATCH CLAMP RECORDING
Organoids were transduced with AAV8.L7-6.eGFP.WPRE.hBG (Addgene, #126462) using 1.5 µl
of virus in 150 µl of CerDM3 medium overnight on static condition in a single well of a 24 well
low attachment plates. One day after transduction, a full media change was performed using
CerDM3 media. On day three after transduction organoids weres transferred to a 6 well low
attachment plate and a full media change to BrainPhys Medium (STEMCELL Technologies,
#05792) supplemented with NeuroCult SM1 Neuronal Supplement (STEMCELL Technologies,
#05711 was performed. Organoids were kept for at least one week in culture before the recordings.
Prior to recording, each organoid was transferred into 35mm petri dishes (Ibidi, #80136) on a 10
µl geltrex drop (Thermo Fisher Scientific, #A1413301) and let for 15 mins in a 37°C incubator.
Afterwards, organoids were incubated in BrainPhys Optimized Medium (STEMCELL
Technologies, #05796) until recording. PCP/L7+ neurons were visualized under a fluorescence
microscope (Olympus BX51 WI). Recordings were performed at RT. Bath application of
Tetrodotoxin, TTX (Tocris, #1078/1), at a final concentration of 1 μM was performed to assess
presence of TTX-sensitive APs in Current-clamp mode. Multi-clamp 700B (Molecular Devices)
was used for recordings and signals were acquired at 10 kHz using pClamp10 software and filtered
at 2 kHz for voltage clamp recordings. Data acquisition was performed with Digidata 1440A
(Molecular Devices). Borosilicate glass capillaries were used for patch pipettes ranging between
4.5-8 MOhm. Patch pipettes were filled with intracellular solution (in mM): 122.5 potassium
gluconate, 12.5 KCl, 0.2 EGTA, 10 Hepes, 2 MgATP, 0.3 Na3GTP and 8 NaCl adjusted to pH 7.3
with KOH. Intracellular solution was kept on ice during recordings. Passive properties of the
membrane were monitored immediately after break-in in current clamp mode. Membrane potential
was kept at -70 mV and currents were injected from -100 pA to + 100 pA with 10 pA increments
to induce action potentials. Inward sodium and delayed rectifying potassium currents were
measured in voltage clamp at depolarising steps of 10mV. Recording data were analyzed by using
pClamp 11 software (Molecular Devices). Baseline gap free signals acquired in current mode were
manually adjusted and frequency calculated with automated event detection with event rejection
set to 1 ms and 67.26 dV. Gap free traces acquired in voltage clamp mode were used to monitor

102
postsynaptic currents and filtered with a low-pass 2000 Hz Gaussian filter during post hoc traces
analysis.
6.10 DISSOCIATION OF hCerOs AND SINGLE-CELL RNA
SEQ
Individual brain organoids were dissociated into single cells using the Worthington Papain
Dissociation System kit (Worthington Biochemical). To give an estimated recovery of 6,000 cells
per channel dissociated cells were resuspended in ice-cold PBS containing 0.04% BSA at a
concentration of 1,000 cells/μl, loaded onto a Chromium Single Cell 3′ Chip (10x Genomics) and
processed through the Chromium controller to generate single-cell gel beads in emulsion (GEMs).
scRNA-seq libraries were prepared with the Chromium Single Cell 3′ Library & Gel Bead Kit v.2
(10x Genomics). Libraries from different samples were pooled based on molar concentrations and
sequenced on a NextSeq 500 instrument (Illumina) with 26 bases for read 1, 57 bases for read 2
and 8 bases for index 1. After the first round of sequencing, libraries were re-pooled on the basis
of the actual number of cells in each and re-sequenced to give an equal number of reads per cell in
each sample and to reach a sequencing saturation of at least 50% (in most cases >70%).
6.11 SINGLE-CELL RNA-SEQ DATA ANALYSIS
Bioinformatic analysis was performed using the Seurat R package. Briefly, all single cell
transcriptomes from 3 organoids were combined, and a set of highly variable genes was
determined. The initial PCA analysis was performed using this gene set and the number of
statistically significant principal components (PCA) was identified via Jackstraw analysis. Cluster
structures were remapped to two dimensions using Uniform Manifold Approximation and
Projection (UMAP). Finally, a graph-based clustering approach was used to cluster the cells to
produce putative cell-type clusters and return sets of differentially expressed marker genes for each
cluster. These were compared to known marker genes to ascribe a cellular identity to each cluster.
This data was then integrated with the human dataset to confirm cell type identifications.
6.12 DETAILED EXPLANATION OF DATA INTEGRATION
AND ANALYSIS
For co-clustering of our hCerOs with human fetal samples, we merged the organoid data set with
a downsampled fetal data set using the Seurat pipeline based on Canonical Correlation Analysis
(CCA) integration. First, to downsample the fetal data set, we randomly took 500 cells from each

103
of 21 cell types identified in the fetal data set, (01-PC to 21-BS Choroid/Ependymal). Next, we
used the SCTransformation function to both normalize and regress out cycling cells based on the
difference between the G2M and S phase scores. We recovered features for integration with the
SelectIntegrationFeatures function (features = 3000 genes) and PrepSCTIntegration. Finally, we
found the anchoring features with the FindIntegrationAnchors function using
normalization.method = "SCT" and anchor.features = features, and integrated the two datasets with
the IntegrateData set function.
For the remainder of the analysis, we performed the standard Seurat pipeline to generate PCA and
UMAP coordinates before using the FindNeighbors and FindClusters functions (resolution = 0.2,
0.8 and 1.2), to define cell-types clusters. Each cell-type cluster was classified based on known
markers and co-clustering with the fetal data.
To calculate the Markers genes, we used the FindMarker function with the test.use attribute set to
the default Wilcoxon rank sum test.
6.13 SEQFISH (SPATIAL TRANSCRIPTOMICS)
RNA expression data from multiple sections from two 6-month-old organoids were obtained on
the Spatial Genomics platform. Briefly, sections were hybridized overnight at 37 °C with probes
against a set of RNA transcripts for Purkinje cell and other cell type specific markers. Samples
were washed the following day and loaded onto the automated Spatial Genomics microscope for
imaging. Threshold values were manually chosen to select transcript dots with signal above
background. Cells were segmented based on expanding the nuclear DAPI stain. Cell-by-gene
count matrices were then used for further downstream analyses. Raw counts of transcripts were
used for all analyses.
6.14 SINGLE-CELL PRE-PROCESSING (SPATIAL
TRANSCRIPTOMICS)
The cell-by-gene matrix of counts was pre-processed using the scanpy pre-processing module. For
sample 1, cells with less than 1 or greater than 1500 transcript counts were removed. For sample
2, cells with less than 30 or greater than 1500 transcript counts were removed; the higher minimum
count threshold was used to remove likely necrotic cells in the center of the organoid.
6.15 IDENTIFICATION OF PURKINJE CELLS (SPATIAL
TRANSCRIPTOMICS)

104
Purkinje cells (PCs) were identified using CA8 expression. Cells with CA8 counts greater than or
equal to the 99th quantile were defined as PCs; cells with zero CA8 counts were defined as nonPCs and cells with CA8 counts between zero and the 99th quantile were defined as ambiguous.
6.16 DIFFERENTIAL EXPRESSION BETWEEN PURKINJE
AND NON-PURKINJE CELLS (SPATIAL
TRANSCRIPTOMICS)
We tested for differential expression of gene transcripts between Purkinje cells and non-Purkinje
cells as identified above. Ambiguous cells were removed. To avoid psuedo-replication we
aggregated cellular transcript counts within samples prior to fitting a gamma-Poisson generalized
linear model with glmGamPoi 145. We accounted for the difference in cell size between Purkinje
cells and non-Purkinje cells by dividing cell counts by cell area prior to aggregation. Total
transcript counts are often included as a size factor in RNA-seq based psueudo-bulk analysis. This
accounts for variation in sequencing depth and cell count per sample. However, Spatial Genomics
transcript counts are image based and therefore transcripts are not resampled. Size factors for the
model were the number of cells per pseudo-bulked sample.
6.17 SUPPLEMENTARY TEXT
hCerO cell type enrichment in various diseases
The cerebellum’s protracted development makes it susceptible to early neurodevelopmental
disorders, and it is also vulnerable to adult-onset degenerative diseases. To validate the fidelity of
hCerOs as a viable in vitro model to assess the selective vulnerability of distinct cerebellar cell
types to cerebellar disorders, we crossed our 2-month-old hCerO single cell dataset with genes
associated with intellectual disability (ID), autism spectrum disorder (ASD), spinocerebellar ataxia
(SCA), Joubert Syndrome, cerebellar malformations, Alzheimer’s disease (AD). The gene lists
associated with these disorders were obtained from Aldinger et al37.
First, we assessed gene enrichment for intellectual disability (ID) (14) and found enrichment in
nearly all neuronal cell types. In addition to finding enrichment within inhibitory neuronal clusters
such as PIP, MLI, immature iCN, iCN, Purkinje Cells (PC), and immature PC, the ID gene set was
also enriched in the excitatory granule cells (GC) with high levels of expression of SOX5, SATB2,
and KCNQ3. Of the ID gene set, notable expression of a subset of these genes was observed within

105
the MLI (EHMT1, ZC4H2, PPKAR1A, PPP1CB, QRICH1, KCNH1, U2AF2, NALCN, GNAI1,
KCNB1, CHD2, CLTC, NSD1, CYP27C1, and GOLPH3).
Next, we examined the enrichment of autism spectrum disorder (ASD), a disorder known to result
in structural and functional cerebellar abnormalities146. The risk gene set was most significantly
enriched in the neuronal cell types including PIP, molecular layer interneurons (MLI), inhibitory
cerebellar nuclei (iCN), immature iCN, Purkinje Cells (PC), and immature PC with most notable
levels of expression in MLI (TNRC6B, SMARCC2, FAM98C,CHD2, UBN2, DSCAM, TBL1XR1,
CUL3), PIP (USP45, ERBIN, PYHIN1), immature PC and PCs (ASXL3, SHANK2, CACNA2D3,
SLC6A1, PTEN, P2RX5, UIMC1), and immature CN and CNs (PAX5 and ACHE).
We also investigated spinocerebellar ataxia (SCA), which is defined by Purkinje cell loss and
associated with movement control and muscle coordination issues147. At this stage in development,
we found notable levels expression of SCA risk genes (DAB1, KIF26B, and ITPR1) in the PC
cluster including a significant enrichment of the SCA gene set in immature PC (TBP) and the MLI
(NOP56, FGF14, ATXN8OS, and PRKCG).
We continued our examination with Joubert Syndrome, an autosomal recessive ciliopathy. The
genes associated with this syndrome were significantly enriched in the choroid plexus and roof
plate cell types with most notable levels of expression in TMEM231, B9D1, and NPHP1 in choroid
plexus and CC2D2A in the roof plate cells.
Finally, we investigated the enrichment of genes associated with Dandy-Walker malformations
and hypoplasia under the umbrella of cerebellar malformations148. We see this gene set enriched
significantly in PAX2+ interneuron precursors (PIP) with high levels of expression in PTF1A and
EBF2. As expected, we found no significant enrichment in any of the genes associated with
Alzheimer’s disease, a progressive disease leading to adult-onset cerebellar atrophy149. Together,
these data demonstrate the ability of the hCerOs to uncover the cell type specific mechanisms
underlying these and other disorders stemming from dysfunctions of early human cerebellar
development.

106
CHAPTER 4
Uncovering the function of autism spectrum disorder-associated gene
SYNGAP1 in cerebellar development using human cerebellar organoids
(hCerOs)
Alexander Atamian1,2, Negar Hosseini1,2, Tuan Nguyen1,2, Marcelo P. Coba4,5,6, Giorgia
Quadrato1,2*
Affiliations:
1
Department of Stem Cell Biology and Regenerative Medicine, Keck School of Medicine,
University of Southern California, Los Angeles, CA 90033, USA
2
Eli and Edythe Broad CIRM Center for Regenerative Medicine and Stem Cell Research at
USC, Keck School of Medicine, University of Southern California, Los Angeles, CA 90033,
USA
4
Department of Psychiatry and Behavioral Sciences, Keck School of Medicine, University of
Southern California, Los Angeles, CA 90033, USA
5
Zilkha Neurogenetic Institute, Keck School of Medicine, University of Southern California;
Los Angeles, CA 90033, USA
6
Department of Physiology and Neuroscience, Keck School of Medicine, University of
Southern California, 1501 San Pablo Street, Los Angeles, CA 90033, USA
Author contributions:
A.A. and G.Q. conceived the experiments. A.A. and N.H. generated, cultured, and
characterized all organoids used in this study; A.A., T.N. performed scRNA-seq
experiments; T.N. performed scRNA-seq analysis and worked on cell type assignments
and data analysis with assistance from A.A.; M.P.C. generated the D2 iPSC line; G.Q.
supervised all aspects of the project; A.A. and G.Q. wrote the manuscript with
contributions from all authors.
Unpublished work

107
1 ABSTRACT
Genes involved in synaptic function are enriched among those with autism spectrum disorder
(ASD)-associated rare genetic variants. Dysregulation of neurogenesis has been implicated as a
convergent mechanism in ASD pathology, yet it remains unknown how ‘synpatic’ ASD risk genes
contribute to these phenotypes, which arise before synaptogenesis. Here, we show that synaptic
Ras GTPase-activating (RASGAP) protein1 (SYNGAP1, a top ASD risk gene) is expressed within
cerebellar progenitor cell populations in addition to its well-known synaptic functions in neurons.
In a human cerebellar organoid model of SYNGAP1 haploinsufficiency, we find dysregulated
neuron projection development, cytoskeletal remodeling, and synaptic organization. Thus
SYNGAP1-related brain disorders may arise through non-synaptic mechanisms, highlighting the
need to study genes associated with neuronal disorders in diverse human cell types and
developmental stages.
2 INTRODUCTION
The greatly protracted development of the cerebellum compared to the neocortex makes it
particularly susceptible to various neurodevelopmental disorders (NDD) including autism
spectrum disorder (ASD)150. Specifically, this brain region is one of the first to initiate
differentiation and one of the last to fully mature151. Additionally, NDDs are often characterized
by motor dysfunctions such as ataxia, which implicates cerebellar involvement in these disorders,
a brain region critical in governing motor behavior112. SYNGAP1 is a top ASD genetic risk
factor152 with various studies reporting that 21%-51% of patients with SYNGAP1 mutations with
ataxia or gait abnormalities, suggesting cerebellar dysfunctions occurring within this disease153–
156.
The SynGAP protein has various structural isoforms due to differential transcriptional start sites
and post-transcriptional processing157. Three N-terminal isoforms (A-C) have been identified and
result from alternative transcriptional start site usage, while at least four C-terminal SynGAP splice
variants (α1, α2, β, and γ) are currently known157. SynGAP isoforms display distinct distribution
patterns in neuronal subcellular compartments and brain regions as well as differentially regulate
synaptic strength in cultured neurons157,158. More specifically, the expression pattern of SynGAP
α and β variants were similar between the forebrain and the cerebellum, however, the level of

108
expression of these variants were lower in the cerebellum than that of the forebrain3
. Additionally,
a previous study has pointed out that SynGAP is expressed in the cerebellum prior to P4 when
there are not too many synaptic connections and is detected as early as E16 in mouse
development159. Although there are cortical issues that arise from mutations in SynGAP, recent
evidence has shown that complex developmental brain disorders such as ASD are also linked with
dysfunctional cerebellar development112. More importantly, this is true for brain disorders that are
characterized by motor or non-motor deficits. Despite these implications, most of our
understanding about SynGAP mutations has derived from animal models of cortical
development160–165 coupled with our recent study on early human cortical development conducted
in cortical organoids166.
SYNGAP1 is one of the most abundant proteins found at the postsynaptic density (PSD) of cortical
excitatory synapses152,167. Within the PSD, SYNGAP1 functions as a RAS GTPase-activating
(RASGAP) protein that regulates synaptic plasticity167–172. Through its RASGAP domain,
SYNGAP1 limits the activity of the mitogen-activated protein kinase 1 (Mapk1/Erk2), whereas
through its PDZ-binding domain, SYNGAP1 helps assemble the core scaffold machinery of the
PSD164,173–175. Disruptions in SYNGAP1 protein function has been associated with synaptic
dysfunction. However, recent evidence points toward a role in governing early neurogenesis: First,
homozygous deletion of Syngap1 in embryonic mice leads to early developmental lethality160.
Second, decreased syngap1 levels have been shown to affect the ratios of neural progenitor cells
to mature neurons in Xenopus tropicalis161. Third, disruption of the Syngap1 signaling complex in
embryonic mice results in deficits in the tangential migration of GABAergic interneurons162.
Fourth, in addition to being an ASD genetic risk factor, de novo mutations in SYNGAP1 have
been found in patients with intellectual disability, epilepsy, neurodevelopmental disability, and
global developmental delay163,165,176. Lastly, a cortical organoid model of SYNGAP1
haploinsufficiency displayed dysregulated cytoskeletal dynamics that disrupted lamination and
resulted in accelerated maturation of cortical projection neurons166. This evidence, combined with
the high frequency and penetrance of pathogenic SYNGAP1 variants, indicates a major and unique
role for SYNGAP1 in human cortical development. However, mutations in SYNGAP1 have not
been as closely examined in cerebellar development and how it affects early cerebellar
progenitors/neurogenesis.

109
To address this question, we used hCerOs derived from human induced pluripotent stem cells
(hPSCs), which allows for the longitudinal characterization of ASD-risk genes in an all-human
system. Here, we show the expression of SYNGAP1 protein in human cerebellar progenitors in
addition to neuronal cell types. We find that SYNGAP1 regulates cytoskeletal remodeling of
subcellular and intercellular components of cerebellar neuronal types as was seen in human radial
glial cells in cortical organoids, which led to the disorganization of the developing cortical plate
in cortical organoids166. Through scRNAseq coupled with structural analysis of mutant organoids,
we discovered that SYNGAP1 regulates the differentiation trajectory within SYNGAP1
haploinsufficient organoids exhibiting decelerated maturation of granule cells and Purkinje cell,
the major excitatory and inhibitory neurons of the cerebellum, respectively. Taking this data
together, we hypothesized that SYNGAP1 has a role in cytoskeletal organization as was observed
in human cortical organoids166, which led to the disruption of the cytoskeletal organization in early
progenitors and altered neurodevelopmental trajectory.
3 RESULTS
3.1 Generation and characterization of haploinsufficient SYNGAP1
(Patientp.Q503X) and isogenic control (PatientCorrected) cerebellum
SYNGAP1 has been shown to be involved not only in early neurodevelopment, but also is largely
restricted to the PSD of mature synapses167,168. To determine the function of SYNGAP1 in
cerebellar development, we developed hCerOs from a patient carrying a SYNGAP1-truncating
(p.Q503X) mutation. This patient presented with intellectual disability, developmental delay,
autistic features, and epilepsy. For this, we generated iPSC lines from the patient and the
correspondent corrected isogenic control (Patientp.Q503X and PatientCorrected, respectively).
Characterization of the patient-derived iPSC cell line showed that the p.Q503X mutation produces
a haploinsufficient model of SYNGAP1 dysfunction with a 51.6% decrease in SYNGAP1 total
protein levels and undetectable protein fragments as evidenced by western blot and quantitative
mass spectrometry assays (Extended Data Figure 3 in Birtele et al166).
To start addressing the role of SYNGAP1 in early cerebellar development, we derived hCerOs
from Patientp.Q503X and PatientCorrected lines as described previously in Chapter 2 (Figure 1A).

110
Around Day 10, there is a noticeable difference in the circularity of these organoids (Figure 1B),
which indicates that SYNGAP1 may have a role in early cerebellar neurodevelopment as was
previously shown in cortical organoids166. Additionally, both cell lines (Patientp.Q503X and
PatientCorrected) produces the major inhibitory and excitatory neuronal populations of the
developing cerebellum as indicated by the presence of SKOR2 (post-mitotoic Purkinje cell
progenitors) and BARHL1 (granule cells), respectively (Figure 1C,D). We see the SKOR2+ cell
are directly adjacent to the SOX2+ neural rosette structures of PatientCorrected derived hCerOs as
described in previous chapters. However, there are two key differences between Patientp.Q503X and
PatientCorrected derived hCerOs. The first striking difference is the lack of rosette structures within
the Patientp.Q503X organoids as seen in Figure 1C as well as the presence of bifurcating parallel
fibers emerging from granule cells only in the PatientCorrected hCerOs, a clear sign of maturation of
this cell type (Figure 1E and 1F).
Together, this data indicates that these cell lines have the ability to generate both inhibitory and
excitatory progenitor populations of the developing cerebellum with subtle differences in
morphologies that must examined at a higher resolution to determine a disease phenotype.
3.2 SYNGAP1 haploinsufficiency increases maturation of granule
cells and Purkinje cells.
To dive further into uncovering the function of SYNGAP1 in cerebellar cell types, we analyzed
these organoids at a single cell resolution by performing scRNA-seq analysis on a total of 31,970
cells from six individual organoids at 2 months (three corrected and three haploinsufficient). To
systematically perform cell type classification, we clustered cells from all organoids and compared
signatures of differentially expressed genes (DEGs) with a pre-existing human cerebellar singlecell dataset37. This defined 16 main transcriptionally distinct cell types (Figure 2A), which
included a large diversity of progenitors (Ventricular Zone (VZ), glia, Bergmann glia (BG),
granule cell progenitors (GCP), and Rhombic Lip (RL)) and neuronal cell types (granule cells
(GC), inhibitory cerebellar nuclei (iCN), PAX2+ interneuron progenitors (PIP), and Purkinje cells
(PC)), representing all the main cell types of the endogenous developing human cerebellum.
Additionally, every cluster expressed a variable level of SYNGAP1 indicating a similar expression
pattern as seen in cortical organoids (Figure 2B). Importantly, we found that individual organoids
were highly reproducible in cell type composition, which highlights the robustness of our

111
cerebellar organoid protocol. (Figure 2C). This high level of reproducibility allowed us to preform
differential gene expression analysis between PatientCorrected and Patientp.Q503X cells. In contrast to
observed increase in the differentiative division mode observed in haploinsufficient cortical
organoids166, we found a decreased enrichment in terms related to neuronal differentiation,
synapses, and neuronal projection development the major excitatory and inhibitory neuronal
populations of the developing cerebellum (GCPs, GCs, and PCs) (Figure 2D). These GO terms
analysis include genes involved in neuron projection development, cytoskeletal remodeling, and
synaptic organization. Interestingly, many genes that are downregulated in Patientp.Q503X have also
been implicated in various disorders resulting in intellectual disability, developmental delay, and
autism spectrum disorder (Table 1). Moreover, we observe this morphological difference in 4-
month-old Patientp.Q503X PCP2+ Purkinje cells, which display a decreased dendritic arborization
compared to its isogenic control (PatientCorrected) (Figure 2E). Overall, these data suggest that
SYNGAP1 haploinsufficiency affects the differentiative trajectory of cerebellar neurons.
4 DISCUSSION
Many human genetic studies of neurodevelopmental diseases including ASD have found
enrichment of mutations in genes encoding classically defined synaptic proteins177–183. Currently,
most studies of SYNGAP1 and other synaptic proteins have been accomplished within the context
of mature synapses in rodent models165. Indeed, the lack of a reliable model to study the stagespecific functions of ASD risk genes during human brain development has limited our
understanding of their role to rudimentary functional categories. Here, we leveraged hCerOs to
model probe the function of SYNGAP1 in human cerebellar development.
Our recent work has shown that SYNGAP1 might regulate cytoskeletal organization in hRGCs in
cortical organoids through its RASGAP domain and its localization and assembly in
macromolecular complexes through its PDZ ligand domain. This is particularly intriguing when
considering the critical role of tight junctions and their association with the cytoskeleton in forming
and preserving the apical junctional belt, which maintains radial glial apico–basal polarity and
neuroepithelial cohesion184. It is therefore important to determine the role of SYNGAP1 at earlier
time points in cerebellar development and probe the function of SYNGAP1 in earlier timepoints
when the hCerO is composed of a more homogenous population of early cerebellar progenitors.
Additionally, this dysregulation of cytoskeletal organization is seen within our 2-month-old

112
hCerOs, which is led to a downregulation of a host of genes involved in locomotion, neuronal
projection, and synaptic organization in the Patientp.Q503X line. More specifically, the
downregulation of these processes linked to maturation in Patientp.Q503X Purkinje cells coincides
with findings that have shown that a decrease in SYNGAP1 in GABAergic cells leads to impaired
synaptic connectivity and synaptic inhibition185.
A growing body of literature suggests that a dysregulated neurogenesis program may lead to
impaired neuronal wiring. This is because establishing proper neuronal circuits requires
spatiotemporally precise control of neuronal positioning, neurogenesis and afferent and efferent
synaptic connectivity186,187. Therefore, our finding that cerebellar neurons in SYNGAP1-
haploinsufficient organoids display delayed development reshapes the current framework for
therapeutic interventions. Additionally, the accelerated maturation observed in cortical organoids
contrasts the decreased maturation observed in hCerOs indicating a brain regions specific
alteration due to this specific mutation in SYNGAP1. This does, however, contradict a finding by
the Hatten lab, which shows a similar accelerated development in mice granule cells159. This
difference may be because the mice granule cells used for the study were from a P6 mouse while
our 2-month-old hCerOs resemble more of an embryonic stage in development (~E14-16).
Nevertheless, these findings in our hCerOs must be confirmed with future work that includes
scRNAseq at later time points in addition to functional analysis of hCerOs through patch-clamp
recordings and analyzing GCaMP-calcium dynamics as was conducted in cortical organoids166.
The expression of SYNGAP1 in progenitors and neuronal synapses in hCerOs underscores the
importance of dissecting the role of ASD risk genes in specific cell types across developmental
stages and across brain regions, which suggests that a similarly approach may have broader
relevance for studying other NDDs. The cerebellum is connected to the main areas of the brain
that are altered in ASD, which display altered structure, function, and connectivity. However, the
broad impact ASD has on various brain regions in addition to the heterogeneity of this disorder
makes parsing out the direct impact an altered cerebellum has on the rest of the brain. hCerOs offer
an opportunity to take a reductionist approach to study the intricacies of the cerebellum and its
contributions to ASD, which has the potential to yield novel therapeutics when combined with
current model systems used to study ASD.

113
5 FIGURES/TABLES
Fig. 1. SYNGAP1 haploinsufficient hCerOs can generate inhibitory and excitatory
progenitors of the developing cerebellum. (A) Protocol schematic for generating hCerOs;
ROCKi: ROCK inhibition, SB: TGFB inhibition, Noggin: BMP inhibitor, CHIR (CHIR-99021 -
GSK3 inhibitor), FGF8b: Fibroblast Growth Factors 8b, T3: triiodothyronine. (B) Quantification
of circularity of Day 10 hCerOs. n=12 across 3 batches. p<0.01=**** (C/D) Day 30 and Day 60

114
Immunofluorescence stain of granule cells (BARHL1), post-mitotic Purkinje cell progenitors
(SKOR2), and neuroepithelial cells (SOX2) in Patientp.Q503X and PatientCorrected hCerOs,
respectively. (E) 40X magnified BARHL1 staining s in Patientp.Q503X and PatientCorrected hCerOs,
respectively at Day 60. (F) Quantification of BARHL1+ cells that have bifurcating extensions.

115
Fig. 2. SYNGAP1 haploinsufficient organoids exhibit accelerated maturation of cerebellar
neurons. (A) Combined UMAP, uniform manifold approximation and projection, plot of all 6
organoids from Patientp.Q503X and PatientCorrected hCerOs with defined clusters for cerebellar cell
types including BG: Bergmann glia, Div-: dividing, GC: granule cells, GCP: granule cell
progenitors, iCN: inhibitory cerebellar nuclei, PC: Purkinje cells, PIP: pax2+ interneuron
progenitors, RL: rhombic lip, VZ: ventricular zone. (B) FeaturePlot of SYNGAP1 expression
pattern in 2-month-old hCerOs. (C) Individual UMAP plots for three PatientCorrected and three
Patientp.Q503X hCerOs at 2-months. organoid 1: 4900 cells; organoid 2: 6949 cells; organoid 3: 3455

116
cells; organoid 4: 8144 cells; organoid 5: 4426 cells; organoid 6: 4096 cells. (D) GO terms from
scRNAseq of 2-month-old Patientp.Q503X and PatientCorrected hCerOs. Graphical representation of
upregulated terms for Patientp.Q503X in Granule Cell Progenitors, Granule Cells, and Purkinje Cells.
(E) 40um section of 4-month-old hCerOs stained with anti-GFP after viral infection with
AAV8.L7-6.eGFP.WPRE.hBG.
Table 1.
DEGs associated with ASD and intellectual disability found in GO terms list.
Table 2.
Primary antibodies used for immunofluorescence
Primary
Antibody
Host Manufacturer Catalog Number Dilution
BARHL1 Rabbit Proteintech HPA004809 1:500
SKOR2 Rabbit Proteintech HPA046206 1:500
SOX2 Goat RD Systems AF2018 1:500
Anti-GFP Chicken Abcam G160 1:100

117
6 Materials and Methods
6.1 HUMAN IPSC LINE GENERATION
6.1.1 Patientp.Q503X cell line.
An iPSC line from a patient carrying a SYN- GAP1 c.1507C>T (p.Q503X) nonsense mutation was
generated. iPSCs were generated using episomal expression of Yamanaka factors in patientderived PBMCs188.
6.1.2 PatientCorrected cell line
The sequence for a single-guide RNA (sgRNA) targeting the patient-specific mutation in
SYNGAP1 was cloned into pSpCas9(BB)-2A-Puro (PX459) version 2.0 (Addgene plasmid 62988).
This, along with a homology-directed repair (HDR) template containing the WT SYNGAP1
sequence, were nucleofected into the patient-derived iPSC line. Individual iPSC colonies were
transferred to 24-well plates and subsequently underwent restriction enzyme-based genotyping.
Positive colonies were then confirmed by Sanger sequencing and expanded in culture. The HDR
template was:
CCGCGAGAACACGCTTGCCACTAAAGCCATAGAAGAGTATATGAGACTGATTGGTC
AGAAATATCTCAAGGATGCCATTGGTATGGCCCACACTCAGGCCCTCTTCTTCCCAA
ACCTGCCA.
The underlined CAG sequence corresponds to the insertion of the WT ‘T’ base pair, and the
underlined T base corresponds to a silent mutation to disrupt the PAM sequence of the sgRNA.
Substitution of the truncating ‘T’ with the WT ‘C’ base pair was screened for by restric- tion
enzyme digestion and then confirmed by Sanger sequencing. The mutation was c.1507C>T
(p.Q503X), a nonsense mutation. For genotyping information, introduction of the correction
destroys the DrdI restriction enzyme site, which was used for initial screening of cell lines. Work
on the patient-derived line Patientp.Q503X is approved by the USC IRB (HS-18-00745).

118
6.2 VIRAL INFECTION OF ORGANOIDS
Organoids were transduced with AAV8.L7-6.eGFP.WPRE.hBG (Addgene, #126462) using 1.5 µl
of virus in 150 µl of CerDM3 medium overnight on static condition in a single well of a 24 well
low attachment plates. One day after transduction, a full media change was performed using
CerDM3 media. On day three after transduction organoids were transferred to a 6 well low
attachment plate and a full media change to BrainPhys Medium (STEMCELL Technologies,
#05792) supplemented with NeuroCult SM1 Neuronal Supplement (STEMCELL Technologies,
#05711 was performed. Organoids were kept for at least one week in culture before sectioning at
40um and staining with anti-GFP to visualize PCP2+ cells.
6.3 IMMUNOCYTOCHEMSITRY
Organoids were fixed in 4% paraformaldehyde for 30 min at room temperature before overnight
incubation at 4C in 30% sucrose solution. Organoids were then embedded in Tissue-Tek O.C.T.
compound (Sakura, #62550) and cryosectioned at 20μm thickness onto glass slides (Globe
Scientific, #1354W). Slides were washed 3x with a 0.1% Tween20 (Sigma #P9416) solution
before a 1-hour incubation in 0.3%TritonX-100 (Sigma, #T9284) and 6% bovine serum albumin
(Sigma, #AA0281) solution. Slides were incubated for 1-hour at room temperature in a primary
antibody solution followed by a 1-hour room temperature incubation in a secondary antibody
solution, both consisting of 0.1% TritonX-100 and 2.5%BSA with 3 washes before and after
secondary antibody incubation. Slides were coverslipped using Fluoromount G (EMS, #50-259-
73). Primary antibodies and dilutions used are specified in Table 2.
6.4 DISSOCIATION OF hCerOs AND SINGLE-CELL RNA-SEQ
Individual brain organoids were dissociated into single cells using the Worthington Papain
Dissociation System kit (Worthington Biochemical). To give an estimated recovery of 6,000 cells
per channel dissociated cells were resuspended in ice-cold PBS containing 0.04% BSA at a
concentration of 1,000 cells/μl, loaded onto a Chromium Single Cell 3′ Chip (10x Genomics) and
processed through the Chromium controller to generate single-cell gel beads in emulsion (GEMs).
scRNA-seq libraries were prepared with the Chromium Single Cell 3′ Library & Gel Bead Kit v.2
(10x Genomics). Libraries from different samples were pooled based on molar concentrations and
sequenced on a NextSeq 500 instrument (Illumina) with 26 bases for read 1, 57 bases for read 2
and 8 bases for index 1. After the first round of sequencing, libraries were re-pooled on the basis

119
of the actual number of cells in each and re-sequenced to give an equal number of reads per cell in
each sample and to reach a sequencing saturation of at least 50% (in most cases >70%).
6.5 SINGLE-CELL RNA-SEQ DATA ANALYSIS
Bioinformatic analysis was performed using the Seurat R package. Briefly, all single cell
transcriptomes from 3 organoids were combined, and a set of highly variable genes was
determined. The initial PCA analysis was performed using this gene set and the number of
statistically significant principal components (PCA) was identified via Jackstraw analysis. Cluster
structures were remapped to two dimensions using Uniform Manifold Approximation and
Projection (UMAP). Finally, a graph-based clustering approach was used to cluster the cells to
produce putative cell-type clusters and return sets of differentially expressed marker genes for each
cluster. These were compared to known marker genes to ascribe a cellular identity to each cluster.
This data was then integrated with the human dataset to confirm cell type identifications.
6.6 ORGANOID SIZE ANALYSIS
ImageJ software was used to measure the circularity and the area of each single organoid. Prism
software was then used to plot the average size of the organoids of each differentiation.

120
CHAPTER 5
Future Directions
1 INTRODUCTION
The complex interplay between cell-to-cell interactions coupled with the array of morphogen
gradients within the developing neural tube leads to the spatially segregated regional identities of
the developing brain. The work presented in Chapter 2 and 3 provides foundational work to
uncover cerebellar developmental dynamics through lineage tracing, improve the physiologically
relevance of hCerOs through the introduction of exogenous cell types, and robustly model
cerebellar disorders to discover potential therapeutics.
2 LINEAGE TRACING CEREBELLAR CELL TYPES
As previously stated in Chapter 1, the gold-standard method of characterizing brain organoids is
scRNAseq, which not only uncovers the cellular diversity, but also can be used to infer the
developmental dynamics of the cells within the organoid. This is because organoids contain a range
of cell types from early progenitors to fully differentiated cells. Pseudotime, the most widely used
technique used to uncover lineage trajectory using RNA transcripts, comes with its caveats as this
method is not accurate in tracing the lineage of distinct cell types189,190. Therefore, more precise
methods of tracing are needed to trace the path progenitors take on their way to becoming more
mature cell types. To achieve this, a PhD student in the lab, Jean Paul Urenda, who is working
closely with the Elowiz laboratory at CalTech to establish several systems that allow for a clearer
understanding of the fate of each individual cells and their decedents within brain organoid
systems. Two of these techniques that are currently being used to study hCerOs include the
Hypercascade system, which incorporates layered barcoding that allows for deeper insight into
clonal dynamics and the Zombie system, which allows for tracking the spatial data of specific
clones.
The Hypercascade system uses a barcoding system to scar and trace the lineage dynamics, which
is analogous to other lineage tracing technologies. However, the Hypercascade system
incorporates barcode targets with a 74-unit arrays divided into four layers, which are edited in a
sequential way. This method of tracing contains a high level of memory given the various levels

121
of editable barcode sites. The level of information obtained from this system can reveal progenitor
dynamics as they make their bifurcation decision leading to an excitatory or inhibitory fate.
Additionally, this system can answer more specific questions relating to fate choice by revealing
the transcriptomic profiles of subpopulation of progenitors from within the current, more broadly
categorized “inhibitory” and “excitatory” progenitor populations, which give rise to distinct
neuronal cell types.
In addition to lineage tracing, the Zombie system maintains the spatial resolution of the
characterized cells. This is done through implementing RNA polymerases to transcribe gnomically
integrated barcodes in fixed cells to produce amplified RNA that can be detected using sequential
FISH (fluorescence in situ hybridization) probes. This technique can uncover dynamics behavior
of inhibitory and excitatory progenitor populations in hCerOs as they decide their fate from a
common progenitor population. Our previous work has shown that these two populations
(KIRREL2+ progenitors and ATOH1+ progenitors) are spatially segregated (Chapter 3, fig. S1S
and T). Additionally, previous work has demonstrated that this bifurcation decision between
inhibitory and excitatory lineages is governed by NOTCH signaling191. Therefore, the Zombie
system can reveal the spatial location of the common progenitor pool and elucidate how these two
populations emerge from a common progenitor pool.
The third technique developed by the Elowitz lab that is currently in the early stages of the project
led by Jean Paul Urenda is the Synthetic RNA Export system. This system allows for the secretion
of vesicle-packaged RNA from living cells into the media they are cultured in. These nanoparticles
are then collected and analyzed to reveal the RNA content of individual cells that they are packaged
from. Additionally, these secreted packages contain stochastically edited barcodes from each
individual cells that can be analyzed to reveal the lineage relationships and cellular dynamics
within the organoids. The main advantage of this system over the previous two mentioned is the
ability to conduct longitudinal analysis on the same batch of organoids given that the downstream
experiments are done on the culture media and not on the organoids themselves. By doing so, one
can study more dynamic conditions such as how a specific diseased state can affect early
neurogenesis in addition to studying those neurons as they mature over time. Moreover, this system
can allow for therapeutic applications by studying the effects of drug treatments on human cells
over a prolonged time frame.

122
Each of these technologies have the ability to uncover the lineage trajectory of each cell type that
develops in these organoids with their own unique advantages. The Hypercascade system is a more
traditional lineage tracing system with a very high level of memory, which can help clearly define
branch points as a lineage tree is constructed. The Zombie system allows for traditional clonal
tracing as well. However, it also provides valuable information on the spatiotemporal dynamics
within hCerOs as they develop. Finally, the RNA export system will be very useful when
conducting more longitudinal analysis without the need for dissociating or fixing the organoids,
which will offer the opportunity to study cerebellar progenitor dynamics over time. These
technologies can be used to better characterize the function of SYNGAP1 in hCerOs and get a
better understanding of progenitor/neuronal dynamics of the developing cerebellum.
3 UPGRADING PHYSIOLOGICAL RELEVANCE OF hCerOs
WITH ADDITIONAL CELL TYPES
While hCerOs generate neuronal and glial populations of the developing cerebellum and can be
used to model neuronal functionality, the incorporation of other cell types will greatly enhance the
physiological relevance of the model and help model the microenvironment of the developing
cerebellum more accurately. The integration of various cell types such as microglia,
oligodendrocytes, and endothelial cells into cerebellar organoids will be a promising step toward
creating a platform to model complex cellular interactions to understand development and disease
pathogenesis that involves the contributions of multiple cell types.
Microglia, the resident immune cells of the CNS, plays a crucial role in synaptic pruning and brain
development192–194. With its distinct functional dynamics and transcriptomic profile compared to
microglia from other brain regions195, microglial integration into hCerOs provides a platform to
study this cell type in a unique context, which has been difficult to study to date given the lack of
a robust protocol that can recapitulate cerebellar development in vitro. Additionally, given
microglia’s role in various disorders and its contributions to disease progression such as ataxia and
ASD 196,197, it is essential to co-culture these cells to properly model these disorders and find
therapeutic targets198. Additionally, a recent study has shown that microglia can aid in maturation
of cortical organoids post-integration199. Therefore, the integration of microglia into hCerOs has

123
the potential to further mature the organoids in addition to creating a more physiologically relevant
model system.
Oligodendrocytes are another important cell type involved in myelinating the axons in the CNS,
which facilitates the efficient transmission of action potentials in neurons. Dysregulation of
myelination is implicated in various neurological disorders such as multiple sclerosis with patients
commonly displaying cerebellar impairments leading to tremor, ataxia, and dysarthria200–203. With
the advent of protocols that aim to generate hPSC derived oligodendrocytes101,102, integration of
these cells into hCerOs is a great step forward in improving the model system to study
oligodendrocyte-axon interactions in circuit development in the context of these diseases.
The co-culture of endothelial cells in hCerOs is also crucial in understanding the blood-brain
barrier, which delivers nutrients to the brain and mediates the efflux of metabolic waste.
Additionally, this cell type has been implicated in various neurodegenerative disorders such as
Alzheimer’s and amyotrophic lateral sclerosis (ALS)204. By including this cell type into organoids,
researchers can get a better understanding of how the neurovascular unit can influence the
development and function of the cerebellum. Although attempts have been made to integrate
endothelial cells into brain organoids205–207, it is important to note that a unidirectional flow system
(an in-and-out system) is necessary to properly model the neurovasculature. The use of
microfluidics devices to guide and arrange these endothelial cells into a network or creating an
artificial vasculature would be necessary to have a physiologically relevant model.
By introducing various cell types into hCerOs at physiologically relevant time points, we can start
to model how these exogenously introduced cell types affect cerebellar ontogenesis and circuit
formation as well as uncover their contributing roles in the emergence and progression of
cerebellar disorders.
4 CEREBELLAR DISEASE MODELLING
hCerOs offer a revolutionary platform for modeling cerebellar disease in a way that was previously
inaccessible. These hPSC derived model system recapitulates key aspects of early human
cerebellar development and provides researchers with a system to study disease mechanism that
underly human cerebellar disorders such as ataxia, medulloblastoma, and neurodevelopmental
disorders. Previous 2D culture systems lacked the ability to recapitulate the cellular diversity and

124
complex network activity while animal models simply could not replicate certain human specific
features. However, by recapitulating cytoarchitectural features as well as distinct
electrophysiological properties, hCerOs offer a unique avenue to investigate human disease
pathogenesis, screen potential therapeutics, and develop personalized treatments.
Ataxia is a form of cerebellar disorder that is characterized in patients by a loss of balance and
coordination and accompanied by slurred speech. hCerOs can be an ideal model to probe
monogenic disorders such as autosomal dominant spinocerebellar ataxias (SCAs) or the autosomal
recessive Friedreich’s ataxia208. This model can allow for the study of the mutation and can be
compared to its CRISPR corrected isogenic control hCerO. This way the background of the patient
will not be a confounding variable similar to the comparison between Patientp.Q503X and
PatientCorrected hCerOs in Chapter 4.
Recent studies have revealed that aggressive medulloblastoma subgroups originate from the RLsvz
of the developing cerebellum, which is a human specific subgroup of cells that is not present in
mice81,82. This is the most common malignant childhood brain tumor yet the ability to model this
disease was not possible until now. Given that our study has revealed the presence of RLsvz within
our organoids, the ability to model this disease in vitro is now possible. Additionally, the ability to
create patient specific iPSC lines offers the potential to study the pathogenesis of various
subgroups of medulloblastoma and uncover the mechanisms of each subgroup within an all-human
system. Recently, we have initiated a collaboration with Dr. Huang Miller at USC to study the
origins of medulloblastoma to gain a deeper understanding of the disease.
Using hCerOs, researchers can gain insight into the underlying molecular mechanisms that drive
pathogenesis and disease progression. Moreover, hCerOs can be employed in a high-throughput
drug screening assay that can help identify specific compounds that target pathways that give rise
to disease-associated phenotypes found in ataxia, medulloblastoma, and neurodevelopmental
disorders. This can pave the way toward developing targeted therapies for cerebellar disorders.

125
REFERENCES
1. Barton, R. A. & Venditti, C. Rapid Evolution of the Cerebellum in Humans and Other Great
Apes. Current Biology 27, 1249–1250 (2017).
2. Ito, M. The Cerebellum and Neural Control. Raven Press, New York (1984).
3. Adrian, E. D. Afferent areas in the cerebellum connected with the limbs. Brain 66, 289–315
(1943).
4. Snider, R. S. & Stowell, A. RECEIVING AREAS OF THE TACTILE, AUDITORY, AND
VISUAL SYSTEMS IN THE CEREBELLUM. Journal of Neurophysiology 7, 331–357
(1944).
5. F.Nitschke, M., Kleinschmidt, A., Wessel, K. & Frahm, J. Somatotopic motor representation
in the human anterior cerebellum: A high-resolution functional MRI study. Brain 119,
1023–1029 (1996).
6. Rijntjes, M., Buechel, C., Kiebel, S. & Weiller, C. Multiple somatotopic representations in
the human cerebellum: NeuroReport 10, 3653–3658 (1999).
7. Grodd, W., Hülsmann, E., Lotze, M., Wildgruber, D. & Erb, M. Sensorimotor mapping of
the human cerebellum: fMRI evidence of somatotopic organization. Human Brain Mapping
13, 55–73 (2001).
8. Haines, D. E. Mario Manto, Donna L. Gruol, Jeremy D. Schmahmann, Noriyuki Koibuchi,
and Ferdinando Rossi (Eds.) Handbook of the Cerebellum and Cerebellar Disorders:
Springer New York Heidelberg Dordrecht London 2013, Vol 1–4, 2,424 p. ISBN 978-94-
007-1404-5 (Print and Online); ISBN 978-94-007-1333-8 (eBook). Cerebellum 13, 303–
306 (2014).
9. Schmahmann, J. D. The cerebellum and cognition. Neuroscience Letters 688, 62–75 (2019).
10. Petersen, S. E., Fox, P. T., Posner, M. I., Mintun, M. & Raichle, M. E. Positron Emission
Tomographic Studies of the Processing of Singe Words. Journal of Cognitive Neuroscience
1, 153–170 (1989).
11. Raichle, M. E. et al. Practice-related Changes in Human Brain Functional Anatomy during
Nonmotor Learning. Cereb Cortex 4, 8–26 (1994).
12. Desmond, J. E., Gabrieli, J. D. E. & Glover, G. H. Dissociation of Frontal and Cerebellar
Activity in a Cognitive Task: Evidence for a Distinction between Selection and Search.
NeuroImage 7, 368–376 (1998).

126
13. Kim, S.-G., Uğurbil, K. & Strick, P. L. Activation of a Cerebellar Output Nucleus During
Cognitive Processing. Science 265, 949–951 (1994).
14. Schmahmann, J. D. & Pandya, D. N. Anatomical investigation of projections to the basis
pontis from posterior parietal association cortices in rhesus monkey. J of Comparative
Neurology 289, 53–73 (1989).
15. Schmahmann, J. D. & Pandya, D. N. Projections to the basis pontis from the superior
temporal sulcus and superior temporal region in the rhesus monkey. J of Comparative
Neurology 308, 224–248 (1991).
16. Schmahmann, J. D. & Pandya, D. N. Anatomic Organization of the Basilar Pontine
Projections from Prefrontal Cortices in Rhesus Monkey. J. Neurosci. 17, 438–458 (1997).
17. Middleton, F. A. & Strick, P. L. Cerebellar Projections to the Prefrontal Cortex of the
Primate. J. Neurosci. 21, 700–712 (2001).
18. Kelly, R. M. & Strick, P. L. Cerebellar Loops with Motor Cortex and Prefrontal Cortex of a
Nonhuman Primate. J. Neurosci. 23, 8432–8444 (2003).
19. Arkell, R. M. & Tam, P. P. L. Initiating head development in mouse embryos: integrating
signalling and transcriptional activity. Open Biol 2, 120030 (2012).
20. Davies, J., Evans, R. H., Jones, A. W., Smith, D. A. S. & Watkins, J. C. Differential
activation and blockade of excitatory amino acid receptors in the mammalian and
amphibian central nervous systems. Comparative Biochemistry and Physiology Part C:
Comparative Pharmacology 72, 211–224 (1982).
21. Stern, C. D. et al. Head-tail patterning of the vertebrate embryo: one, two or many
unresolved problems? Int. J. Dev. Biol. 50, 3–15 (2003).
22. Chambers, S. M. et al. Highly efficient neural conversion of human ES and iPS cells by dual
inhibition of SMAD signaling. Nat Biotechnol 27, 275–280 (2009).
23. Rosebrock, D. et al. Enhanced cortical neural stem cell identity through short SMAD and
WNT inhibition in human cerebral organoids facilitates emergence of outer radial glial
cells. Nat Cell Biol 24, 981–995 (2022).
24. Lupo, G. et al. From pluripotency to forebrain patterning: an in vitro journey astride
embryonic stem cells. Cell Mol Life Sci 71, 2917–2930 (2014).
25. Hoshino, M. et al. Ptf1a, a bHLH Transcriptional Gene, Defines GABAergic Neuronal Fates
in Cerebellum. Neuron 47, 201–213 (2005).

127
26. Machold, R. & Fishell, G. Math1 Is Expressed in Temporally Discrete Pools of Cerebellar
Rhombic-Lip Neural Progenitors. Neuron 48, 17–24 (2005).
27. Wang, V. Y., Rose, M. F. & Zoghbi, H. Y. Math1 Expression Redefines the Rhombic Lip
Derivatives and Reveals Novel Lineages within the Brainstem and Cerebellum. Neuron 48,
31–43 (2005).
28. Yamada, M. et al. Specification of Spatial Identities of Cerebellar Neuron Progenitors by
Ptf1a and Atoh1 for Proper Production of GABAergic and Glutamatergic Neurons. J
Neurosci 34, 4786–4800 (2014).
29. Pascual, M. et al. Cerebellar GABAergic progenitors adopt an external granule cell-like
phenotype in the absence of Ptf1a transcription factor expression. Proc. Natl. Acad. Sci.
U.S.A. 104, 5193–5198 (2007).
30. Ben-Arie, N. et al. Math1 is essential for genesis of cerebellar granule neurons. Nature 390,
169–172 (1997).
31. Akazawa, C., Ishibashi, M., Shimizu, C., Nakanishi, S. & Kageyama, R. A Mammalian
Helix-Loop-Helix Factor Structurally Related to the Product of Drosophila Proneural Gene
atonal Is a Positive Transcriptional Regulator Expressed in the Developing Nervous
System. Journal of Biological Chemistry 270, 8730–8738 (1995).
32. Carter, R. A. et al. A Single-Cell Transcriptional Atlas of the Developing Murine
Cerebellum. Current Biology 28, 2910-2920.e2 (2018).
33. Wizeman, J. W., Guo, Q., Wilion, E. M. & Li, J. Y. Specification of diverse cell types during
early neurogenesis of the mouse cerebellum. eLife 8, e42388 (2019).
34. Kozareva, V. et al. A transcriptomic atlas of mouse cerebellar cortex comprehensively
defines cell types. Nature 598, 214–219 (2021).
35. Sarropoulos, I. et al. Developmental and evolutionary dynamics of cis-regulatory elements in
mouse cerebellar cells. Science 373, eabg4696 (2021).
36. Rodriques, S. G. et al. Slide-seq: A Scalable Technology for Measuring Genome-Wide
Expression at High Spatial Resolution. Science 363, 1463–1467 (2019).
37. Aldinger, K. A. et al. Spatial and cell type transcriptional landscape of human cerebellar
development. Nat Neurosci 24, 1163–1175 (2021).
38. Sepp, M. et al. Cellular Development and Evolution of the Mammalian Cerebellum.
http://biorxiv.org/lookup/doi/10.1101/2021.12.20.473443 (2021)
doi:10.1101/2021.12.20.473443.

128
39. Haldipur, P. et al. Spatiotemporal expansion of primary progenitor zones in the developing
human cerebellum. Science 366, 454–460 (2019).
40. Haldipur, P., Millen, K. J. & Aldinger, K. A. Human Cerebellar Development and
Transcriptomics: Implications for Neurodevelopmental Disorders. Annu. Rev. Neurosci. 45,
annurev-neuro-111020-091953 (2022).
41. Zhao, Z. et al. Organoids. Nat Rev Methods Primers 2, 94 (2022).
42. Takahashi, T., Fujishima, K. & Kengaku, M. Modeling Intestinal Stem Cell Function with
Organoids. Int J Mol Sci 22, 10912 (2021).
43. Grebenyuk, S. & Ranga, A. Engineering Organoid Vascularization. Front Bioeng Biotechnol
7, 39 (2019).
44. Qian, X. et al. Sliced Human Cortical Organoids for Modeling Distinct Cortical Layer
Formation. Cell Stem Cell 26, 766-781.e9 (2020).
45. Atamian, A., Cordón-Barris, L. & Quadrato, G. Taming human brain organoids one cell at a
time. Seminars in Cell & Developmental Biology 111, 23–31 (2021).
46. Del Dosso, A., Urenda, J.-P., Nguyen, T. & Quadrato, G. Upgrading the Physiological
Relevance of Human Brain Organoids. Neuron 107, 1014–1028 (2020).
47. Espuny-Camacho, I. et al. Pyramidal Neurons Derived from Human Pluripotent Stem Cells
Integrate Efficiently into Mouse Brain Circuits In Vivo. Neuron 77, 440–456 (2013).
48. Watanabe, K. et al. Directed differentiation of telencephalic precursors from embryonic stem
cells. Nat Neurosci 8, 288–296 (2005).
49. Eiraku, M. et al. Self-Organized Formation of Polarized Cortical Tissues from ESCs and Its
Active Manipulation by Extrinsic Signals. Cell Stem Cell 3, 519–532 (2008).
50. Elkabetz, Y. et al. Human ES cell-derived neural rosettes reveal a functionally distinct early
neural stem cell stage. Genes Dev 22, 152–165 (2008).
51. ten Berge, D., Brugmann, S. A., Helms, J. A. & Nusse, R. Wnt and FGF signals interact to
coordinate growth with cell fate specification during limb development. Development 135,
3247–3257 (2008).
52. Atamian, A. et al. Human cerebellar organoids with functional Purkinje cells. Cell Stem Cell
31, 39-51.e6 (2024).

129
53. He, Z. et al. An Integrated Transcriptomic Cell Atlas of Human Neural Organoids.
http://biorxiv.org/lookup/doi/10.1101/2023.10.05.561097 (2023)
doi:10.1101/2023.10.05.561097.
54. Camp, J. G. et al. Human cerebral organoids recapitulate gene expression programs of fetal
neocortex development. Proc Natl Acad Sci U S A 112, 15672–15677 (2015).
55. Marton, R. M. & Pașca, S. P. Organoid and Assembloid Technologies for Investigating
Cellular Crosstalk in Human Brain Development and Disease. Trends in Cell Biology 30,
133–143 (2020).
56. Osaki, T. et al. Complex activity and short-term plasticity of human cerebral organoids
reciprocally connected with axons. Nat Commun 15, 2945 (2024).
57. Kadoshima, T. et al. Self-organization of axial polarity, inside-out layer pattern, and speciesspecific progenitor dynamics in human ES cell–derived neocortex. Proc. Natl. Acad. Sci.
U.S.A. 110, 20284–20289 (2013).
58. Velasco, S. et al. Individual brain organoids reproducibly form cell diversity of the human
cerebral cortex. Nature 570, 523–527 (2019).
59. Paşca, A. M. et al. Functional cortical neurons and astrocytes from human pluripotent stem
cells in 3D culture. Nat Methods 12, 671–678 (2015).
60. Yoon, S.-J. et al. Reliability of human 3D cortical organoid generation. Nat Methods 16, 75–
78 (2019).
61. Trujillo, C. A. et al. Complex Oscillatory Waves Emerging from Cortical Organoids Model
Early Human Brain Network Development. Cell Stem Cell 25, 558-569.e7 (2019).
62. Qian, X. et al. Brain-Region-Specific Organoids Using Mini-bioreactors for Modeling ZIKV
Exposure. Cell 165, 1238–1254 (2016).
63. Rigamonti, A. et al. Large-Scale Production of Mature Neurons from Human Pluripotent
Stem Cells in a Three-Dimensional Suspension Culture System. Stem Cell Reports 6, 993–
1008 (2016).
64. Watanabe, M. et al. Self-Organized Cerebral Organoids with Human-Specific Features
Predict Effective Drugs to Combat Zika Virus Infection. Cell Reports 21, 517–532 (2017).
65. Bagley, J. A., Reumann, D., Bian, S., Lévi-Strauss, J. & Knoblich, J. A. Fused dorsal-ventral
cerebral organoids model complex interactions between diverse brain regions. Nat Methods
14, 743–751 (2017).

130
66. Eichmüller, O. L. & Knoblich, J. A. Human cerebral organoids — a new tool for clinical
neurology research. Nat Rev Neurol 18, 661–680 (2022).
67. Wilson, S. W. & Houart, C. Early Steps in the Development of the Forebrain. Developmental
Cell 6, 167–181 (2004).
68. Schmahmann, J. D. & Sherman, J. C. The cerebellar cognitive affective syndrome.
69. Zhang, P. et al. The cerebellum and cognitive neural networks. Front Hum Neurosci 17,
1197459 (2023).
70. Barton, R. A. & Venditti, C. Rapid Evolution of the Cerebellum in Humans and Other Great
Apes. Current Biology 24, 2440–2444 (2014).
71. Liu, A., Losos, K. & Joyner, A. L. FGF8 can activate Gbx2 and transform regions of the
rostral mouse brain into a hindbrain fate. Development 126, 4827–4838 (1999).
72. Martinez, S., Crossley, P. H., Cobos, I., Rubenstein, J. L. R. & Martin, G. R. FGF8 induces
formation of an ectopic isthmic organizer and isthmocerebellar development via a
repressive effect on Otx2 expression. Development 126, 1189–1200 (1999).
73. Garda, A.-L., Echevarrı́
a, D. & Martı́
nez, S. Neuroepithelial co-expression of Gbx2 and Otx2
precedes Fgf8 expression in the isthmic organizer. Mechanisms of Development 101, 111–
118 (2001).
74. Buchholz, D. E. et al. Novel genetic features of human and mouse Purkinje cell
differentiation defined by comparative transcriptomics. Proceedings of the National
Academy of Sciences 117, 15085–15095 (2020).
75. Sundberg, M. et al. Purkinje cells derived from TSC patients display hypoexcitability and
synaptic deficits associated with reduced FMRP levels and reversed by rapamycin. Mol
Psychiatry 23, 2167–2183 (2018).
76. Quadrato, G. et al. Cell diversity and network dynamics in photosensitive human brain
organoids. Nature 545, 48–53 (2017).
77. Nordström, U., Jessell, T. M. & Edlund, T. Progressive induction of caudal neural character
by graded Wnt signaling. Nat Neurosci 5, 525–532 (2002).
78. Green, D., Whitener, A. E., Mohanty, S. & Lekven, A. C. Vertebrate nervous system
posteriorization: Grading the function of Wnt signaling. Developmental Dynamics 244,
507–512 (2015).
79. Nakamura, H., Katahira, T., Matsunaga, E. & Sato, T. Isthmus organizer for midbrain and
hindbrain development. Brain Research Reviews 49, 120–126 (2005).

131
80. Chi, C. L., Martinez, S., Wurst, W. & Martin, G. R. The isthmic organizer signal FGF8 is
required for cell survival in the prospective midbrain and cerebellum. Development 130,
2633–2644 (2003).
81. Smith, K. S. et al. Unified rhombic lip origins of group 3 and group 4 medulloblastoma.
Nature 609, 1012–1020 (2022).
82. Hendrikse, L. D. et al. Failure of human rhombic lip differentiation underlies
medulloblastoma formation. Nature 609, 1021–1028 (2022).
83. Tanaka, D., Nakaya, Y., Yanagawa, Y., Obata, K. & Murakami, F. Multimodal tangential
migration of neocortical GABAergic neurons independent of GPI-anchored proteins.
Development 130, 5803–5813 (2003).
84. Ang, E. S. B. C., Haydar, T. F., Gluncic, V. & Rakic, P. Four-Dimensional Migratory
Coordinates of GABAergic Interneurons in the Developing Mouse Cortex. J Neurosci 23,
5805–5815 (2003).
85. Tanaka, D. H., Maekawa, K., Yanagawa, Y., Obata, K. & Murakami, F. Multidirectional and
multizonal tangential migration of GABAergic interneurons in the developing cerebral
cortex. Development 133, 2167–2176 (2006).
86. Martini, F. J. et al. Biased selection of leading process branches mediates chemotaxis during
tangential neuronal migration. Development 136, 41–50 (2009).
87. Britto, J. M., Johnston, L. A. & Tan, S.-S. The Stochastic Search Dynamics of Interneuron
Migration. Biophys J 97, 699–709 (2009).
88. Tanaka, D. H. et al. Random Walk Behavior of Migrating Cortical Interneurons in the
Marginal Zone: Time-Lapse Analysis in Flat-Mount Cortex. J Neurosci 29, 1300–1311
(2009).
89. Hansen, D. V. et al. Non-epithelial stem cells and cortical interneuron production in the
human ganglionic eminences. Nat Neurosci 16, 1576–1587 (2013).
90. Ma, T. et al. Subcortical origins of human and monkey neocortical interneurons. Nat
Neurosci 16, 1588–1597 (2013).
91. Barlow, C. et al. Atm-Deficient Mice: A Paradigm of Ataxia Telangiectasia. Cell 86, 159–
171 (1996).
92. Choy, K. R. & Watters, D. J. Neurodegeneration in ataxia-telangiectasia: Multiple roles of
ATM kinase in cellular homeostasis: ATM and Cellular Homeostasis. Dev. Dyn. 247, 33–
46 (2018).

132
93. Rahimi-Balaei, M. et al. Embryonic stages in cerebellar afferent development. Cerebellum
Ataxias 2, 7 (2015).
94. Muguruma, K., Nishiyama, A., Kawakami, H., Hashimoto, K. & Sasai, Y. Self-Organization
of Polarized Cerebellar Tissue in 3D Culture of Human Pluripotent Stem Cells. Cell
Reports 10, 537–550 (2015).
95. Silva, T. P. et al. Maturation of Human Pluripotent Stem Cell-Derived Cerebellar Neurons in
the Absence of Co-culture. Frontiers in Bioengineering and Biotechnology 8, (2020).
96. Silva, T. P. et al. Transcriptome profiling of human pluripotent stem cell-derived cerebellar
organoids reveals faster commitment under dynamic conditions. Biotechnology and
Bioengineering 118, 2781–2803 (2021).
97. Nayler, S., Agarwal, D., Curion, F., Bowden, R. & Becker, E. B. E. High-resolution
transcriptional landscape of xeno-free human induced pluripotent stem cell-derived
cerebellar organoids. Sci Rep 11, 12959 (2021).
98. Chen, Y. et al. Generation of advanced cerebellar organoids for neurogenesis and neuronal
network development. Human Molecular Genetics 32, 2832–2841 (2023).
99. Muguruma, K. et al. Ontogeny-recapitulating generation and tissue integration of ES cell–
derived Purkinje cells. Nat Neurosci 13, 1171–1180 (2010).
100. Hidalgo-Sánchez, M., Millet, S., Simeone, A., & Rosa-Magda Alvarado-Mallart.
Comparative analysis of Otx2, Gbx2, Pax2, Fgf8 and Wnt1 gene expressions during the
formation of the chick midbrain/hindbrain domain. Mechanisms of Development 81, 175–
178 (1999).
101. Madhavan, M. et al. Induction of Myelinating Oligodendrocytes in Human Cortical
Spheroids. Nat Methods 15, 700–706 (2018).
102. Marton, R. M. et al. Differentiation and maturation of oligodendrocytes in human threedimensional neural cultures. Nat Neurosci 22, 484–491 (2019).
103. Cakir, B. et al. Development of human brain organoids with functional vascular-like
system. Nat Methods 16, 1169–1175 (2019).
104. Lin, Y.-T. et al. APOE4 Causes Widespread Molecular and Cellular Alterations Associated
with Alzheimer’s Disease Phenotypes in Human iPSC-Derived Brain Cell Types. Neuron
98, 1141-1154.e7 (2018).
105. Song, L. et al. Functionalization of Brain Region-specific Spheroids with Isogenic
Microglia-like Cells. Sci Rep 9, 11055 (2019).

133
106. Bejoy, J. et al. Genomics Analysis of Metabolic Pathways of Human Stem Cell-Derived
Microglia-Like Cells and the Integrated Cortical Spheroids. Stem Cells Int 2019, 2382534
(2019).
107. Popova, G. et al. Human microglia states are conserved across experimental models and
regulate neural stem cell responses in chimeric organoids. Cell Stem Cell S1934-
5909(21)00376–3 (2021) doi:10.1016/j.stem.2021.08.015.
108. Marek, S. et al. Spatial and Temporal Organization of the Individual Human Cerebellum.
Neuron 100, 977-993.e7 (2018).
109. Cutando, L. et al. Cerebellar dopamine D2 receptors regulate social behaviors. Nat
Neurosci 25, 900–911 (2022).
110. Wang, J., Garancher, A., Ramaswamy, V. & Wechsler-Reya, R. J. Medulloblastoma: From
Molecular Subgroups to Molecular Targeted Therapies. Annu. Rev. Neurosci. 41, 207–232
(2018).
111. Mapelli, L., Soda, T., D’Angelo, E. & Prestori, F. The Cerebellar Involvement in Autism
Spectrum Disorders: From the Social Brain to Mouse Models. IJMS 23, 3894 (2022).
112. Sathyanesan, A. et al. Emerging connections between cerebellar development, behaviour
and complex brain disorders. Nat Rev Neurosci 20, 298–313 (2019).
113. Fatemi, S. H. & Folsom, T. D. Autism Spectrum Disorders and Ataxia. in Handbook of the
Cerebellum and Cerebellar Disorders (eds. Manto, M., Schmahmann, J. D., Rossi, F.,
Gruol, D. L. & Koibuchi, N.) 1895–1906 (Springer Netherlands, Dordrecht, 2013).
doi:10.1007/978-94-007-1333-8_87.
114. Watson, L. M., Wong, M. M. K., Vowles, J., Cowley, S. A. & Becker, E. B. E. A
Simplified Method for Generating Purkinje Cells from Human-Induced Pluripotent Stem
Cells. Cerebellum 17, 419–427 (2018).
115. Behesti, H., Kocabas, A., Buchholz, D. E., Carroll, T. S. & Hatten, M. E. Altered temporal
sequence of transcriptional regulators in the generation of human cerebellar granule cells.
eLife 10, e67074 (2021).
116. Jo, J. et al. Midbrain-like organoids from human pluripotent stem cells contain functional
dopaminergic and neuromelanin producing neurons. Cell Stem Cell 19, 248–257 (2016).
117. Kumamaru, H. et al. Generation and post-injury integration of human spinal cord neural
stem cells. Nat Methods 15, 723–731 (2018).
118. Sato, T., Araki, I. & Nakamura, H. Inductive signal and tissue responsiveness defining the
tectum and the cerebellum. Development 128, 2461–2469 (2001).

134
119. Wataya, T. et al. Minimization of exogenous signals in ES cell culture induces rostral
hypothalamic differentiation. Proc. Natl. Acad. Sci. U.S.A. 105, 11796–11801 (2008).
120. Fleck, J. S. et al. Resolving organoid brain region identities by mapping single-cell genomic
data to reference atlases. Cell Stem Cell 28, 1148-1159.e8 (2021).
121. He, X. et al. Integrated Model of De Novo and Inherited Genetic Variants Yields Greater
Power to Identify Risk Genes. PLoS Genet 9, e1003671 (2013).
122. Siletti, K. et al. Transcriptomic Diversity of Cell Types across the Adult Human Brain.
http://biorxiv.org/lookup/doi/10.1101/2022.10.12.511898 (2022)
doi:10.1101/2022.10.12.511898.
123. Xu, H. et al. Bergmann Glia Function in Granule Cell Migration During Cerebellum
Development. Mol Neurobiol 47, 833–844 (2013).
124. Kompaníková, P. & Bryja, V. Regulation of choroid plexus development and its functions.
Cell. Mol. Life Sci. 79, 304 (2022).
125. Chizhikov, V. V. et al. The roof plate regulates cerebellar cell-type specification and
proliferation. Development 133, 2793–2804 (2006).
126. Marzban, H. et al. Cellular commitment in the developing cerebellum. Front Cell Neurosci
8, 450 (2015).
127. Klein, R. S. et al. SDF-1α induces chemotaxis and enhances Sonic hedgehog-induced
proliferation of cerebellar granule cells. Development 128, 1971–1981 (2001).
128. Zou, Y.-R., Kottmann, A. H., Kuroda, M., Taniuchi, I. & Littman, D. R. Function of the
chemokine receptor CXCR4 in haematopoiesis and in cerebellar development. Nature 393,
595–599 (1998).
129. Reiss, K., Mentlein, R., Sievers, J. & Hartmann, D. Stromal cell-derived factor 1 is secreted
by meningeal cells and acts as chemotactic factor on neuronal stem cells of the cerebellar
external granular layer. Neuroscience 115, 295–305 (2002).
130. Haldipur, P. et al. Foxc1 dependent mesenchymal signalling drives embryonic cerebellar
growth. eLife 3, e03962.
131. Stumm, R. K. et al. CXCR4 Regulates Interneuron Migration in the Developing Neocortex.
J. Neurosci. 23, 5123–5130 (2003).
132. Tran, P. B., Banisadr, G., Ren, D., Chenn, A. & Miller, R. J. Chemokine Receptor
Expression by Neural Progenitor Cells in Neurogenic Regions of Mouse Brain. J Comp
Neurol 500, 1007–1033 (2007).

135
133. Samarasinghe, R. A. et al. Identification of neural oscillations and epileptiform changes in
human brain organoids. Nat Neurosci 24, 1488–1500 (2021).
134. Giovannucci, A. et al. CaImAn an open source tool for scalable calcium imaging data
analysis. eLife 8, e38173 (2019).
135. Mahn, M. et al. Efficient optogenetic silencing of neurotransmitter release with a mosquito
rhodopsin. Neuron 109, 1621-1635.e8 (2021).
136. Gruol, D. & Franklin, C. Morphological and physiological differentiation of Purkinje
neurons in cultures of rat cerebellum. J Neurosci 7, 1271–1293 (1987).
137. Raman, I. M. & Bean, B. P. Ionic Currents Underlying Spontaneous Action Potentials in
Isolated Cerebellar Purkinje Neurons. J Neurosci 19, 1663–1674 (1999).
138. Rakic, P. & Sidman, R. L. Histogenesis of cortical layers in human cerebellum, particularly
the lamina dissecans. J. Comp. Neurol. 139, 473–500 (1970).
139. Haldipur, P. et al. Evidence of disrupted rhombic lip development in the pathogenesis of
Dandy-Walker malformation. Acta Neuropathol 142, 761–776 (2021).
140. Lobsiger, C. S. & Cleveland, D. W. Glial cells as intrinsic components of non-cellautonomous neurodegenerative disease. Nat Neurosci 10, 1355–1360 (2007).
141. Urenda, J.-P., Del Dosso, A., Birtele, M. & Quadrato, G. Present and Future Modeling of
Human Psychiatric Connectopathies With Brain Organoids. Biological Psychiatry 93, 606–
615 (2023).
142. Ogura, T., Sakaguchi, H., Miyamoto, S. & Takahashi, J. Three-dimensional induction of
dorsal, intermediate and ventral spinal cord tissues from human pluripotent stem cells.
Development 145, dev162214 (2018).
143. Giovannucci, A. et al. CaImAn an open source tool for scalable calcium imaging data
analysis. eLife 8, e38173 (2019).
144. Lois, C., Hong, E. J., Pease, S., Brown, E. J. & Baltimore, D. Germline Transmission and
Tissue-Specific Expression of Transgenes Delivered by Lentiviral Vectors. Science 295,
868–872 (2002).
145. Ahlmann-Eltze, C. & Huber, W. Comparison of Transformations for Single-Cell RNA-Seq
Data. http://biorxiv.org/lookup/doi/10.1101/2021.06.24.449781 (2021)
doi:10.1101/2021.06.24.449781.
146. Fatemi, S. H. et al. Consensus Paper: Pathological Role of the Cerebellum in Autism.
Cerebellum 11, 777–807 (2012).

136
147. Klockgether, T., Mariotti, C. & Paulson, H. L. Spinocerebellar ataxia. Nat Rev Dis Primers
5, 24 (2019).
148. Aldinger, K. A. et al. Redefining the Etiologic Landscape of Cerebellar Malformations. The
American Journal of Human Genetics 105, 606–615 (2019).
149. Guo, C. C. et al. Network-selective vulnerability of the human cerebellum to Alzheimer’s
disease and frontotemporal dementia. Brain 139, 1527–1538 (2016).
150. Ten Donkelaar, H. J., Lammens, M., Wesseling, P., Thijssen, H. O. & Renier, W. O.
Development and developmental disorders of the human cerebellum. Journal of Neurology
250, 1025–1036 (2003).
151. Wang, V. Y. & Zoghbi, H. Y. Genetic regulation of cerebellar development. Nat Rev
Neurosci 2, 484–491 (2001).
152. Satterstrom, F. K. et al. Large-Scale Exome Sequencing Study Implicates Both
Developmental and Functional Changes in the Neurobiology of Autism. Cell 180, 568-
584.e23 (2020).
153. Jimenez-Gomez, A. et al. Phenotypic characterization of individuals with SYNGAP1
pathogenic variants reveals a potential correlation between posterior dominant rhythm and
developmental progression. J Neurodev Disord 11, 18 (2019).
154. Mignot, C. et al. Genetic and neurodevelopmental spectrum of SYNGAP1 -associated
intellectual disability and epilepsy. J Med Genet 53, 511–522 (2016).
155. Vlaskamp, D. R. M. et al. SYNGAP1 encephalopathy. Neurology 92, e96–e107 (2019).
156. Wright, D., Kenny, A., Eley, S., McKechanie, A. G. & Stanfield, A. C. Clinical and
behavioural features of SYNGAP1-related intellectual disability: a parent and caregiver
description. J Neurodevelop Disord 14, 34 (2022).
157. Li, W. et al. Characterization of a Novel synGAP Isoform, synGAP-β. Journal of
Biological Chemistry 276, 21417–21424 (2001).
158. McMahon, A. C. et al. SynGAP isoforms exert opposing effects on synaptic strength. Nat
Commun 3, 900 (2012).
159. Tomoda, T., Kim, J. H., Zhan, C. & Hatten, M. E. Role of Unc51.1 and its binding partners
in CNS axon outgrowth. Genes Dev 18, 541–558 (2004).
160. Knuesel, I., Elliott, A., Chen, H., Mansuy, I. M. & Kennedy, M. B. A role for synGAP in
regulating neuronal apoptosis. Eur J of Neuroscience 21, 611–621 (2005).

137
161. Willsey, H. R. et al. Parallel in vivo analysis of large-effect autism genes implicates cortical
neurogenesis and estrogen in risk and resilience. Neuron 109, 788-804.e8 (2021).
162. Su, P. et al. Disruption of SynGAP–dopamine D1 receptor complexes alters actin and
microtubule dynamics and impairs GABAergic interneuron migration. Sci. Signal. 12,
eaau9122 (2019).
163. Berryer, M. H. et al. Mutations in SYNGAP1 Cause Intellectual Disability, Autism, and a
Specific Form of Epilepsy by Inducing Haploinsufficiency. Human Mutation 34, 385–394
(2013).
164. Kilinc, M. et al. Endogenous Syngap1 alpha splice forms promote cognitive function and
seizure protection. eLife 11, e75707.
165. Gamache, T. R., Araki, Y. & Huganir, R. L. Twenty Years of SynGAP Research: From
Synapses to Cognition. J Neurosci 40, 1596–1605 (2020).
166. Birtele, M. et al. Non-synaptic function of the autism spectrum disorder-associated gene
SYNGAP1 in cortical neurogenesis. Nat Neurosci 26, 2090–2103 (2023).
167. Chen, H.-J., Rojas-Soto, M., Oguni, A. & Kennedy, M. B. A Synaptic Ras-GTPase
Activating Protein (p135 SynGAP) Inhibited by CaM Kinase II. Neuron 20, 895–904
(1998).
168. Kim, J. H., Liao, D., Lau, L.-F. & Huganir, R. L. SynGAP: a Synaptic RasGAP that
Associates with the PSD-95/SAP90 Protein Family. Neuron 20, 683–691 (1998).
169. Komiyama, N. H. et al. SynGAP Regulates ERK/MAPK Signaling, Synaptic Plasticity, and
Learning in the Complex with Postsynaptic Density 95 and NMDA Receptor. J Neurosci
22, 9721–9732 (2002).
170. Kim, J. H., Lee, H.-K., Takamiya, K. & Huganir, R. L. The Role of Synaptic GTPaseActivating Protein in Neuronal Development and Synaptic Plasticity. J. Neurosci. 23,
1119–1124 (2003).
171. Zhu, J. J., Qin, Y., Zhao, M., Van Aelst, L. & Malinow, R. Ras and Rap Control AMPA
Receptor Trafficking during Synaptic Plasticity. Cell 110, 443–455 (2002).
172. Araki, Y., Zeng, M., Zhang, M. & Huganir, R. L. Rapid Dispersion of SynGAP from
Synaptic Spines Triggers AMPA Receptor Insertion and Spine Enlargement during LTP.
Neuron 85, 173–189 (2015).
173. Walkup, W. G. et al. A model for regulation by SynGAP-α1 of binding of synaptic proteins
to PDZ-domain ‘Slots’ in the postsynaptic density. eLife 5, e16813.

138
174. Zeng, M. et al. Phase Transition in Postsynaptic Densities Underlies Formation of Syaptic
Complexes and Synaptic Plasticity. Cell 166, 1163-1175.e12 (2016).
175. Zeng, M., Bai, G. & Zhang, M. Anchoring high concentrations of SynGAP at postsynaptic
densities via liquid-liquid phase separation. Small GTPases 10, 296–304 (2017).
176. Kilinc, M. et al. Species-Conserved SYNGAP1 Phenotypes Associated with
Neurodevelopmental Disorders. Mol Cell Neurosci 91, 140–150 (2018).
177. The DDD Study et al. Synaptic, transcriptional and chromatin genes disrupted in autism.
Nature 515, 209–215 (2014).
178. Paulsen, B. et al. Autism genes converge on asynchronous development of shared neuron
classes. Nature 602, 268–273 (2022).
179. Mariani, J. et al. FOXG1-dependent dysregulation of GABA/glutamate neuron
differentiation in autism spectrum disorders. Cell 162, 375–390 (2015).
180. Jourdon, A. et al. Modelling idiopathic autism in forebrain organoids reveals an imbalance
of excitatory cortical neuron subtypes during early neurogenesis. Nat Neurosci 26, 1505–
1515 (2023).
181. Villa, C. E. et al. CHD8 haploinsufficiency links autism to transient alterations in excitatory
and inhibitory trajectories. Cell Reports 39, 110615 (2022).
182. Schafer, S. T. et al. Pathological priming causes developmental gene network
heterochronicity in autism patient-derived neurons. Nat Neurosci 22, 243–255 (2019).
183. De Jong, J. O. et al. Cortical overgrowth in a preclinical forebrain organoid model of
CNTNAP2-associated autism spectrum disorder. Nat Commun 12, 4087 (2021).
184. Aaku-Saraste, E., Hellwig, A. & Huttner, W. B. Loss of Occludin and Functional Tight
Junctions, but Not ZO-1, during Neural Tube Closure—Remodeling of the
Neuroepithelium Prior to Neurogenesis. Developmental Biology 180, 664–679 (1996).
185. Berryer, M. H. et al. Decrease of SYNGAP1 in GABAergic cells impairs inhibitory
synapse connectivity, synaptic inhibition and cognitive function. Nat Commun 7, 13340
(2016).
186. Noctor, S. C., Martínez-Cerdeño, V., Ivic, L. & Kriegstein, A. R. Cortical neurons arise in
symmetric and asymmetric division zones and migrate through specific phases. Nat
Neurosci 7, 136–144 (2004).

139
187. Shuijin, H., Zhizhong, L., Shaoyu, G., Yong-Chun, Y. & Song-Hai, S. Inside-out radial
migration facilitates lineage-dependent neocortical microcircuit assembly. Neuron 86,
1159–1166 (2015).
188. Okita, K. et al. An Efficient Nonviral Method to Generate Integration-Free Human-Induced
Pluripotent Stem Cells from Cord Blood and Peripheral Blood Cells. Stem Cells 31, 458–
466 (2013).
189. Campbell, K. R. & Yau, C. Order Under Uncertainty: Robust Differential Expression
Analysis Using Probabilistic Models for Pseudotime Inference. PLoS Comput Biol 12,
e1005212 (2016).
190. Tritschler, S. et al. Concepts and limitations for learning developmental trajectories from
single cell genomics. Development 146, dev170506 (2019).
191. Zhang, T. et al. Generation of excitatory and inhibitory neurons from common progenitors
via Notch signaling in the cerebellum. Cell Reports 35, 109208 (2021).
192. Eyo, U. B. & Dailey, M. E. Microglia: Key Elements in Neural Development, Plasticity,
and Pathology. J Neuroimmune Pharmacol 8, 494–509 (2013).
193. Parkhurst, C. N. et al. Microglia promote learning-dependent synapse formation through
BDNF. Cell 155, 1596–1609 (2013).
194. Andoh, M. & Koyama, R. Microglia regulate synaptic development and plasticity. Dev
Neurobiol 81, 568–590 (2021).
195. Grabert, K. et al. Microglial brain region-dependent diversity and selective regional
sensitivities to ageing. Nat Neurosci 19, 504–516 (2016).
196. Coomey, R., Stowell, R., Majewska, A. & Tropea, D. The Role of Microglia in
Neurodevelopmental Disorders and Their Therapeutics. Curr Top Med Chem 20, 272–276
(2020).
197. Ferro, A., Sheeler, C., Rosa, J.-G. & Cvetanovic, M. Role of Microglia in Ataxias. J Mol
Biol 431, 1792–1804 (2019).
198. Schafer, D. P. & Stevens, B. Microglia Function in Central Nervous System Development
and Plasticity. Cold Spring Harb Perspect Biol 7, a020545 (2015).
199. Park, D. S. et al. iPS-cell-derived microglia promote brain organoid maturation via
cholesterol transfer. Nature 623, 397–405 (2023).
200. Cottrell, D. A. et al. The natural history of multiple sclerosis: a geographically based study.

140
201. Koch, M., Mostert, J., Heersema, D. & De Keyser, J. Tremor in multiple sclerosis. J Neurol
254, 133–145 (2007).
202. D’Ambrosio, A. et al. Cerebellar contribution to motor and cognitive performance in
multiple sclerosis: An MRI sub-regional volumetric analysis. Mult Scler 23, 1194–1203
(2017).
203. Blanco, Y., Compta, Y., Graus, F. & Saiz, A. Midbrain lesions and paroxysmal dysarthria
in multiple sclerosis. Mult Scler 14, 694–697 (2008).
204. Zlokovic, B. V. Neurovascular pathways to neurodegeneration in Alzheimer’s disease and
other disorders. Nat Rev Neurosci 12, 723–738 (2011).
205. Sun, X.-Y. et al. Generation of vascularized brain organoids to study neurovascular
interactions. eLife 11, e76707 (2022).
206. Wörsdörfer, P. et al. Generation of complex human organoid models including vascular
networks by incorporation of mesodermal progenitor cells. Sci Rep 9, 15663 (2019).
207. Matsui, T. K., Tsuru, Y., Hasegawa, K. & Kuwako, K. Vascularization of Human Brain
Organoids. Stem Cells 39, 1017–1024 (2021).
208. Karamazovova, S., Matuskova, V., Ismail, Z. & Vyhnalek, M. Neuropsychiatric symptoms
in spinocerebellar ataxias and Friedreich ataxia. Neuroscience & Biobehavioral Reviews
150, 105205 (2023).

Abstract (if available)

Abstract In this thesis, I will discuss a new protocol to obtain reproducible cerebellar organoids derived from human pluripotent stem cells (hCerOs). When I first joined Giorgia’s laboratory as a technician, I began my work trying to optimize a protocol that would allow me to generate cerebellar organoid with the least amount of forebrain contamination. Long-term culture of this optimized protocol yielded organoids that had a high degree of reproducibility in terms of its cellular composition determined through single cell RNA sequencing and immunohistochemistry (IHC) as well as reproducible functional outputs determined by neuronal calcium dynamics (GCaMP recordings). Moreover, we were able to identify the human specific rhombic lip subventricular zone from our 2-month-old single cell RNA seq data, which has recently been identified as a human specific feature. Additionally, through whole cell patch clamping and spatial transcriptomics, we determined that these organoids could generate Purkinje cells that are transcriptomically and functionally similar to their in vivo counterparts after 6-months in culture. This model system was therefore an ideal model to probe the cerebellum’s phenotype within a SYNGAP1 happloinsufficient (SYNGAP1+/-) context. Through scRNAseq, we see a trend toward a delayed maturation phenotype within our hCerOs opposite to what our lab has previously shown with cortical projection neurons in cortical organoids.
In summary, with great help and support from my mentor Dr. Giorgia Quadrato and my co- authors, we have created a model system to study the human cerebellar ontogenesis. This model can now be used to probe the emergence of cerebellar cell types and study various human cerebellar disorders such as ataxia and medulloblastoma. The use of these hCerOs to study SYNGAP1+/- also highlights the importance of having a system that can explore how neurodevelopmental disorders present themselves in various brain regions to gain a wholistic picture of disease pathogenesis, which in turn will aid in the emergence of better therapeutics.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The autism-associated gene SYNGAP1 regulates human cortical neurogenesis

PDF

Elucidating neurodevelopmental consequences of syngap1 mutations and inactivated functional regions in human iPSC-derived neurons

PDF

Understanding human nephrogenesis and scaling synthesis of organoids facilitate modeling of kidney development and disease

PDF

Identification of therapeutic targets in human cerebral brain organoid models of neurodegeneration

PDF

Modeling SynGAP1 truncating mutations in neurodevelopmental disease using iPSC-derived neurons

PDF

Cross-species comparison of non-canonical Wnt signaling in the developing retina

PDF

Transcriptomic maturation of developing human cone precursors in fetal and 3D hESC-derived tissues

PDF

Signaling networks in complex brain disorders

PDF

Blood-brain barrier breakdown and vessel integrity in Alzheimer’s disease

PDF

Investigating molecular roadblocks to enhance direct cellular reprogramming

PDF

The role of KMT2C(MLL3) in precursor biology during early tumor development in HNSCC

PDF

Derivation, expansion and characterization of human hippocampal primordial cells from normal and diseased iPSCs

PDF

Blood-brain barrier pathophysiology in cognitive impairment and injury

PDF

Development and evolution of skeletal joints: lessons learned from studying zebrafish

PDF

Long-term expansion of human alveolar epithelial cells as a novel model system to study lung disease progression in vitro

PDF

DNA methylation in the mouse cochlea promotes maturation of supporting cells and contributes to the failure of hair cell regeneration

PDF

The epigenetic landscapes underlying differentiation and plasticity in the developing organ of Corti

PDF

Transcriptional and epigenetic mechanisms underlying sensory hair cell differentiation and regeneration

PDF

Functional study of Snord118 in microglila

PDF

Vascular contributions to brain aging along the Alzheimer's disease continuum

Asset Metadata

Creator Atamian, Alexander (author)

Core Title Generation and long-term culture of human cerebellar organoids to study development and disease.

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Neuroscience

Degree Conferral Date 2024-08

Publication Date 08/26/2024

Defense Date 08/24/2024

Publisher Los Angeles, California (original), University of Southern California (original), University of Southern California. Libraries (digital)

Tag brain organoid,cerebellar development,cerebellar organoid,disease modeling,iPSC,OAI-PMH Harvest,organoids,SynGAP1

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Ichida, Justin (committee chair), McMahon, Andrew (committee member), Quadrato, Giorgia (committee member)

Creator Email atamiana@usc.edu;alexander.atamian@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113999JQK

Unique identifier UC113999JQK

Identifier etd-AtamianAle-13428.pdf (filename)

Legacy Identifier etd-AtamianAle-13428

Document Type Dissertation

Format theses (aat)

Rights Atamian, Alexander

Internet Media Type application/pdf

Type texts

Source 20240827-usctheses-batch-1202 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu