University of Southern California Dissertations and Theses

Generation of patterned and functional kidney organoids that recapitulate the adult kidney collecting duct system

(USC Thesis Other)

Generation of patterned and functional kidney organoids that recapitulate the adult kidney collecting duct system

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Generation of Patterned and Functional Kidney Organoids that Recapitulate the Adult Kidney
Collecting Duct System

by

Zipeng Zeng

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requitements for the Degree
DOCTOR OF PHILOSOPHY
(DEVELOPMENT, STEM CELLS, AND REGENERATIVE MEDICINE)

May 2023

Copyright 2023 Zipeng Zeng
ii
Acknowledgements
I would like to sincerely thank Dr. Zhongwei Li for his guidance and support throughout
my graduate study journey. In the past six years I have experienced the best mentorship and
friendship in my life from him that not only made me a better scientist, but also a better person.
The positive influences I received from him will for sure live with me throughout my life and be
passed to more people. I would also like to express deep gratitude to my dissertation committee
Dr. Qi-long Ying (Chair), Dr. Andrew McMahon, Dr. Nuria Pastor-Soler, and Dr. Unmesh Jadhav
for their guidance and support throughout the years of my graduate study.
iii
Table of Contents
Acknowledgements……………………………………………………………………………………….ii
List of Tables………………………………………………………………………………………………iv
List of Figures...……………………………………………………………………………………………v
Abbreviations….…………………………………………………………………………………………vii
Abstract…………………………………………………………………………………………...…......viii
Introduction………………………………………………………………………………………………..1
Kidney function and development……………………………………………………..1
Current kidney organoid models and Knowledge gaps……………………………...3
Dissertation aims and key findings…………………………………………………….4
Chapter 1. Generation of expandable ureteric bud organoids from primary cells or human
pluripotent stem cells………………………………………………………………………..6
Abstract…………………………………………………………………………………..6
Introduction………………………………………………………………………….......7
Results……………………………………………………………………………………9
Discussion………………………………………………………………………………24
Chapter 2. Generation of patterned kidney organoids that recapitulate the adult kidney
collecting duct system from expandable ureteric bud progenitors ………………........26
Abstract………………………………………………………………………………....26
Introduction…………………………………………………………………….............27
Results………………………………………………………………………………….29
Discussion………………………………………………………………………………35
Conclusions and Future Directions………………………………………………………………........37
Methods…………………………………………………………………………………………............38
Supplemental Information……………………………………………………………………..............48
Bibliography……………………………………………………………………………………………...60
Appendix…………………………………………………………………………………………………67

iv
List of Tables
Supplementary Table 1. Summary of available in vitro models for mouse kidney branching
morphogenesis………………………………………………………………………………….59
Supplementary Table 2. Summary of available in vitro models for human kidney branching
morphogenesis………………………………………………………………………………….59
Supplementary Tables 3. Key resources…………………………………………………………...76
Supplementary table 4. Genotyping primer sequences for detecting PAX2-mCherry and
WNT11-GFP knockin (KI)……………………………………………………………………..79
Supplementary table 5. qRT-PCR Primer sequences ……………………………………………79
Supplementary table 6. Medium recipe of mUBCM (mouse UB culture medium) …………….80
Supplementary table 7. Medium recipe of hUBCM (Human UB culture medium).…………….81
Supplementary table 8. Medium recipe of CDDM (CD differentiation medium).……………….82
Supplementary table 9. Medium recipe of stepwise directed differentiation to ND cells
(D1 to D7 from hPSC)………………………………………………………………………….83

v
List of Figures
Figure 1. Diagram showing the human kidney structure…………………………………………….1
Figure 2. Diagrams showing the different perspectives of kidney development…………………..2
Figure 3. Schematic showing the broad application of organoids for studying development
and diseases……………………………………………………………..……………………4
Figure 1-1. Expanding mouse UB progenitor cells into 3D branching UB organoids…………...11
Figure 1-2. Generating engineered mouse kidney from expandable NPCs and UBs and
gene editing of the mouse UB organoid…………………………………………………..13
Figure 1-3. Generating human UB organoids from primary human UPCs and dual-reporter
human pluripotent stem cells………………………………………………………………16
Figure 1-4. Generating human iUB organoids from human pluripotent stem cells
independent of reporters……………………………………………………………………19
Figure 1-5. Modeling kidney development and disease using mouse and human UB
organoids…………………………………………………………………………………….22
Figure 2-1. Generating mature and highly organized mouse CD organoids from UB
organoids…………………………………………………………………………………….30
Figure 2-2. Generating human iCD organoids from hPSC-derived iUB organoids……………...32
Figure 2-3. Generating matured human iCD organoids from iUB organoids…………………….34
Supplementary Figure 1. Screening for optimal UB culture condition…………………………..48
Supplementary Figure 2. Derivation and characterization of mouse UB organoid…………….49
Supplementary Figure 3. Expandable UB organoid-based screening for CD differentiation…50
Supplementary Figure 4. Genetic engineering hESC with a dual reporter system by
CRISPR/Cas9 and generation of iUB organoids from various hPSC reporter lines…52
Supplementary Figure 5. Generating human iUB and iCD organoids from the dual-reporter
hPSC line independent of its reporters……………………………………………………53

vi
Supplementary Figure 6. Generating human iUB organoids from the SOX9-GFP iPSC
independent of its reporter…………………………………………………………………54
Supplementary Figure 7. Immunostaining of whole-mount samples of control and RET KO
iUB organoids and human CD antibodies testing………………………………………..56
Supplementary Figure 8. FACS sequential gating/sorting strategies used in iUB
differentiation………………………………………………………………………………...57

vii
Abbreviations
BF: Bright field
CAKUT: Congenital anomalies of kidney and urinary tract
CD: Collect duct
CDDM: Collect duct differentiation medium
GFP: Green fluorescence protein
hPSC: Human pluripotent stem cell
hESC: Human embryonic stem cell
hiPSC: Human induced pluripotent stem cell
IC: Intercalated cell
IPC: Interstitial progenitor cell
KO: Knockout
MM: Metanephric mesenchyme
ND: Nephric duct
NPC: Nephron progenitor cell
OE: Over expression
PC: Principal cell
PCA: Principal component analysis
qRT-PCR: quantitative reverse transcription polymerase chain reaction
RFP: Red fluorescence protein
UB: Ureteric bud
UBCM: Ureteric bud culture medium
UPC: Ureteric progenitor cell

viii
Abstract
Human kidney is responsible for filtering and processing blood and maintaining body fluid
homeostasis. Normal kidney function is critical for supporting human life. However, there are still
many knowledge gaps in the study of kidney development and diseases. One of the reasons is
the lack of high-quality model that represent authentic human kidney for modeling development
and disease pathogenesis. In the past decade, organoid systems leveraging the power of cell
self-organization and 3-dementional cell culture have been adopted and became powerful tools
for kidney research. Although various nephron organoid models have been reported, high-quality
organoid system modeling the kidney collecting duct (CD) system, or it’s precursor, ureteric bud
(UB) is still lacking. Here in this dissertation, in Chapter 1, I report the generation of expandable,
3D branching ureteric bud (UB) organoid culture model that can be derived from primary UB
progenitors from mouse and human fetal kidneys, or generated de novo from human pluripotent
stem cells. In Chapter 2, I report the generation of mouse and human CD organoids from these
expandable UB organoids that recapitulate the adult kidney collecting duct system. These
platforms will facilitate an enhanced understanding of development, regeneration and diseases
of the mammalian collecting duct system.
1
Introduction

Kidney function and development
Kidney is the critical organ that filters the blood and remove wastes from our body. The
human kidneys filter all the blood in our body every 30 minutes and adjust water, electrolytes, and
pH to maintain the homeostasis of body fluid. Human kidney contains roughly one million
nephrons all connected to a highly branched collecting duct (CD) system (Fig. 1). Nephrons filter
and process the blood to form the primitive urine, which is collected in the CD system and further
refined through various water and ion channels
1,2
. The refined urine was then drained into the
bladder through the ureter.

Human kidney development starts around gestation week five, when anterior intermediate
mesoderm-derived nephric duct (ND) invades into the posterior intermediate mesoderm-derived
metanephric mesenchyme (MM) and becomes the ureteric bud (UB)
3
(Fig. 2a). Within this
Fig. 1 Diagram showing the human kidney structure. Image adapted from the PKD Foundation.
2
metanephric kidney, the area where the tip of the UB is capped by MM is called the nephrogenic
niche (Fig. 2b). And within each of these niches, signals from the MM, including Wnt, FGF, GDNF,
Retinoic Acid, etc., induce the repeated branching of UB which eventually gives rise to the entire
CD network
4
(Fig. 2c). During this process, the ureteric progenitor cells (UPCs) reside on the tip-

-most region constantly divide and proliferate to maintain the self-renewing UPC pool on the tip,
while also migrate out of the tip niche and start maturing into the UB stalk cell fate as they move
away from the self-renewal signals such as GDNF, and eventually differentiate towards the
mature CD cell types
5
. In the renal CD tubules, there are mainly two different cell types, the more
abundant principal cells (PCs) that concentrate the urine and regulate Na
+
/K
+
homeostasis via
water and sodium/potassium transporters, and the less abundant α- and β-intercalated cells (ICs)
regulate normal acid-base homeostasis via secretion of H+ or HCO3− into the urine. During the
UB to CD differentiation, Notch signaling pathway plays an critical role in determining the cell fate
between PC (Notch on) and IC (Notch off) through juxtacrine regulation and therefore create a
salt-and-pepper patterning where fewer ICs are dispersed within the surrounding more abundant
PCs
6,7
(Fig. 2d). On the other hand, signals from the UB maintain the self-renewal of the nephron
Fig. 2 Diagrams showing the different perspectives of kidney development. a Diagram showing the initiation
of metanephric kidney development and UB/CD branching morphogenesis. Image adapted from Costantini and
Kopan, Developmental Cell, 2010
1
. b Diagram showing the different progenitor cell populations in the nephrogenic
niche. Image adapted from McMahon, Curr Top Dev Biol, 2016
7
. c Diagram showing the signals within the
nephrogenic niche. Image adapted from Little et al., Curr Top Dev Biol, 2010
4
. d Notch signaling determines the
CD cell fate between PC and IC. Image adapted from Awqati and Gao, Physiology, 2011
7
.
3
progenitor cells (NPCs) within the MM while also induce it’s differentiation to form different
segments of nephrons, from glomerulus at the proximal end that directly filters the blood, to the
distal convoluted tubule and connecting tubule at the distal end that connects immediately to the
UB/CD
4,8
. Beside these two major progenitor populations, other important cell types that take part
in kidney organogenesis include interstitial progenitor cells (IPCs) and vasculature progenitor cells
(VPCs) that will form kidney stroma and kidney vasculature, respectively (Fig. 2b).
Given this central role of the UB/CD in kidney organogenesis and functionality, defects in
UB/CD development often lead to malformation of the kidney, low endowment of nephrons at
birth, and congenital anomalies of kidney and urinary tract (CAKUT)
2,8,9
. Thus, a better
understanding of kidney branching morphogenesis is needed for in vitro efforts toward rebuilding
the kidney. It is also required for developing novel preventive, diagnostic, and therapeutic
approaches for various kidney diseases.

Current kidney organoid models and Knowledge gaps
Three-dimensional (3D) multicellular mini-organ structures, or organoids, have broad
applications for modeling organ development and disease, and for regenerating organs through
cell or tissue replacement therapies
10-12
(Fig. 3). Recently, we and others have been able to
generate functional kidney organoids from human pluripotent stem cells (hPSCs)
13-17
or from
expandable nephron progenitor cells
18-20
. These organoid models have greatly aided studies of
the role of nephrons in kidney development, as well as modeling kidney injury and diseases such
as the Autosomal Dominant Polycystic Kidney Disease (ADPKD), Autosomal Recessive
Polycystic Kidney Disease (ARPKD)
17,21-23
. However, despite these efforts toward the expansion
or de novo generation of the NPCs and nephrons, the generation and expansion of the UB/CD
lineage cell types has been unsatisfactory. The existing culture systems or differentiation
protocols for generating UB/CD from primary mouse/rat tissue
24-28
, mouse embryonic stem cells
29-
31
, or hPSCs
31-38
can neither capture and expand the ureteric progenitor cells for long-term, nor
4

fully differentiate the UBs into mature CDs with patterned and functional PCs and ICs. Therefore,
we still lack a robust kidney organoid model that can generate and expand the UB progenitor cells
and recapitulate the maturation and spatial patterning of the adult CD (see Supplementary Tables
1 and 2 for a side-by-side comparison of this study and published literature).

Dissertation aims and key findings
Here in this dissertation, I report the development of a UB/CD organoid system that will
facilitate an enhanced understanding of development, regeneration and diseases of the
mammalian collecting duct system. In Chapter 1, I report the generation of expandable, 3D
branching ureteric bud (UB) organoid that can be derived from primary UB progenitors from
mouse and human fetal kidneys, or generated de novo from human pluripotent stem cells. This
study has been published in Nature Communications in 2021
39
. In this study, I also show proof-
Fig. 3 Schematic showing the broad application of organoids for studying development and diseases.
Image adapted from Fatehullah et al., Nature Cell Biology, 2016
12
.
5
of-concept experiments using these models for disease modeling and generating engineered
kidneys. In Chapter 2, I report the generation of mouse and human CD organoids (mouse CD
organoid and human CD organoid version 1 published in Zeng et al., 2021
39
, manuscript for
human CD organoid version 2 is in preparation) from these expandable UB organoids that
recapitulate the adult kidney collecting duct system.

6
Chapter 1. Generation of expandable ureteric bud organoids from primary cells or
human pluripotent stem cells

Abstract
Current kidney organoids model development and diseases of the nephron but not the contiguous
epithelial network of the kidney’s ureteric bud (UB) or collecting duct (CD) system. Here, we report
the generation of an expandable, 3D branching ureteric bud (UB) organoid culture model that can
be derived from primary UB progenitors from mouse and human fetal kidneys, or generated de
novo from human pluripotent stem cells. Aggregating 3D-cultured nephron progenitor cells with
UB organoids in vitro results in a reiterative process of branching morphogenesis and nephron
induction, similar to kidney development. Applying an efficient gene editing strategy to remove
RET activity, we demonstrate genetically modified UB organoids can model congenital anomalies
of kidney and urinary tract. Taken together, these platforms will facilitate an enhanced
understanding of development, regeneration, and diseases of the mammalian ureteric bud
epithelial network. This study has been published in Nature Communications in 2021
39
.

7
Chapter 1 Introduction
During kidney development, within the nephrogenic niche, signals from the MM induce the
repeated branching of UB which eventually gives rise to the entire CD network
4
. During this
process, the ureteric progenitor cells (UPCs) reside on the tip-most region constantly divide and
proliferate to maintain the self-renewing UPC pool on the tip, while also migrate out of the tip niche
and start maturing into the UB stalk cell fate as they move away from the self-renewal signals,
and eventually differentiate towards the mature CD cell types
5
. The signals involved in the UPC
self-renewal have been studied in the past few decades by different groups (reviewed by
Costantini, 2012)
2
. These signals secreted from the NPCs and IPCs of the MM include GDNF
signaling, WNT signaling, FGF7, FGF10, vitamin A/retinoic acid, HGF, EGF, etc., in which the
GDNF signaling has been considered the most critical factor for maintaining UPC self-renewal.
It’s receptor, RET, and co-receptor GFRA1, are localized on the UPC surface. Mutations in either
GDNF or RET/GFRA1 can lead to kidney agenesis
40
. Downstream of GDNF-RET signaling,
among the three known downstream pathways, MEK/ERK, PI3-K, and PLC , it was suggested
that UPC self-renewal is mainly maintained through the PI3-K pathway and further the activation
of two key target genes, Etv4 and Etv5, to promoting other downstream target genes that’s directly
involved in UPC proliferation, cell migration, and ECM degradation
2,41
. Other target genes include
Ret itself, and Wnt11 which promotes GDNF expression in MM, forming two positive feedback
loops. Currently, detail mechanism of how GDNF-RET and other signals regulating those target
genes through PI3-K mediated Etv4/Etv5 expression remains unclear.
Recent attempt for in vitro culture and expansion of UB has been reported but with limited
success
28
. Using this system, mouse UB was only able to be cultured in vitro for maximum 9 days
and cannot be passaged. And there is no culture/expansion system for the human UPC/UB. On
the other hand, attempts to generate mouse or human UPCs from PSCs have been reported from
many groups
31,33,35,37,38
. However, although some of these UB organoids showed branching
8
morphogenesis, they still lack the expansion ability and the ability to generate mature CD cell
types. Therefore, a robust system to generate high-quality UB/UPCs then capture and expand
these progenitor cells will be needed for the generation of high-quality CD organoid, as well as
for the studying of kidney development and disease.
Even though these previous studies mentioned above were not very successful, they
serve as a good foundation together with the previous developmental studies for us to develop a
better synthetic niche for the capture and expansion of UPCs. Here in this chapter, I report the
establishment of both the mouse and human UB culture systems that can capture and expand
mouse and human UPCs in the form of 3D organoids. I also developed a directed differentiation
protocol to generate high-quality human UPCs de novo from hPSCs.

9
Results

Expanding mouse UB progenitor cells into 3D branching UB organoids.
We previously developed a 3D culture system for the long-term expansion of mouse and
human NPCs, which can generate nephron organoids that recapitulate kidney development and
disease
20,42
. UB branching morphogenesis is driven by another kidney progenitor population, the
UPCs. UPCs are specified around embryonic Day 10.5 (E10.5), when the UB starts to invade the
MM. UPCs disappear around postnatal Day 2 (P2), when nephrogenesis ceases. Self-renewing
UPCs reside in the tip region of the branching UB. During their ~10-day lifespan, some UPCs
migrate out of UB tip niche to the UB trunk, and differentiate into the renal CD network. Other
UPCs proliferate and replenish the self-renewing progenitor cell population of the UB tip. Ret
43,44

and Wnt11
40
have been identified as specific markers for UPCs and regulate UPC programs
directly (Ret) or through feedback mechanisms (Wnt11). A transgenic reporter mouse strain
Wnt11-myrTagRFP-IRES-CE (“Wnt11-RFP” for short) facilitates the real-time tracking of Wnt11-
expresing cells based on RFP expression, and the lineage tracing of their progeny via a Cre-
mediated recombination system
45,46
.
We employed this Wnt11-RFP reporter system as a readout to screen for a culture
condition that maintained the progenitor identity of UPCs in vitro. T-shaped UBs were manually
isolated from E11.5 kidneys of Wnt11-RFP mice, and immediately embedded into Matrigel to set
up a 3D culture platform that supported epithelial branching. In this 3D culture format, built on
prior efforts toward the ex vivo culture of UB
24-28
, starting from the medium components used by
Yuri et al.
28
, hundreds of different combinations of growth factors and small molecules were tested,
following strategies similar to those we had used to establish optimal NPC culture
20

(Supplementary Fig. 1a–l, see also the “Supplementary Methods” for details). This screening
allowed us to identify a cocktail, which we named “UB culture medium” (UBCM, Supplementary
Table 6), that maintained self-renewing UPCs as a 3D branching UB organoid (Fig. 1-1a). Under
this culture condition, the T-shaped UB formed a rapidly expanding branching epithelial
10
morphology. More importantly, in contrast to prior UB culture system that generated a mixture of
both UB tip and trunk cell types
28
, uniform Wnt11-RFP expression was maintained throughout the
3D structure in the UBCM-derived UB organoid, suggesting the capture of a relatively pure UPC
population (Fig. 1-1b). Resected Wnt11-RFP+ UB organoid tips, re-embedded in Matrigel,
branched and grew into additional Wnt11-RFP+ UB organoids. Repetitive passaging and
embedding for up to 3 weeks, resulted in more than a hundred thousand-fold expansion in the
number of cells (Fig. 1-1c). Wnt11-RFP levels remained uniform for the first 10 days but
progressively dropped thereafter, similar to the normal time course of UPCs in vivo. Consistent
with the uniform expression of Wnt11-RFP throughout UB organoids at 10 days, whole-mount
immunostaining confirmed the homogenous expression of the critical UPC regulators Ret, Etv5,
and Sox9, as well as broad UB lineage markers Gata3, Pax2, Krt8, and Cdh1 (Fig. 1-1d, e and
Supplementary Fig. 2a, b).
To better define the identity of the UB organoids, we used RNA-seq to profile the
transcriptome of the organoids after 5, 10, and 20 days in culture. These data were compared
with prior RNA-seq data for primary UB tip and UB trunk populations
46
, as well as for NPCs and
interstitial progenitor cells (IPCs)
47
. Unsupervised clustering (Supplementary Fig. 2c) and principal
component analysis (PCA) (Fig. 1-1f) placed the cultured UB organoids closer to the primary UB
tip samples than to differentiated stalk derivatives of the UB trunk. Taken together, these findings
indicate that the UB organoid culture system enables a substantial expansion of cells retaining
molecular characteristics of UPCs in vitro.
Next, we tested whether the UB organoid culture system could be applied to mouse strains
other than Wnt11-RFP. For this, we successfully derived UB organoids from E11.5 UB from
various other mouse strains, such as Sox9-GFP
48
and Rosa26-Cas9/GFP
49
(Supplementary Fig.
2d, e). All of these UB organoids retained the typical branching morphology and showed very
similar growth rates, compared to Wnt11-RFP UB organoids (Fig. 1-1g), indicating the robustness
of the 3D/UBCM culture system. Importantly, UB organoids self-organized into branching
11
organoids after a freeze-thaw cycle, enabling cryostorage and reseeding of UB cultures
(Supplementary Fig. 2f).

Fig. 1-1 Expanding mouse UB progenitor cells into 3D branching UB organoids. a Schematic of mouse UB
isolation and screening for UB organoid culture condition. b Representative bright field (BF, left panel) and Wnt11-
RFP (right panel) images of UB organoid. Scale bars, 200 μm. c Cumulative growth curve of UB organoid culture
starting from 2000 cells. Each time point represents three biological replicates. d Whole-mount immunostaining of
UB organoid for various UB markers at Day 10 of culture. The four panels on the right represent the boxed region
in the left panel. Scale bars, 100 μm. e Quantification of percentages of UB cells stained positive for different UB
markers in Fig. 1d and Supplementary Fig. 2a, b. Each column represents counts from three different fields of view
(n = 3). f Principal component analysis (PCA) of RNA-seq data. Different colors and oval circles represent different
primary kidney cell populations (NPC, IPC, UB tip, and UB trunk) or UB organoids cultured for 5 (D5), 10 (D10),
and 20 days (D20). g Summary of UB organoid derivation from mouse strains with different genetic backgrounds
or with different derivation methods. h Bright field (BF) images showing single UB cells, derived from Wnt11-RFP
mice, cultured in the UBCM on Days 1 (left panel) and 5 (middle panel), as well as Wnt11-RFP image on Day 5
(right panel). Scale bars, 200 μm. i UB organoid derivation from a single UB cell-derived budding structure isolated
from the boxed region in (h) at Days 0 (D0, the day of isolation and re-embedding into Matrigel), 2 (D2), 4 (D4), and
6 (D6). All images in (i) have been scaled to share the same scale bar with the D6 image. Scale bar, 400 μm. All
data are presented as mean ± s.d. Source data are provided as a Source Data file.

12
To determine whether UBCM culture conditions enabled clonal growth from a single UPC,
dissociated E11.5 UBs were embedded at clonal density in Matrigel and cultured in UBCM
medium. Around 30% of the single cells self-organized into E11.5 UB-like budding structures
within 5 days, though a smaller percentage (3–5%) maintained Wnt11-RFP (Fig. 1-1h, and
Supplementary Fig. 2g, h), an efficiency similar to clonal organoid formation for Lgr5+ intestinal
stem cells
50
. Importantly, the clonally derived Wnt11-RFP+ budding structures were identical to
intact E11.5 UB-derived organoids in both branching morphology and growth rate (Fig. 1-1g–i).
Furthermore, with-drawal of the major medium components from UBCM resulted in either growth
arrest (CHIR99021 and GDNF) or rapid loss of Wnt11-RFP (all components except for
CHIR99021), suggesting that each component was essential for optimal UB organoid culture
(Supplementary Fig. 1l). These data, taken together, suggest that UBCM represents a synthetic
niche for the in vitro expansion of UPCs.

Generating engineered mouse kidney from expandable NPCs and UBs.
The availability of expandable NPCs
20,42
and UPCs provides the scalable building blocks
required for making a kidney. As a proof-of-concept, we examined whether combining these cell
types could generate a model mimicking key features of in vivo kidney development, such as
reiterative ureteric branching and nephron induction, and morphogenesis and patterning of
differentiating derivatives (Fig. 1-2a).
NPCs in our long-term culture model grow as 3D aggregates. To mimic the natural
organization of NPCs capping UB tips in the kidney anlagen, we manually excavated a cavity in
3D cultured NPCs (expanded several billion-fold over for 6–12 months of culture) and inserted a
cultured UB organoid tip. The engineered kidney structures were transferred onto an air–liquid
interface to facilitate further kidney organogenesis. Over 7 days of culture, the inserted Hoxb7-
Venus UB organoid tip underwent extensive branching (Fig. 1-2b) generating a tubular network
13
extending from the center of the structure to the periphery. Further, NPCs generated nephron-like
cell types including PODXL+/WT1+ podocytes and LTL+ proximal tubules (Fig. 1-2c).
To determine whether the engineered kidney also formed a connection between nephron
and CD, we engineered kidneys comprising Hoxb7-Venus UB and wild-type NPCs. In this way,
all progeny of the UB organoid could be tracked by Venus expression. Co-staining of the
engineered kidney structure with CDH1 and GATA3-specific antibodies identified a clear fusion
of CDH1+/Venus− distal nephron with CDH1+/Venus+ CD. Importantly, GATA3 expression was
strong in the entire Venus + CD structure, but progressively dropped along the distal-to-proximal

Fig. 1-2 Generating engineered mouse kidney from expandable NPCs and UBs and gene editing of the
mouse UB organoid. a Schematic of the engineered kidney reconstruction and organotypic culture procedure. b
Time course images (bright field and Hoxb7-Venus) showing the branching morphogenesis of the engineered
kidney reconstructed at air–liquid interface at Days 2, 4, 5, and 7. Scale bars, 200 μm. c Immunostaining of the
engineered kidney (Day 7) constructed from Wnt11-RFP UB organoid and wild-type NPCs for UB/CD marker KRT8,
nephron marker PODXL and WT1 (podocytes) and LTL (proximal tubule). Note that both KRT8 and PODXL were
stained green. The round structures that co-stain with WT1 are podocytes of the nephron. UB-derived structures
do not co-stain with WT1. Scale bar, 100 μm. d Immunostaining of the engineered kidney constructed from Hoxb7-
Venus UB organoid and wild-type NPC (Day 10) for GATA3 and CDH1. Scale bars, 50 μm. e Summary of
engineered kidney generation experiments. f Schematic of gene overexpression and gene knockout procedures in
the UB organoid. OE overexpression, KO knockout. g Fluorescence image of GFP overexpression (GFP OE) in
wild-type UB organoid. Scale bar, 200 μm. h Knockout of GFP in Rosa26-Cas9/GFP UB organoid using multiplexed
sgRNAs (“GFP KO,” right panels) targeting the coding sequence of GFP. Multiplexed non-targeting sgRNAs were
introduced to the organoid as control (“Ctrl KO,” left panel). Note the gene-edited single cells self-organized into
typical branching organoid morphology. Scale bars, 400 μm.

14
axis of the distal nephron, as observed in vivo
51,52
(Fig. 1-2d). Thus, the engineered kidney
established a luminal interconnection between the nephron and CD, an essential morphological
event for kidney function. Engineered kidney development was robust: ~80% of engineered
kidneys underwent a similar developmental program, with most failures likely reflecting technical
issues in manual construction (Fig. 1-2e). Taken together, engineered kidney with interconnected
nephron and CD can be efficiently generated from expandable NPCs and UBs.

Performing efficient gene editing in the expandable UB organoid.
The UB model could provide an accessible in vitro complement to the mouse models for
in-depth mechanistic studies and drug screening. Here, efficient gene overexpression (OE) or
gene knockout (KO) would significantly extend the capability and utility of the in vitro model (Fig.
1-2f). As a proof-of-concept, GFP OE and GFP KO UB organoids were generated. For GFP OE,
we used a standard lentiviral system to introduce GFP under the control of a CMV promoter
53
.
However, even at a very high titer, the lentiviral infection efficiency of the intact T-shaped UB was
low. However, dissociating T-shaped UB or UB organoids into a single-cell suspension prior to
infection dramatically improved the infection efficiency. Widespread GFP activity was observed in
resulting UB organoids after reaggregation of infected cells (Fig. 1-2g). To test KO, we targeted
GFP in Rosa26-Cas9/GFP UB organoid, in which Cas9 and GFP are constitutively expressed
from the Rosa26 loci
49
(Supplementary Fig. 2e). A mix of three different lentiviral constructs,
expressing three different single-guide RNAs (sgRNAs) with Cas9 targeting sites 100–150 bp
apart
54
, gave a highly efficient, GFP sgRNA-specific, multiplexed CRISPR/Cas9 KO,
demonstrating effective KO in UB organoid cultures (Fig. 1-2h).

Generating human UB organoids from primary human UPCs (hUPCs).
The successful generation of mouse UB organoids prompted us to test whether the
system can also derive human UB organoids. To achieve this, we developed a method to
15
generate expandable human UB organoids from primary human UPCs (hUPCs) (Fig. 1-3a).
Similar to their murine counterparts, hUPCs within UB tips, express RET
51,52,55
and WNT11
(GUDMAP/RBK Resources, https://www.gudmap.org). Using an anti-RET antibody raised against
the extracellular domain of RET that recognizes RET+ hUPCs (Fig. 1-3b), we performed FACS
enrichment of RET+ cells from human kidneys between 9 and 13 weeks of gestational age and
examined their growth in modified UBCM conditions. A robust human UBCM (hUBCM,
Supplementary Table 7) culture condition was identified that sustained the long-term expansion
(an estimated 108–109 fold expansion over 70 days) as branching UB organoids (Fig. 1-3c). UPC
marker gene expression was maintained in human UB cultures at level comparable to that of the
human fetal kidney (Fig. 1-3d). Thus, we provide the proof-of-concept that expandable human UB
organoid can be derived from purified RET+ primary human UPCs.

Generating iUB organoids from hPSCs.
To determine whether UB organoids could be generated from hPSC-derived UPCs, we
first genetically engineered H1 human embryonic stem cells (hESCs) with a knockin dual-reporter
system where mCherry was expressed from the PAX2 locus (PAX2-mCherry) and GFP from the
WNT11 locus (WNT11-GFP) (Supplementary Fig. 4a-e and Supplementary Methods). Using this
reporter line, we first tested whether PAX2+/WNT11+ hUPCs can be generated following
previously reported directed differentiation protocols that generated UB-like cells
31-34
. After
directed differentiation, we confirmed the expression of PAX2-mCherry, but failed to observe the
expression of WNT11-GFP, suggesting that the differentiation efficiency to generate hUPCs was
relatively low following existing protocols. Relying on hUBCM’s role in de novo hUPC induction
and stabilization and a modified differentiation protocol, we were able to establish a stepwise
protocol that resulted in high-quality hUPC cultures, which generated branching UB (iUB)
organoids (Fig. 1-3e and Supplementary Table 9).
16

The UB is derived from the nephric duct (ND), which originates from primitive streak
(mesendoderm (ME))-derived anterior intermediate mesoderm
2,8
. Consistent with this
developmental trajectory, following a 7-day directed differentiation, we were able to first observe
the expression of ME marker T on Day 3 of differentiation in most cells (Supplementary Fig. 4f),
followed by the formation of large numbers of compact cell colonies that are
GATA3+/SOX9+/PAX2+/PAX8+/KIT+/KRT8+ on Day 7 of differentiation, suggesting the
generation of potential precursor cells of the UB lineage (Supplementary Fig. 4g). Consistent with
the immunostaining results, we were able to identify a PAX2-mCherry+ population (13.1%) by
Fig. 1-3 Generating human UB organoids from primary human UPCs and dual-reporter human pluripotent
stem cells. a Schematic of the purification of primary human UPCs from the nephrogenic zone (illustrated as boxed
region) of the human fetal kidney (9–13 weeks of gestational age) and the derivation of human UB organoid. b
Immunostaining of the human fetal kidney nephrogenic zone for UB tip marker RET (red), broad UB lineage marker
KRT8 (green), and NPC marker SIX2 (cyan). Scale bars, 50 μm. c Time course bright field images showing the
growth of human UB organoid derived from primary human UPCs in a typical passage cycle at Days 1 (D1), 5 (D5),
and 9 (D9). Scale bar, 200 μm. All images in (c) have been scaled to share the same scale bar with the D9 image.
d qRT-PCR analyses of human UB organoid (cultured for 54 days) derived from primary human UPCs for various
UB markers as indicated. Human fetal kidney from 11.2-week (11.2 weeks) gestational age was used as control. e
Schematic of the stepwise differentiation from WNT11-GFP/PAX2-mCherry dual-reporter hESC line into iUB
organoid. (TeSR mTeSR1 medium, Y Y27632, ME mesendoderm stage medium, UB-I UB Stage I medium, UB-II
UB Stage II medium). f qRT-PCR analyses of the FACS purified mCherry+ cells (orange) and the iUB organoid
(blue, cultured for 50 days) for various UB markers as indicated. Undifferentiated H1 hESCs (gray) and human fetal
kidney (green, 11.2-week gestational age) were used as controls. g Whole-mount immunostaining of the
expandable iUB organoid for UB markers SOX9 and CDH1. Scale bars, 50 μm. h Summary of human UB organoid
derivation from different sources and their expansion in vitro. ***, this is the culture time we achieved before our lab
shutdown due to coronavirus outbreak. The maximum organoid culture time and expansion could be much greater.
All data are presented as mean ± s.d. In d and f, the significance was determined by two-tailed unpaired Student’s
t tests; NS not significant; n = 3. Source data are provided as a Source Data file.
17
FACS on Day 7. However, at this stage, the PAX2-mCherry+/WNT11-GFP+ population was very
rare (0.4%), preventing further characterization or culture (Supplementary Fig. 4h and 8b).
However, further culture of PAX2-mCherry+ cells in the 3D/hUBCM culture conditions activated
WNT11-GFP reporter expression at around 3 weeks, and the structure started to show a
branching morphology (Supplementary Fig. 4i). We refer to the PAX2-mCherry +/WNT11-GFP+
branching structure an “iUB” organoid hereafter. Importantly, these iUB organoids could be
expanded stably in 3D/hUBCM for at least 2 months without losing reporter gene expression (Fig.
1-3h). Consistently, qRT-PCR analysis confirmed that WNT11 expression was low in the
mCherry+ cells purified from FACS, but was dramatically elevated in the iUB organoid.
Furthermore, even though UB marker genes PAX2, GATA3, LHX1, and RET were greatly
elevated on Day 7 of differentiation, while WNT11, CDH1, EMX2, and HNF1B, showed
comparable levels of expression to the human fetal kidney only after extended hUBCM culture,
suggesting that hUBCM promoted transition from a common ND to a specific ureteric epithelial
precursor (Fig. 1-3f). In addition to qRT-PCR, expression of marker genes SOX9 and CDH1 in
the iUB organoid was detected at the protein level by immunostaining (Fig. 1-3g), further
confirming the identity of the iUB organoid.
To test whether expandable iUB organoids could be generated from human-induced
pluripotent stem cells (hiPSCs), we employed SOX9-GFP hiPSC
56
for differentiation and purified
the SOX9-GFP+ UB precursor cells on Day 7 of differentiation (Supplementary Figs. 4j and 8c).
Similar to hESC-derived iUBs, following an extended culture in hUBCM, we were able to derive
SOX9-GFP iUB organoids that expanded stably with retained SOX9-GFP expression throughout
(Fig. 1-3h and Supplementary Fig. 4k–m). Taken together, these results support the conclusion
that expandable iUB organoids can be derived from hESC and hiPSC lines.

18
Generating iUB organoids independent of reporter hPSC lines.
The reporter hPSC lines are useful in developing iUB differentiation protocols, but if iUB
organoids can only be derived from these reporter hPSCs, its applications will be significantly
limited. To solve this problem, we next developed a method to derive iUB organoid from any given
hPSC line in the absence of reporter (Fig. 1-4a). In this method, after 7 days of differentiation,
sorting of KIT+ cells was used to enrich the precursor population, rather than sorting based on
PAX2-mCherry or SOX9-GFP reporters. With further refinement of our stepwise iUB
differentiation protocol and a refined hUBCM (hUBCM-v2, Supplementary Tables 7 and 9), long-
term expandable branching iUB organoids can be derived within 12 days from hPSCs with high
efficiency. Following this protocol, as proof-of-concept, we have successfully derived iUB
organoids from three different hPSC lines, including two hESC lines (Fig. 1-4 and Supplementary
Fig. 5) and one hiPSC line (Supplementary Fig. 6).
KIT was previously reported as a surface marker that can be used to enrich UB-like cells
upon hPSC differentiation
31
. Interestingly, we noticed that KIT+ cells are frequently co-stained
with PAX2 (Supplementary Fig. 4g). We thus hypothesized that FACS of KIT+ cells will enrich for
PAX2+ precursor cells similar to the sorting of PAX2-mCherry+ cells using our PAX2-mCherry
reporter line. Starting from our WNT11-GFP/ PAX2-mCherry hESC line, after 7 days of
differentiation, 36.1% of the cells were KIT+ (Supplementary Fig. 5a and 8d). Further culture of
these KIT+ cells in hUBCM-v2 showed a much faster induction of WNT11-GFP expression than
using hUBCM (5–7 days with hUBCM-v2, as shown in Supplementary Fig. 5b, vs. ~3 weeks with
hUBCM as shown in Supplementary Fig. 4i). Importantly, accompanying the expression of
WNT11-GFP, the organoid started to show the typical branching morphology (Supplementary Fig.
5b), and can since be stably passaged and expanded billions of folds either manually or as single
cells for at least 70 days by the time our manuscript is submitted (Supplementary Fig. 5c, h).
To determine whether gene expression in the iUB organoid is also stably maintained over
long-term culture, we collected iUB organoids 33, 49, and 66 days after the initiation of culture in
19

Fig. 1-4 Generating human iUB organoids from human pluripotent stem cells independent of reporters. a
Schematic of the stepwise differentiation from any hPSC line into iUB organoid without using reporter. (TeSR
mTeSR1 medium, CR CloneR, ME (v2) mesendoderm stage medium version 2, UB-I UB Stage I medium, UB-II
UB Stage II medium). b –h Characterizations of iUB or iCD organoids derived from the WNT11-GFP/PAX2-mCherry
hESC line independent of its reporters. b qRT-PCR analyses of the FACS purified KIT+ precursor (orange), KIT+
precursor-derived iUB organoids cultured for 33 (D33, gray), 49 (D49, yellow), and 66 days (D66, light blue) for
various UB markers as indicated. Undifferentiated H1 hESCs (dark blue) and human fetal kidney (green, 11.2-week
gestational age) were used as controls. c –e Whole-mount immunostaining of the expandable iUB organoid for
various UB markers as indicated. The four panels on the right represent the boxed region in the left panel. Scale
bars, left panel, 100 μm; right four panels, 40 μm. f Quantification of percentages of iUB cells stained positive for
different UB markers in Fig. 1-4 c–e. Each column represents counts from three different fields of view (n = 3).
20

hUBCM-v2, and compared their gene expression by qRT-PCR with undifferentiated hPSCs, KIT+
precursor cells, and human fetal kidney tissue. Consistent with our previous finding with sorted
PAX2-mCherry+ precursors (Fig. 1-3f), the sorted KIT+ precursor cells also showed strong
expression of PAX2, GATA3, LHX1, and RET, but the expression of WNT11, CDH1, EMX2, and
HNF1B was only induced after further programming in the presence of hUBCM-v2 (Fig. 1-4b).
Importantly, gene expression of all these markers in the iUB organoids were maintained stably
throughout the culture period, at levels comparable or higher than that of human fetal kidney
tissue, indicating the robustness of our method. Whole-mount immunostaining (Fig. 1-4c–e) or
section staining (Supplementary Fig. 5d–f) of the iUB organoids for various general UB lineage
markers (KRT8, PAX2, PAX8, GATA3, and CDH1) or UB tip markers (RET and SOX9) further
confirmed the expression of all these genes at the protein level. Importantly, quantification of the
immunostaining results indicated that more than 95% of cells in the organoids showed
homogeneous expression of all markers (Fig. 1-4f), suggesting that the majority of the cells in the
iUB organoid are of the UB progenitor identity.
Importantly, similar iUB organoids were also generated from a second hESC line (H1) and
from the SOX9-GFP reporter hiPSC line. After 7 days of differentiation, 55.4% of H1 cells and
43.9% of SOX9-GFP hiPSCs were identified as KIT+ on FACS (Fig. 1-4g and Supplementary
Figs. 6a, 8a, e). Further culturing of these KIT+ cells in hUBCM-v2 derived iUB that can be stably
expanded as branching iUB organoids (Fig. 1-4h and Supplementary Fig. 6b, c). qRT-PCR further
confirmed that gene expression in the H1 hESC-derived iUB organoid is similar to the iUB
organoid derived from our dual-reporter hESC line shown above (Fig. 1-4i). Similarly, qRT-PCR
(Supplementary Fig. 6d), whole-mount immunostaining (Supplementary Fig. 6e–h), and section
g Flow cytometry analysis of KIT+ precursor cells differentiated from wild-type H1 hESC. h Bright field image of a
typical iUB derived from wild-type H1 hESC. Scale bar, 250 μm. i qRT-PCR analyses of the iUB organoids derived
from the dual-reporter hESC line (W/P reporter, orange, cultured for 33 days) or wild-type H1 hESC line (H1, gray,
cultured for 30 days) for various UB markers. Undifferentiated H1 hESCs (blue) and human fetal kidney (yellow,
11.2-week gestational age) were used as controls. All data are presented as mean ± s.d. In b and i, the significance
was determined by two-tailed unpaired Student’s t tests; n = 3. Source data are provided as a Source Data file.
21
staining (Supplementary Fig. 6i–l) confirmed that the majority of the cells in the hiPSC-derived
iUB organoids are of the UB progenitor identity.

Modeling kidney development and disease using mouse and human UB organoids.
GDNF is a critical signal in both mouse and human UB culture. In vivo, GDNF secreted
by MM cells surrounding UPC-containing branch tips signals via RET, with its co-receptor GFRA1,
to maintain the UPC state and stimulate UB branching morphogenesis
2,8
. Loss of the activity of
these genes results in a CAKUT syndome
2,8,9,57,58
. We employed CRISPR/Cas9 system to KO
Ret/RET in mouse and human UB organoids predicting as in vivo, UB organoid development in
vitro would be Ret/RET-dependent (Fig. 1-5a). UB organoids were infected with lentivirus
expressing Cas9 and two independent sgRNAs targeting Ret/RET (in lentiCRISPR-v2 vector),
while a control group received sgRNA without Cas9 (in lentiGuide-puro vector). As expected, the
control mouse UB organoids grew normally with maintained branching morphogenesis upon
lentiviral infection and puromycin selection, while both Ret KO UB organoids stopped branching
(Fig. 1-5b). Whole-mount immunostaining of control and Ret KO UB organoids confirmed a
dramatic reduction (more than 95%) of RET expression 6 days after lentiviral infection in the Ret
KO UB organoids, demonstrating the successful removal of Ret (Fig. 1-5c, d). Consistent with the
defect in branching morphogenesis in the Ret KO organoids, genes enriched in UB tip, Wnt11,
and Lhx1
46
were reduced dramatically, and the expression of common UB lineage markers Pax2
and Gata3 were also decreased (Fig. 1-5e). KO of RET in the human iUB organoid also resulted
in the arrest of branching morphogenesis in both of the RET KO organoids receiving two different
sgRNAs (Fig. 1-5f) and a loss of RET immunostaining in more than 95% of cells with both sgRNAs
(Fig. 1-5g, h and Supplementary Fig. 7a–f). Interestingly, even though WNT11 and GFRA1
expression was significantly reduced in both RET KO iUB organoids, the expression of LHX,
GATA3, PAX2, as well as ETV5 and SOX9, did not show consistent changes in both RET KO
22

Fig. 1-5 Modeling kidney development and disease using mouse and human UB organoids. a Schematic of
Ret/RET gene knockout procedures in the mouse or human UB organoid. b Bright field images showing the
branching morphogenesis of mouse UB organoids 2 days (Day 2) and 6 days (Day 6) after lentiviral infection. Scale
bars, 200 μm. c Whole-mount immunostaining of the control or Ret KO mouse UB organoids for UB markers GATA3
and RET after 6 days of lentiviral infection. Arrow heads indicate the few RET
+
cells in the Ret KO organoids. Lower
panels represent the boxed region in the upper panels. Scale bars, upper panels, 200 μm; lower panels, 40 μm.
23

organoids (Fig. 1-5i). These results suggest that species-specific regulatory network downstream
of Ret/RET might govern UB progenitor fate, consistent with previous observations of convergent
and divergent mechanisms of nephrogenesis between mouse and human
47,51
. Taken together,
we provide a proof-of-concept for recapitulating kidney development and disease using mouse
and human UB organoid models.
d Quantification of percentages of cells stained positive for RET in Fig. 1-5c. Each column represents counts from
three different fields of view (n = 3). e qRT-PCR analyses of the control mouse UB organoids (blue and orange)
and Ret KO mouse UB organoids (gray and yellow) for various UB markers 6 days after lentiviral infection. f Bright
field images showing the branching morphogenesis of human iUB organoids 3 (Day 3) and 12 days (Day 12) after
lentiviral infection. Scale bars, 200 μm. g Whole-mount immunostaining of the control or RET KO human iUB
organoids for UB markers PAX2 and RET 12 days after lentiviral infection. Arrow heads indicate the few RET
+
cells
in the RET KO organoids. Scale bars, 40 μm. See also Supplementary Fig. 7a–f for images of individual fluorescent
channels. h Quantification of percentages of human iUB cells stained positive for RET in Fig. 1-5g. Each column
represents counts from three different fields of view (n = 3). i qRT-PCR analyses of the control human iUB organoids
(blue and orange) and RET KO iUB organoids (gray and yellow) for various UB markers as indicated 12 days after
lentiviral infection. All data are presented as mean ± s.d. In d, e, h, i, the significance was determined by two-tailed
unpaired Student’s t tests; n = 3. Source data are provided as a Source Data file.
24
Chapter 1 Discussion
In this study, we report 3D culture models enabling the expansion and differentiation of
mouse and human UPCs. The organoid culture medium effectively replaces cell interactions
within the nephrogenic niche of the developing mammalian kidney, with a chemically defined
synthetic niche capable of maintaining UPC identity. Consistent with mouse genetics studies,
signaling pathways that play key roles in kidney branching morphogenesis, such as GDNF
44,59,60
,
FGF
61
, RA
62-64
, and Wnt
65,66
signaling, are also essential in maintaining UPC identity in UB
organoids. UPC cloning efficiency is not very high in the UBCM culture. Similar to our 3D NPC
culture
20,42
, it is likely that cell–cell contact is important for maintaining the best tip identity, as
aggregated UPCs, or manually passaged UB organoids as small cell clusters, can maintain
Wnt11-RFP homogeneously. Better understanding of cell–cell contact and/or potential additional
paracrine signals might help further improve the culture, thus allowing the development of a more
robust clonal expansion method.
An interesting observation during the de novo human UB directed differentiation process
was the induction of hUPC fate (as observed through the elevation of WNT11-GFP signal) by
hUBCM or hUBCM-v2 in the FACS purified PAX2-mCherry or KIT+ cells. One possibility to
explain this phenomenon is that these media could stabilize the rare and transient hUPC
population generated from directed differentiation. Another possibility is that these media could
promote cell fate transition from earlier WD-staged precursor cells (PAX2-mCherry+ and WNT11-
GFP-) to the hUPC fate (PAX2-mCherry+ and WNT11-GFP+). In support of both possibilities, our
NPC culture medium, NPSR, has recently been reported to facilitate the generation of NPC-like
cells in both directed differentiation
67
and transdifferentiation
68
settings. Future studies are
warranted to understand how hUBCM and hUBCM-v2 contributes to hUPC fate specification.
The generation of an engineered kidney from expandable mouse NPCs and mouse UPCs
provides a proof-of-concept for rebuilding a kidney in vitro from kidney-specific progenitor cells.
The availability of expandable NPCs and UPCs provides the scalable building blocks required for
25
making a kidney. The interaction between NPCs and UPCs is faithfully recapitulated, leading to
the autonomous differentiation into interconnected nephron and CD structures. Interestingly,
different from prior study
31
, in our engineered kidney system, IPCs appear to be dispensable in
reconstructing a branching kidney structure in vitro. It is likely that one of our engineered kidney
culture medium components, TTNPB, a small molecule analog of RA, substitutes RA production
by IPCs, an essential mechanism for proper UB branching and kidney development in vivo
62-64
.
Future efforts will require the integration of vascular progenitor cells, and a more in-depth
evaluation of IPCs, to develop engineered structures for testing in animal models of organ
transplantation.
Efficient genome editing in UB organoids opens up many applications using the UB
organoid platform and might be further extended to our UB organoid-derived CD organoid that
will be discussed in chapter 2. UB and CD organoids can be generated from available transgenic
mouse strains that bear genetic mutations related to kidney development and disease. In addition,
disease-relevant mutations can be introduced into the UB organoid directly, enabling the
investigation of pathophysiology throughout the entire course of kidney branching morphogenesis,
from the UB branching period to the mature CD stage. Our proof-of-concept Ret/RET KO
experiment demonstrated that our UB organoid system can recapitulate genetic malfunction of
branching morphogenesis in vitro as that of in vivo. Our results also shed light on the potential
different regulatory mechanisms downstream of Ret/RET in mouse and human. Our system offers
a unique platform to further investigate how human RET mutations identified in the human CAKUT
patients might contribute to congenital kidney malformation. The ability to produce large quantities
of UB organoids also provides a platform for drug screening. In conclusion, the UB organoid
system provides a powerful tool for studying kidney development, modeling kidney disease,
discovering drugs and, ultimately, regenerating the kidney.

26
Chapter 2. Generation of patterned kidney organoids that recapitulate the adult
kidney collecting duct system from expandable ureteric bud progenitors

Abstract
In recent years, protocols generating collecting duct-like cells from primary mouse/human
tissues or de novo from human pluripotent stem cells have been reported but with limited quality,
maturity, and functionality, fail to recapitulate the authentic adult human cortical/outer medullary
collecting duct. Here in this chapter, I report the generation of high-quality, mature, and patterned
mouse/human CD organoids derived from our expandable mouse/human UB organoids. These
CD organoids express some of the more mature markers of different postnatal/adult CD cell types.
And more importantly, they show similar pattern of PC and IC distribution to the adult cortical/outer
medullary collecting duct, demonstrating the robustness of our organoid model. Further
characterization and functional studies of these organoids are currently on going. This will provide
a powerful platform for the study of development and diseases of the mammalian collecting duct
system. The mouse CD organoid and human CD organoid version 1 study have been published
in Nature Communications in 2021
39
, manuscript for human CD organoid version 2 is currently in
preparation.

27
Chapter 2 Introduction
During the kidney branching morphogenesis, the ureteric progenitor cells (UPCs) reside
on the tip-most region constantly divide and proliferate to maintain the self-renewing UPC pool on
the tip, while also migrate out of the tip niche and start maturing into the UB stalk cell fate as they
move away from the self-renewal signals such as GDNF, and eventually differentiate towards the
mature CD cell types
5
. In the renal CD tubules, there are mainly two different cell types, the more
abundant principal cells (PCs) that concentrate the urine and regulate Na
+
/K
+
homeostasis via
water and sodium/potassium transporters, and the less abundant intercalated cells (ICs) that
regulate acid-base homeostasis via secretion of H+ or HCO3− into the urine. During the UB to
CD differentiation, Notch signaling pathway plays an critical role in determining the cell fate
between PC (Notch on) and IC (Notch off) through juxtacrine regulation from the Notch ligand
Jag1 on the IC membrane surface to promote its surrounding cells to adopt PC cell-fate and
therefore create a salt-and-pepper patterning where fewer ICs are dispersed within the
surrounding more abundant PCs
6,7
. Further, there are two major subtypes of ICs from the UB/CD
lineage, type-A and type-B. Type-A IC secretes acid/proton through its apical proton pump V-
ATPase, and is expressing type-A specific Cl
-
/HCO 3
-
anion exchanger, AE1 (gene name Slc4a1),
on the basolateral membrane of the cell. Type-B IC secretes bicarbonate through its apical type-
B specific Cl
-
/HCO 3
-
anion exchanger, PENDRIN (gene name Slc26a4), and its proton pump V-
ATPase is on the basolateral side. How the type-A and type-B ICs are specified is currently
unclear.
Across different regions of the kidney, collecting duct tubules have different compositions
of PC/type-A IC/type-B IC. In the outer region of the adult mouse or rat kidney, the cortex and
outer medulla, 60%-65% of the cells in the CD are PCs and the rest 35%-40% are ICs; and in the
inner medulla, only 10% of the CD cells are ICs and the rest 90% are PCs
69
. More specifically, in
adult mouse kidney, type-B ICs were mostly only found in the cortical and outer medullary CD,
where type-A ICs are presented across the entire CD tubule
70
.
28
Previously, many groups reported different protocols generating human CD-like cells from
hPSCs-derived human UPCs/UB organoids
33,35-38
. Nevertheless, as also mentioned in chapter 1,
none of these protocols can generate high-quality and expandable human UPCs, or mature and
patterned CD organoids with both PCs and ICs, showing the limitation of these existing protocols.
Here I report the generation of high-quality, mature, and patterned mouse and human CD
organoids derived from our expandable mouse and human UB organoids. These CD organoids
show typical pattern of PC and IC distribution as seem in vivo in the adult cortical/outer medullary
collecting duct, demonstrating the robustness of our organoid models.

29
Results
Screening for conditions to mature mouse UB organoids into CD organoids.
The functions of the mature renal CD system are carried out by two major cell populations
that are intermingled throughout the entire CD network. The more abundant principal cells (PCs)
concentrate the urine and regulate Na+/K+ homeostasis via water and Na+/K+ transporters. The
less abundant α- and β-intercalated cells (ICs) regulate normal acid-base homeostasis via
secretion of H+ or HCO3− into the urine. The absence of an in vitro system recapitulating PC and
IC development in an appropriate 3D context, constrains physiological exploration, disease
modeling, and drug screening on the renal CD system. With this limitation in mind, we developed
a screen to establish conditions supporting the differentiation of CD organoids, assaying
expression of Aqp2
71
and Foxi1
72
, definitive markers for PC and IC lineages, respectively, by
quantitative reverse transcription PCR (qRT-PCR), following 7 days of culture under variable but
defined culture conditions (Fig. 2-1a).
In a 1st round of screening, we determined the base condition in which minimal growth
factors/small molecules sustained the survival of the organoids and permitted their differentiation.
The base medium used for UBCM—hBI
20
—was tested, together with the commercially available
APEL medium for sustaining kidney organoid generation
14
. Combinations of FGF9, EGF, and
Y27632 were tested, together with the two different base media (Supplementary Fig. 3a). After 7
days of differentiation in the various conditions, we observed that the hBI + FGF9 + Y27632
condition enabled the survival of organoids and permitted spontaneous basal differentiation, as
assayed by a modest induction of both PC (Aqp2) and IC (Foxi1) specific gene expression
(Supplementary Fig. 3b, c).
To enhance the efficiency of differentiation, we carried out a 2nd round of screening
identifying molecules that strongly induced the expression of Aqp2 and/or Foxi1 under the hBI +
FGF9 + Y27632 condition. Agonists or antagonists targeting major developmental pathways (e.g.,
TGF-β, BMP, Wnt, FGF, Hedgehog, and Notch) were tested, together with hormonal inputs known
30

Fig. 2-1 Generating mature and highly organized mouse CD organoids from UB organoids. a Schematic of the
screening strategy for identifying differentiation culture condition to generate mouse CD organoids from UB organoids.
b qRT-PCR analyses of UB (red) and CD (green) organoids for UB progenitor markers Wnt11 and Ret; PC markers
Aqp2 and Aqp3; IC markers Foxi1, Atp6v1b1, Slc4a1, and Slc26a4; and Tfcp2l1 that is expressed in both PC and IC.
Adult mouse kidney (blue) was used as control. The significance was determined by two-tailed unpaired Student’s t
tests. n = 3. c Bright field images showing the morphologic differences between UB and CD organoids. d –g Whole-
mount immunostaining analyses of CD organoids for PC marker AQP2, and IC markers FOXI1 and ATP6V1B1,
showing the distribution of these two cell types within the organoid. f Shows the higher-power images for the boxed
region in (e). h Comparison of PC and IC ratios in postnatal Day 0 (P0) mouse CD, adult mouse CD, and CD organoids.
Whole-mount immunostaining images (CD organoids), or section staining images (P0 and adult kidneys) stained for
AQP2 (PC) and FOXI1 (IC) were quantified for ratios of PC and IC in the CD organoid or the kidney’s collecting duct.
Each column represents counts from nine different fields of view (n = 9, three fields were randomly selected from each
of the three biologically independent samples). i Principal component analysis (PCA) of RNA-seq data. Different colors
and oval circles represent different primary kidney cell populations (NPC, IPC, UB tip, UB trunk, and adult CD cells),
or UB organoids cultured for 5 (D5), 10 (D10), and 20 days (D20), or CD organoids. Scale bars: c, d 200 μm, e 100
μm, f 25 μm, g 40 μm. All data are presented as mean ± s.d. Source data are provided as a Source Data file.

31
to regulate PC or IC activity (aldosterone and vasopressin). BMP7, DAPT (a Notch pathway
inhibitor), JAKI (JAK inhibitor I), and PD0325901 (MEK inhibitor) dramatically increased both Aqp2
and Foxi1 expression, while JAG-1
73
(Notch agonist) and aldosterone led to a preferential
increase in Foxi1 expression, and vasopressin to enhanced Aqp2 expression (Supplementary Fig.
3d–f). In a 3rd round of screening, testing of various combinations of these factors led to the
identification of an optimized CD differentiation medium (CDDM, Supplementary Table 8)
supplemented with FGF9, Y27632, DAPT, PD0325901, aldosterone, and vasopressin.

Generating mature and highly organized mouse CD organoids from UB organoids.
Seven days of UB organoid culture in CDDM resulted in a morphologically elongated CD
organoid phenotype (Fig. 2-1c). qRT-PCR revealed a marked decrease in the expression of the
UPC genes (Wnt11 and Ret) and a concomitant elevation in the expression of PC-specific water
transporter encoding genes (Aqp2 and Aqp3
71
) and IC-specific transcription factor (Foxi1), proton
pump (Atp6v1b1
74
), and Cl−/HCO3− exchangers (Slc4a1/Ae1
75
, α-IC-specific; Slc26a4/Pendrin
76
,
β-IC-specific) (Fig. 2-1b). Immunostaining confirmed the presence of AQP2, AQP3, FOXI1,
TFCP2L1
77
, and ATP6V1B1 in the CD organoids (Fig. 2-1d–g and Supplementary Fig. 3g–j).
Differentiating CD organoids displayed a clear lumen (Fig. 2-1e, f), and the organization of PC
and IC cell types reflected that of the postnatal mouse kidney CD in which
FOXI1+/ATP6V1B1+/TFCP2L1+/KIT+
70,78
ICs were dispersed in AQP2+/AQP3+ PCs
(Supplementary Fig. 3k–n). Further, in some areas, AQP2 and AQP3 showed a differential
subcellular localization, AQP2 to the apical luminal facing surface and AQP3 to basolateral
plasma membrane, reflecting the normal cellular distribution of these critical components of water
trafficking through PC cells (Fig. 2-1e, f and Supplementary Fig. 3i, j). PC and IC ratios in the CD
organoids indicated a higher IC portion (50–55%) over PC portion (40–45%) than observed in the
neonatal and adult kidney, a likely reflection of DAPT-mediated Notch inhibition in CDDM culture
(Fig. 2-1h); in vivo, IC-derived Notch ligand signaling inhibits the IC fate
6,77,79,80
.
32
To better define the identity of the CD organoids, we used RNA-seq to profile the
transcriptome of the organoids. These data were compared with mouse CD (mCD) freshly
isolated by FACS from the kidney of adult Hoxb7-Venus mice, as well as UB organoids, and prior
RNA-seq data for primary UB tip and UB trunk populations
46
, NPCs and IPCs
47
. PCA showed a
clear separating of CD organoids from the immature UB tip and UB organoid populations, and
similar grouping to UB trunk and primary mCD (Fig. 2-1i), supporting an expected cell maturation
of CD organoids. Importantly, the CDDM differentiation protocol was highly reproducible when
testing UB organoids derived from different genetic backgrounds (Supplementary Fig. 3o). The
evidence above support that we have generated a mature and patterned kidney organoid that
recapitulates the adult kidney collecting system, in a chemically defined manner.

Generating human CD organoids from iUB organoids with the mouse CDDM.
To determine whether we can generate iCD organoid from iUB organoid, we developed a
refined CDDM from the mouse version (human CD differentiation medium (hCDDM),
Supplementary Table 8), which efficiently induced the mRNA expression of various PC (AQP2,

Fig. 2-2 Generating human iCD organoids from hPSC-derived iUB organoids. a Schematic of the stepwise
differentiation from any hPSC line into iUB and iCD organoid without using reporter. (TeSR mTeSR1 medium, CR
CloneR, ME (v2) mesendoderm stage medium version 2, UB-I UB Stage I medium, UB-II UB Stage II medium). b
qRT-PCR analyses of the iUB organoid (blue) and iUB-derived iCD organoid (orange), for UB (WNT11, RET), PC
(AQP2, AQP3, AQP4), and IC markers (FOXI1). c Whole-mount immunostaining of the iUB-derived iCD organoid
for PC marker AQP3 and broad CD marker CDH1. Scale bars, 50 μm. All data are presented as mean ± s.d. In b,
the significance was determined by two-tailed unpaired Student’s t tests; n = 3. Source data are provided as a
Source Data file.
33
AQP3, and AQP4) and IC (FOXI1) markers hundreds to thousands of fold within 14 days of
differentiation from iUB organoids, accompanying the reduction of UB tip genes WNT11 and RET
(Fig. 2-2 a, b, and Supplementary Fig. 5g). Given the limited availability of validated antibodies,
we were only able to examine AQP3 and FOXI1 at the protein level (Supplementary Fig. 7g–m).
Approximately, 20–30% of iCD organoid cells were AQP3+ but we were not able to observe
FOXI1+ ICs (Fig. 2-2c). These results suggest that long-term expandable iUB organoids generate
PC and IC-like cells but cells generated in hCDDM do not attain a fully mature CD cell fates.

Generating high-quality mature iCD organoids from iUB organoids with optimized protocol.
Our previous iCD differentiation protocol mentioned above was only a slightly modified
version from the mouse. The iCD organoids generated following this protocol were neither mature
nor healthy. To generate more mature and high-quality organoids that recapitulate mature adult
human collecting duct system, we further optimized our iUB-iCD directed differentiation protocol
based on our previous and unpublished data to mimic the CD development in vivo. Following this
protocol, we were able to first induce the elongation and branching of the iUB organoids (Fig. 2-
3a), sign of the cell fate transition from UB tip progenitors to UB stalk, and then further mature
and differentiate into collecting duct cell type, with significant elevation of various PC (AQP2,
AQP3, AVPR2) and IC (FOXI1, PENDRIN) markers hundreds to thousands of folds from iUB
organoids (Fig. 2-3b). More importantly, we were able to show the expression of FOXI1 on the
protein level, for the first time in the kidney organoid field without force expression ectopically in
the cell, together with PC marker AQP2 (Fig. 2-3c). Meanwhile, these organoids showed typical
lumen structure within each elongated tubule as well as apical localization of AQP2, recapitulating
their in vivo counterpart. These results suggest the optimized protocol can now generate a more
mature and high-quality iCD organoids from our expandable iUB organoids. Further
characterization including immunostaining of more markers, bulk and single-cell RNA-Seq, and
34
functional studies are currently undertaken. Manuscript is in preparation for submission in a few
months.

Fig. 2-3 Generating matured human iCD organoids from iUB organoids. a Bright field time-course images
showing the morphological change during differentiation from iUB organoid to iCD organoid, on day 0 (D0), day 10
(D10), and day 21 (D21). b qRT-PCR analyses of the iUB organoid (blue) and iUB-derived iCD organoid (orange),
for UB (WNT11, RET), PC (AQP2, AQP3, AVPR2), and IC markers (FOXI1, PENDRIN). c Whole-mount
immunostaining of the iUB-derived iCD organoid for PC marker AQP2, IC marker FOXI1, and broad CD marker
CDH1. Note the presence of lumen and the apical localization of AQP2. Scale bars, a and c, 200 μm.

35
Chapter 2 Discussion

Leveraging our ability to produce large quantities of high-quality UPCs in the format of
expandable branching UB organoids, we performed a screening that identified CDDM - a cocktail
of growth factors, small molecules, and hormones that together can differentiate mouse, and for
the first time in the field, human UB organoids into CD organoids with spatially patterned mature
PCs and ICs. The molecular mechanisms underlying the UB-to-CD transition are still largely
unknown. The in vitro organoid system provides a tool to study this process, and the chemically
defined components in CDDM shed light on potential signals that trigger CD maturation in vivo.
Despite the general difficulty of maturing stem-cell-derived tissues, our study shows that it is
possible to achieve proper patterning and maturation in vitro, similar to what we observe in vivo,
when starting from high-quality progenitor cells under appropriate culture conditions. Further
characterization of these CD organoids through detecting more marker genes/proteins expression,
bulk and single-cell level transcriptome analysis, and functional assays are currently on going.
An interesting observation from our mouse CD organoid is that under normal
differentiation condition, we only see the elevation of type-B specific Cl
-
/HCO 3
-
anion exchanger
Slc26a4 expression but no type-A specific Slc4a1, but under a different culture condition, we see
the significant decrease of Slc26a4 and elevation of Slc4a1. This possibly suggests that we
induced type-B ICs in our organoids under normal condition, but how we were able to induce type-
A ICs under a different environmental cue demonstrates the possibility of using our organoid
model to investigate type-A and type-B IC cell fate specification.
In chapter 1, we showed efficient genome editing in UB organoids. Our proof-of-concept
Ret/RET KO experiment demonstrated that our organoid system could recapitulate genetic
malfunction of branching morphogenesis in vitro as that of in vivo. Now with the ability to generate
high-quality CD organoids from the genome edited UB organoids, we open up many applications
in the study of kidney development and disease. UB and CD organoids can be generated from
available transgenic mouse strains or hPSC lines that bear genetic mutations related to kidney
36
development and disease. In addition, disease-relevant mutations can be introduced into the UB
organoid directly, enabling the investigation of pathophysiology throughout the entire course of
kidney branching morphogenesis, from the UB branching period to the mature CD stage.
In conclusion, the UB and CD organoid system provides a powerful tool for studying kidney
development and modeling kidney disease. The mature PCs and ICs present in the CD organoids
are potential sources for cell replacement therapies for patients with CD damage. The ability to
produce large quantities of UB and CD organoids also provides a robust platform for drug
screening. And ultimately, with the ability to generate, capture, and expand both NPCs and UPCs,
two of the key building blocks for kidney organogenesis, we made a big step towards regenerating
the kidney.

37
Conclusion and Future Direction
In this dissertation study, I reported the establishment of culture systems enabling the
expansion of mouse and human UPCs as 3D UB organoids, as well as the generation of mature
CD cell types from these UB organoids. These organoids recapitulate their in vivo counterparts
both morphologically and functionally. These organoid systems can serve as a powerful platform
for the study of kidney development and disease, drug screening, therapeutics, and provide key
building blocks for kidney regenerative medicine.
In the future, besides further characterizing these organoids, especially the different
functions of PCs and ICs in our human CD organoids, we would like to explore the PC/IC as well
as type-A/type-B IC cell fate determination in our organoids under different culture conditions,
which might provide important insights in the study of kidney CD development and cell-type
specification. We would also like to model one of the most common inherited kidney diseases,
polycystic kidney disease (PKD), using our organoid system by introducing gene mutations to the
organoids. The establishment of this organoid disease model will not only provide a powerful tool
for studying PKD pathogenesis, but also provide a platform for drug screening to identify potential
drug for PKD treatment. And lastly, we would like to explore the Organ-On-A-Chip technology
using our organoids to investigate the effect of different environmental cues, such as extracellular
matrix or flow, on the UB/CD cells.

38
Methods

Human tissues
All human fetal kidney samples were collected under Institutional Review Board approval (USC-
HS-13-0399 and CHLA-14-2211). Following the patient decision for pregnancy termination, the
patient was offered the option of donation of the products of conception for research purposes,
and those that agreed signed an informed consent form. No financial gain arose on donation of
tissue. This did not alter the choice of termination procedure, and the products of conception from
those that declined participation were disposed of in a standard fashion. The only information
collected was gestational age and whether there were any known genetic or structural
abnormalities.

Mice
All animal work was performed under Institutional Animal Care and Use Committee approval
(USC IACUC Protocol # 20829). Swiss Webster mice were purchased from Taconic Biosciences
(Model # SW-F, MPF 4 weeks). Sox9-GFP mice were kindly shared from Dr Haruhiko Akiyama
48
.
Wnt11-RFP mice (JAX # 018683), Hoxb7-Venus mice (JAX # 016252), and Rosa26-Cas9/GFP
(JAX #026179) were obtained from the Jackson Laboratory.

hPSC lines
Experiments using hPSCs were approved by the Stem Cell Oversight Committee (SCRO) of
University of Southern California under protocol # 2018-2. hPSCs are routinely cultured in
mTeSR1 medium in monolayer culture format coated with Matrigel and passaged using dispase.

39
3D cultured NPC lines
The 3D cultured NPC lines we used in this study were derived from E11.5 whole kidney cells of
the wild-type Swiss Webster mouse strain, using an improved method we developed that can
derive NPC lines from any mouse strain without the need for prior purification of NPCs
42
. These
NPCs had been cultured 6–12 months (billions of billion-fold of expansion) before used for
reconstruction of engineered kidney with UB organoids.

Derivation of mouse UB organoid
From intact T-shaped UB. Male mice with the desired genotype (Wnt11-RFP, Hoxb7-Venus,
Sox9-GFP, or Rosa26-Cas9/GFP) were mated with female Swiss Webster mice. Plugs were
checked the next morning; midday of plug positive was designated as embryonic Day 0.5 (E0.5).
Timed pregnant mice were euthanized at E11.5. Kidneys were dissected out from embryos using
standard dissection techniques and transferred into a 1.5-mL Eppendorf tube on ice. Next, at least
500-μL fresh, pre-warmed 0.1% (w/v) collagenase IV (Thermo Fisher, Cat. No. 17104019) was
added into the tube and incubated at 37 °C for 20 min. After incubation, collagenase was removed
and at least 500 μL of 10% FBS (1X DMEM, 1X Gluta-MAX-I, 1X MEM NEAA, 0.1 μM 2-
Mercaptoethanol, 1X Pen Strep, 10% FBS) was added to resuspend the kidneys. One to three
kidneys were transferred each time with 80–100-μL medium onto a 100-mm petri dish lid as a
working droplet. UBs were isolated from the surrounding MM and other tissues using sterile
needles (BD, Cat. No. BD305106) without damaging UBs. The isolated UBs can be temporarily
left in the medium at room temperature for <30 min while dissecting other UBs. After all UBs were
isolated, each UB was transferred together with 1–3-μL medium into an 8-μL cold Matrigel droplet
at the bottom of one well of a U- bottom 96-well low-attachment plate, by using a P10 micropipette.
The UB and Matrigel were mixed by pipetting gently 2–3 times. After all UBs were embedded in
Matrigel, the plate was incubated at 37 °C for 20 min for the Matrigel to solidify. Then, 100 μL of
40
mouse UBCM (mUBCM) was slowly added into each well and the plate was then transferred into
an incubator set at 37 °C with 5% CO2.
From dissociated UB single cells. For deriving UB organoid from dissociated UB single cells
(e.g., for gene editing purpose), after the isolation of E11.5 T-shaped UBs from kidneys following
the procedures described above, all UBs were collected into a 1.5-mL Eppendorf tube with the
medium removed as much as possible. An appropriate amount (e.g., for 20 UBs, we use 200 μl,
adjust accordingly) of prewarmed Accumax cell dissociation solution (Innovative Cell
Technologies, # AM105) was added into the tube, and the tube was then incubated at 37 °C for
20 min and gently tapped every 7–10 min. Then, an equal amount of 10% FBS was added into
the tube to neutralize the Accumax and the mixture was pipetted gently 8–10 times to dissociate
the UB into single cells. The tube was then centrifuged at 300 × g for 5 min. After centrifugation,
the supernatant was carefully removed and UB cells were resuspended in an appropriate amount
of mUBCM (Y27632 was supplemented at 10-μM final concentration for the first 24 h) by pipetting
gently 6–8 times. Cell density was measured using automatic cell counter (Bio-Rad, TC20).
Approximately, 2000 cells were transferred into each well of a U-bottom 96-well low-attachment
plate and extra amount of mUBCM (with 10-μM Y27632) was added to the well to make the final
volume 100 μl per well. The plate was then centrifuged at 300 × g for 3 min and transferred and
cultured in a 37 °C incubator. After 24 h, the ~2000 single cells formed an aggregate
autonomously and the aggregate was then transferred together with 1–3-μl medium into an 8-μl
cold Matrigel droplet in another well of the U-bottom 96-well low-attachment plate using a P10
micropipette. The aggregate was pipetted gently 2 - 3 times to mix with Matrigel. After all
aggregates were embedded in Matrigel, the plate was incubated at 37 °C for 20 min for the
Matrigel to solidify. Last, 100 μl of UBCM was added slowly into each well and the plate was then
transferred into an incubator set at 37 °C with 5% CO2.

41
Mouse UB organoid expansion and passaging
Mouse UBCM was renewed with fresh medium every 2 days, and UB organoid was passaged
every 5 days.
Manual passaging as small tips. UB organoid (with Matrigel) was first transferred from U-bottom
96-well plate onto a 100-mm petri dish lid with 80-100-μL medium using a P1000 pipette with the
tip cut 0.5-1 cm to widen the diameter. Most of the Matrigel surrounding the organoid was removed
using sterile needles under a dissecting microscope. A small piece of the organoid with 3-5
branching tips was cut using needles and then re-embedded into Matrigel droplet in a U-bottom
96-well low-attachment plate well and cultured in a 37 °C incubator following the same embedding
procedure described above.
Passaging as single cells. UB organoid (with Matrigel) was first transferred from U-bottom 96-
well plate onto a 100-mm petri dish lid with 80-100-μL medium using a P1000 pipette with the tip
cut 0.5-1 cm to widen the diameter. Most of the Matrigel surrounding the organoid was removed
using sterile needles under a dissecting microscope. Organoid was then cut into small pieces
using sterile needles (the smaller the piece, the easier to dissociate). All the pieces were
transferred into 1.7-mL Eppendorf tubes (1–3 organoids per tube) with as little medium as
possible using a P200 pipette, extra medium was removed from the tube. Two hundred to four
hundred microliters of pre-warmed Accumax cell dissociation solution was added into the tube.
Then followed the same procedure described above (in “Derivation of mouse UB organoid” -
“From dissociated UB single cells,” following the addition of Accumax) to dissociate organoid
pieces into single cells and reaggregate for continuing culture.

Derivation of human UB organoid
Organoid derivation from RET+ primary UPCs purified from human fetal kidney. The kidney
nephrogenic zone was dissected manually from each of fresh 9-13-week human fetal kidney,
chopped into small pieces with surgical blade, and divided into 4-6 1.5-mL Eppendorf tubes.
42
Tissues were washed with PBS and resuspended with 500 μL of pre-warmed Accumax per tube
and the tubes were incubated at 37 °C with shaking for 25 min. Five hundred microliters 10% FBS
was then added to neutralize the Accumax, and the mixture was pipetted ~25 times to dissociate
the tissues into single cells. The mixture medium with kidney cells were then pooled together and
sieved through a 40-μm cell strainer, then transferred into 1.5-mL Eppendorf tubes, centrifuged
at 300 × g for 5 min and the supernatant was carefully removed. All cell pellets from this
preparation were then resuspended and combined into 300-400-μL cold FACS medium (1x PBS,
1X Pen Strep, 2% FBS) supplemented with a human anti-RET antibody at 1:200 dilution into one
tube and incubated for 30 min on ice. The tube was gently tapped every 10 min to ensure mixing.
After 30 min, 1-mL cold FACS medium was added into the tube. The tube was then centrifuged
at 300 × g for 5 min and the supernatant was carefully removed. Cell pellet was resuspended
again in 500-μL cold FACS medium plus secondary antibody (Donkey anti-Goat, Alexa Fluor 568,
Invitrogen, Cat. # A-11057) at 1:1000 dilution and incubated for 30 min on ice, with gentle mixing
every 10 min. After the incubation, 1-mL cold FACS medium was added into the tube. The tube
was then centrifuged at 300 × g for 5 min and the supernatant was carefully removed. The pelleted
cells were resuspended with 300-500-μL cold FACS medium plus DAPI at 1:2000 ratio, placed
through 40-μm cell strainer and transferred into a FACS tube on ice before FACS. RET+ (Alexa
Fluro 568) UPCs were then sorted out by FACS. The RET+ cells were collected in a 1.5-mL
Eppendorf tube with 500-μL 10% FBS. The tube was centrifuged at 300 × g for 5 min and the
supernatant was carefully removed. Cell pellet was then resuspended in an appropriate amount
of hUBCM (with the addition of Y27632 at 10-μM final concentration) and cell density was
measured by automatic cell counter. ~2000–20,000 cells were transferred into each well of a U-
bottom 96-well low-attachment plate and an appropriate amount of hUBCM (with 10-μM Y27632
for the first 24 h) was added to the well to make the final volume of 100 μL per well. After 24 h,
UB cell aggregate was formed and embedded into an 8-μL cold Matrigel droplet in another well
of the U-bottom 96-well low-attachment plate and cultured in a 37 °C incubator following the same
43
embedding procedure described above (with hUBCM). After ~10-15 days of culture, epithelial tip
structures could be seen budding out from the aggregate. These tip structures were dissected out
and re-embedded into Matrigel and expanded as human UB organoid.
Organoid derivation from human ESCs and iPSCs. The hPSCs were pre-treated with 10-μM
Y27632 in mTeSR1 medium for 1 h before dissociation into single cells using Accumax cell
dissociation solution. Following dissociation, ~60,000-80,000 cells (seeding number needs be
optimized for different hPSC lines) were seeded into Matrigel coated 12-well plate with 1-mL
mTeSR1 medium with 10-μM Y27632 (old protocol, Figs. 4) or 1x CloneR (STEMCELL
Technologies, Cat. No. 05888) (refined protocol, Fig. 5) (Day 0). Twenty-four hours later (Day 1),
the medium was removed and 1 mL of pre-warmed ME stage medium was slowly added to the
well. Forty-eight hours later (Day 3), ME stage medium was removed and 1 mL of UB-I stage
medium was slowly added to the well. Twenty-four hours later (Day 4), medium was changed to
1 mL of fresh UB-I medium again. After another 24 h (Day 5), UB-I medium was removed, and
1.5-2 mL of UB-II stage medium was slowly added to the well. Twenty-four hours later (Day 6),
medium was changed to 1.5-2 mL of fresh UB-II medium. At Day 7, differentiated cells were
dissociated into single cells following the standard Accumax dissociation method. For wild-type
hESC line without any reporter, cell pellet after dissociation was resuspended in 250–400-μL cold
FACS medium (1x PBS, 1X Pen Strep, 2% FBS) supplemented with a PE conjugated anti-human
CD117(C-KIT) antibody (Biolegend, Cat. No.313204) at 1:200 dilution and incubated for 30 min
on ice. The tube was gently tapped every 10 min to ensure mixing. After 30 min, 1-mL cold FACS
medium was added into the tube. The tube was then centrifuged at 300 × g for 5 min and the
supernatant was carefully removed. The pelleted cells were resuspended with 300–500-μL cold
FACS medium plus DAPI at 1:2000 ratio, placed through 40-μm cell strainer (Greiner bio-one,
Cat. No. 542040) and transferred into a FACS tube on ice before FACS. C-KIT+ (PE) cells were
then sorted out by FACS. For reporter cell lines (PAX2-mCherry/WNT11-GFP hESC line or
SOX9-GFP hiPSC line), dissociated cells were either resuspended directly in 300-500-μL cold
44
FACS medium plus DAPI at 1:2000 ratio, or first went through the C-KIT staining process
described above before resuspended in FACS medium with DAPI. These cells were then put
through a 40-μm cell strainer and placed on ice before FACS. mCherry+ cells (from the PAX2-
mCherry/WNT11-GFP hESC line), GFP+ cells (from the SOX9-GFP hiPSC line), or C-KIT+ (PE)
cells (after C-KIT staining) were then sorted out by FACS. Upon FACS sorting of the mCherry+
or GFP+ or PE+ cells, these cells were collected in a 1.5-mL Eppendorf tube with 500-μL 10%
FBS. The tube was centrifuged at 300 × g for 5 min, and the supernatant was carefully removed.
Cell pellet was then resuspended in an appropriate amount of hUBCM or hUBCM-v2 (with the
addition of Y27632 at 10-μM final concentration), and cell density was measured by automatic
cell counter. Approximately, 2000–20,000 cells were transferred into each well of a U-bottom 96-
well low-attachment plate, and an appropriate amount of hUBCM or hUBCM-v2 (with 10-μM
Y27632 for the first 24 h) was added to the well to make the final volume of 100 μL per well. After
24 h, UB cell aggregate was formed and embedded into an 8-μL cold Matrigel droplet in another
well of the U-bottom 96-well low-attachment plate and cultured in a 37 °C incubator following the
same embedding procedure described above (with hUBCM or hUBCM-v2). After ~7-10 days of
culturing, epithelial tip structures could be seen budding out from the aggregate. WNT11-GFP
expression was induced within 10 days in the PAX2-mCherry/WNT11-GFP hESC line-derived
organoid. These tip structures were dissected out and re-embedded into Matrigel to continue
culture. From then, the iUB organoid was established and can be passaged stably following the
procedures below.

Human UB organoid expansion and passaging
During the culture, human UBCM was renewed with fresh medium every 2 days. Both human UB
organoid from primary RET+ UPC and iUB organoid from hPSCs were passaged every 6-10 days
depending on the size. The passaging methods (manual or as single cells) are the same as
45
defined above in the mouse UB organoid section, with the change of using hUBCM or hUBCM-
v2 instead of mUBCM.

CD differentiation
mCD differentiation. Mouse UB organoid was passaged at Day 5 of expansion as single cells,
and 2000 cells were seeded for continuing expansion. At Day 10 of mUB expansion, mUBCM
was removed and 150-μL 1x PBS was added and removed to wash the organoid. One hundred
and fifty microliters of mouse CDDM was then added to initiate mCD differentiation (mCD
differentiation Day 0). The organoid was cultured in a 37 °C incubator and medium was changed
every 2 days or daily as needed if the medium turns yellow/orange for a total of 7 days. No
passage of the organoid was needed. At mCD differentiation Day 7, the mCD organoid was
harvested for analyses.
Human CD differentiation. After human UB organoid expansion was stabilized (at least 25 days
post FACS when UB organoid were growing stably) and reached an appropriate size (at least
900-μm diameter), hUBCM was removed and 150-μL 1x PBS was added and removed to wash
the organoid. One hundred and fifty microliters of hCDDM was added to start hCD differentiation
(hCD differentiation Day 0). The organoid was cultured in 37 °C incubator and medium was
changed every 2 days or daily if needed if the medium turns yellow/orange for a total of 14 days.
At hCD differentiation Day 14, hCD organoid was harvested for analyses.

Mouse engineered kidney generation (Refer to Supplementary Methods for more details)
A small piece of mUB organoid was manually dissected out and inserted into a microdissected
hole on a 3D cultured mNPC aggregate to generate a engineered kidney precursor. This
precursor was then carefully transferred into a well of a U-bottom 96-well low-attachment plate
with 100-μL kidney reconstruction medium (APEL2 + 0.1-μM TTNPB) with 10-μM Y27632 and
cultured in 37°C incubator (Day 0). After 24 h (Day 1), the engineered kidney precursor was
46
transferred onto a six-well transwell insert membrane with kidney reconstruction medium for
continuing culture.

RNA sequencing
Adult mCD cells were FACS isolated (Hoxb7-Venus+) from adult (~2 month old) Hoxb7-Venus
mouse kidneys. All samples were collected and lysed in TRIzol reagent and stored under −80 °C.
Total RNA was extracted using the Direct-zol RNA MicroPrep Kit (Zymo). cDNA library was
prepared using the KAPA Stranded mRNA-Seq Kit (KAPA Biosystems). RNA sequencing was
performed by the Children’s Hospital Los Angeles Molecular Pathology Genomics Core.

Gene editing in UB organoids (Refer to Supplementary Methods for more details)
Lentiviral infection was used to edit genes in mUB cells. One hundred microliters mixture of 1-5x
virus, 10-μM Polybrene, and UBCM was added to the U-bottom 96-well low-attachment plate well
with UB single cells suspension. The UBs and virus were centrifuged together at 800 g for 15-30
min for spinfection
81
at room temperature. The infected UB cells were then washed, aggregated
overnight, embedded in Matrigel, and cultured in UBCM in 37 °C incubator following standard UB
organoid culture procedures described above. Appropriate antibiotic was added to the culture to
select for UB cells that have been successfully infected.

RNA-seq data analysis
RNA sequencing data was analyzed using Partek Flow (version 10.0.21.0411), including
published dataset of IPCs and NPCs
47
, ureteric tip and trunk cells
46
. FASTQ files were trimmed
from both ends based on a minimum read length of 25 bps and a shred quality score of 20 or
higher. Reads were aligned to GENCODE mm10 (release M24) using STAR 2.5.3a. Aligned
reads were quantified to the Partek E/M annotation model. Gene counts were normalized by
adding 1 then by TMM values. All TMM values can be found in Supplementary Data 1 (Online).
47
Samples were filtered to include differentially expressed genes of UB tip compared to UB trunk,
with false discovery rate ≤0.01, fold change <−4 or >4, total counts ≥10, and p value < 0.05,
resulting in 1413 UB tip/trunk signature genes (Supplementary Data 2, online). Then, hierarchical
clustering was produced on by clustering samples and features with average linkage cluster
distance and Euclidean point distance. PCA was performed using the EDASeq R/Bioconductor
packages and the plot was rendered with the ggplot2 R package.

Statistics and reproducibility
All statistical analysis shown in figures are specified in the corresponding legends. All statistical
analysis was performed using Prism 8 (version 8.2.1). All bright field and immunofluorescence
micrographs shown are representative images selected from at least three independent
experiments all showing similar results. Genetic engineering of hESCs and DNA gel
electrophoresis (Supplementary Fig. 4a-e) were performed once.

Data availability
RNA-seq data have been submitted to Gene Expression Omnibus (GEO) with accession number
“GSE149109 [https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE149109].” For
GUDMAP/RBK resources, visit https://www.gudmap.org. Key resources such as antibodies,
chemicals, recombinant proteins, genotyping, and qRT-PCR primers are provided in
Supplementary Tables 3–5. All other relevant data supporting the key findings of this study are
available within the article and its Supplementary Information files or from the corresponding
author upon reasonable request. A reporting summary for this Article is available as a
Supplementary Information file. Source data can be found online.

48
Supplemental Information

Supplementary Figures

Supplementary Figure 1. Screening for optimal UB culture condition. a Summary of UB culture condition
screening steps. See Supplementary Methods for more details. b Bright field (BF) and fluorescence (Wnt11-RFP)
images of E11.5 Wnt11-RFP UB cultured in previously described UB culture condition
21
, from day 0 (D0), day 3
(D3), day 6 (D6) to day 9 (D9). Scale bars, 200 μm. c Bright field images showing the effect of withdrawal of each
individual factor from previously described UB culturing condition

as indicated. C1, CHIR99021 at the concentration
of 1 μM. Scale bars, 200 μm. d Bright field images showing the improved UB branching by increasing CHIR99021
concentration from 1μM (C1) to 3μM (C3). Scale bars, 200 μm. e Bright field (BF) and fluorescence (Wnt11-RFP)
images showing improved Wnt11-RFP expression by replacing FGF1 with FGF9. Scale bars, 200 μm. f, g Summary
of screening results of the 1
st

round screening in Stage III for identifying individual hits. In the columns of “Wnt11-
RFP” and “Organoid growth”: “-” indicates no significant differences were observed compared to the control group
(FGF9+C3+TT+GDNF). In the column of “Selected for R2 screening”: “Y” indicates the factor was selected for
2
nd
round (R2) screening; “-” indicates the factor was not selected. h-j Bright field (BF) and fluorescence (Wnt11-
RFP) images showing improved Wnt11-RFP expression in UB organoids with the addition of LDN (h), A83 (i), and
Rspo1 (j), as compared to their corresponding controls, in the 1
st

round screening in Stage III. LDN, LDN193189;
A83, A83-01; Rspo1, R-Spondin 1. Scale bars, 200 μm. k Bright field (BF) and fluorescence (Wnt11-RFP) images
showing improved Wnt11-RFP expression in UB organoid with the addition of both JAK Inhibitor I and SB202190
(+JK/SB), as compared to the corresponding control (-JK/SB), in the 2
nd

round screening in Stage III. Scale bars,
200 μm. l Bright field (BF) and fluorescence (Wnt11-RFP) images showing the morphology and Wnt11-RFP reporter
expression in the UB organoids cultured in complete UBCM or upon withdrawal of indicated components from the
UBCM for 9 days. TT, TTNPB. Scale bars, 200 μm.
49

Supplementary Figure 2. Derivation and characterization of mouse UB organoid. a, b Immunostaining of the
branching UB organoid at day 10 of culture for various UB markers. Note the red signals that do not overlay with
DAPI from PAX2 and ETV5 staining panels are non-specific signals. In b, the four panels on the right represent the
boxed region in the left panel. Scale bars, a, 200 μm; b, 100 μm. c Unsupervised clustering analysis of RNA-seq
data. d Bright field (BF) and fluorescence images of a UB organoid derived from Sox9-GFP genetic background.
Scale bars, 200 μm. e Bright field (BF) and fluorescence images of a UB organoid derived from Rosa26-Cas9/GFP
(R26-Cas9/GFP) genetic background. Scale bars, 200 μm. f Bright field (BF) and fluorescence images of a Wnt11-
RFP UB organoid revived from freezing. Scale bars, 200 μm. g Bright field images showing 200 single cells
embedded into a drop of Matrigel and cultured in mUBCM for 1 day (D1) and 5 days (D5). Scale bars, 500 μm. h
Efficiency of UB organoid formation from 200 single UB cells. Each group represents 3 biological replicates. All
data are presented as mean ± s.d. Source data are provided as a Source Data file.
50

51

Supplementary Figure 3. Expandable UB organoid-based screening for CD differentiation. a Summary of
conditions tested in the 1st round of CD differentiation condition screening. b, c qRT-PCR analyses of the 1
st

round
of CD differentiation condition screening for PC marker gene Aqp2 (b, in blue) and IC marker gene Foxi1 (c, in
orange). d Summary of chemicals tested in the 2
nd

round of CD differentiation condition screening with the R1-2
medium as base medium. e, f qRT-PCR analysis of the 2nd round of CD differentiation condition screening for PC
marker gene Aqp2 (e, in blue) and IC marker gene Foxi1 (f, in orange). Note that data for R2-2 and R2-21 are not
presented here, because dramatic cell death was observed in those conditions and were thus excluded from this
analysis. g-j Immunostaining of cryo-section (g-i) and whole-mount (j) samples of differentiated CD organoids for
ureteric lineage marker (GATA3) and various PC (AQP2 and AQP3) and IC markers (FOXI1, ATP6V1B1, and KIT),
and TFCP2L1, which is expressed stronger in the IC than PC at the protein level
48
. Note the sporadic distribution
of the IC in the organoid. Scale bars, 50 μm. k-n Immunostaining of cryo-section samples of postnatal day 0 kidney
(P0, k and l) and adult mouse kidney (m and n) CD cells for ureteric lineage marker (GATA3) and various PC
(AQP2 and AQP3) and IC markers (FOXI1 and ATP6V1B1). Note the sporadic distribution of the IC in the organoid.
Scale bars, k and l, 50 μm; m and n, 100 μm. o Summary of CD organoid derivation efficiency. All data are
presented as mean ± s.d. Source data are provided as a Source Data file.
52

Supplementary Figure 4. Genetic engineering hESC with a dual reporter system by CRISPR/Cas9 and
generation of iUB organoids from various hPSC reporter lines. a Schematic of the genetic engineering of
PAX2-mCherry reporter into the hESC line. See Supplementary Methods for more details. b PCR-based
genotyping results of PAX2-mCherry reporter knockin. Wild-type (wt) PCR product is 2011 bp, knockin (KI) product
is 4308 bp. *, biallelic KI clones; **, monoallelic KI clones; wild-type hESC “hESC (wt)” is used as a non-editing
control. c PCR-based genotyping result for Cre-based excision of PGK-Neo cassette. PCR product for PAX2-
mCherry knockin with PGK-Neo excised (KI w.o. Neo) is 2779 bp. *, clones with biallelic excision of PGK-Neo; **,
clones without PGK-Neo excision; wild-type hESC “hESC (wt)”, and biallelic KI parental clone “KI (homo)” is used
as controls. d Schematic of the engineering of WNT11-GFP reporter into the PAX2-mCherry reporter hESC line.
See Supplementary Methods for more details. e PCR-based genotyping result for WNT11-GFP reporter knockin.
Wild-type (wt) PCR product is 219 bp, knockin (KI) product is 2528 bp. *, biallelic KI clones; **, monoallelic KI
clones; wild-type hESC “hESC (wt)” is used as a non-editing control. f Immunostaining of hESC-derived
mesendoderm cells (at the end of ME stage) for mesendoderm marker T (Brachyury). Scale bars, 100 μm. g
Immunostaining of hESC-derived UB precursor cells (at the end of UB-II stage and prior to FACS sorting) for
various UB markers. Scale bars, 100 μm. h Flow cytometry analysis of mCherry
+

and GFP
+

cells differentiated
from WNT11-GFP/PAX2-mCherry dual reporter hESCs. i Bright field (BF) and fluorescence images showing the
induction of WNT11-GFP expression in the mCherry
+

aggregate upon extended culture in hUBCM. Scale bars,
200 μm. j Flow cytometry analysis of GFP+ cells differentiated from SOX9-GFP reporter hiPSCs. k-m Bright field
(k and l) and SOX9-GFP (m) fluorescence images of branching iUB organoid derived from SOX9-GFP reporter
hiPSCs in a typical passage cycle at day 0 (D0) and day 9 (D9). The indicated budding structure shown in (k) was
dissected and re-embedded into Matrigel for iUB expansion shown in (l and m). Scale bars, 200 μm.

53

Supplementary Figure 5. Generating human iUB and iCD organoids from the dual-reporter hPSC line
independent of its reporters. a Flow cytometry analysis of KIT+

precursor cells differentiated from the dual
reporter hESC line. b Bright field (BF) and fluorescence images showing the induction of WNT11-GFP expression
in the KIT+

aggregate upon continued culture in hUBCM-v2. Scale bars, 200 μm. c Cumulative growth curve of
iUB organoid culture starting from 20,000 cells at day 20. Each time point represents 3 biological replicates. d-f
Immunostaining of cryo-section samples of the expandable iUB organoid for various UB markers. The four panels
on the right represent the boxed region in the left panel. Scale bars, left panel, 100 μm; right 4 panels, 50 μm. g
Bright field and fluorescence images showing the morphological changes and decreasing of WNT11-GFP from
iUB organoid (left panels) to mature iCD organoid (right panels). Scale bars, 200 μm. h Summary of human iUB
organoids derivation from different hPSC lines independent of reporter and their expansion in vitro. Total culture
time are up until manuscript submission. The maximum organoid culture time and expansion could be longer. All
data are presented as mean ± s.d. Source data are provided as a Source Data file.

54

55

Supplementary Figure 6. Generating human iUB organoids from the SOX9-GFP iPSC independent of its
reporter. a Flow cytometry analysis of KIT+

precursor cells differentiated from the SOX9-GFP hiPSC line. Note
the SOX9-GFP reporter expression is induced in the majority of the cells using this new differentiation protocol. b
Bright field (BF) and fluorescence images showing the SOX9-GFP expression in the KIT+

iUB organoid cultured in
hUBCM-v2. Scale bars, 200 μm. c Cumulative growth curve of iUB organoid culture starting from 20,000 cells at
day 20. Each time point represents 3 biological replicates. d qRT-PCR analyses of the FACS purified KIT+

precursor (orange) and KIT+

iUB organoids cultured for 53 days (D53, green) for various UB markers.
Undifferentiated hiPSCs (dark blue) and human fetal kidney (gray, 11.2-week gestational age) were used as
controls. The significance was determined by two-tailed unpaired Student’s t-tests; n=3. e-g and i-l Immunostaining
of whole-mount (e-g) and cryo-section (i-l) samples of the expandable iUB organoid for various UB markers. The
four panels on the right represent the boxed region in the left panel. Scale bars, e-g, left panels, 100 μm; right 4
panels, 40 μm; i-l, left panels, 100 μm; right 4 panels, 50 μm. h Quantification of percentages of iUB cells stained
positive for different UB markers in Supplementary Fig. 6e-g. Each column represents counts from 3 different fields
of view (n=3). All data are presented as mean ± s.d. Source data are provided as a Source Data file.

56

Supplementary Figure 7. Immunostaining of whole-mount samples of control and RET KO iUB organoids
and human CD antibodies testing. a-f Corresponding to Fig. 1-5g. Whole-mount immunostaining of the control
(a-d) and RET KO (e and f) human iUB organoids for various UB markers. a, c, and e were from the experimental
group with sgRNA #1; b, d, and f were from the experimental group with sgRNA #2. Scale bars, a and b, 100 μm;
c-f, 40 μm. g-l Immunostaining of 17.5-week gestational age human fetal kidney cryo-section samples for testing
various human CD marker antibodies, including PC markers AQP2, AQP3, AQP4, and IC markers FOXI1. Arrow
heads in (h) and (l) indicate non-specific signals. Scale bars, g-i, k, l, 50 μm; j, 100 μm. m Summary of all CD
antibodies we have tested in human kidney samples.

57

58

Supplementary Figure 8. FACS sequential gating/sorting strategies used in iUB differentiation. a
Corresponding to Fig. 1-4i, gating strategy for detection of KIT+

precursor cells differentiated from wild-type H1
hESC. b Corresponding to Supplementary Fig. 4h, gating strategy for detection of mCherry+

and GFP+

cells
differentiated from WNT11-GFP/PAX2-mCherry dual reporter hESCs. c Corresponding to Supplementary Fig. 4j,
gating strategy for detection of GFP+ cells differentiated from SOX9-GFP reporter hiPSCs. d Corresponding to
Supplementary Fig. 5a, gating strategy for detection of KIT+

precursor cells differentiated from the dual reporter
hESC line. e Corresponding to Supplementary Fig. 6a, gating strategy for detection of KIT+

pre-cursor cells
differentiated from the SOX9-GFP hiPSC line.

59
Supplementary Tables
Supplementary Table 1. Summary of available in vitro models for mouse kidney
branching morphogenesis

Supplementary Table 2. Summary of available in vitro models for human kidney
branching morphogenesis

Supplementary Table 3-9. Key resources and medium recipes
See Appendix, Key resources.
60
Bibliography
1 Costantini, F. & Kopan, R. Patterning a complex organ: branching morphogenesis and
nephron segmentation in kidney development. Dev Cell 18, 698-712 (2010).
https://doi.org:10.1016/j.devcel.2010.04.008
2 Costantini, F. Genetic controls and cellular behaviors in branching morphogenesis of the
renal collecting system. Wiley Interdiscip Rev Dev Biol 1, 693-713 (2012).
https://doi.org:10.1002/wdev.52
3 Rosenblum, S., Pal, A. & Reidy, K. Renal development in the fetus and premature infant.
Semin Fetal Neonatal Med 22, 58-66 (2017). https://doi.org:10.1016/j.siny.2017.01.001
4 Little, M. H. & McMahon, A. P. Mammalian kidney development: principles, progress, and
projections. Cold Spring Harb Perspect Biol 4 (2012).
https://doi.org:10.1101/cshperspect.a008300
5 Shakya, R., Watanabe, T. & Costantini, F. The role of GDNF/Ret signaling in ureteric bud
cell fate and branching morphogenesis. Dev Cell 8, 65-74 (2005).
https://doi.org:10.1016/j.devcel.2004.11.008
6 Jeong, H. W. et al. Inactivation of Notch signaling in the renal collecting duct causes
nephrogenic diabetes insipidus in mice. J Clin Invest 119, 3290-3300 (2009).
https://doi.org:10.1172/JCI38416
7 Al-Awqati, Q. & Gao, X. B. Differentiation of intercalated cells in the kidney. Physiology
(Bethesda) 26, 266-272 (2011). https://doi.org:10.1152/physiol.00008.2011
8 McMahon, A. P. Development of the Mammalian Kidney. Curr Top Dev Biol 117, 31-64
(2016). https://doi.org:10.1016/bs.ctdb.2015.10.010
9 Nicolaou, N., Renkema, K. Y., Bongers, E. M., Giles, R. H. & Knoers, N. V. Genetic,
environmental, and epigenetic factors involved in CAKUT. Nat Rev Nephrol 11, 720-731
(2015). https://doi.org:10.1038/nrneph.2015.140
10 Clevers, H. Modeling Development and Disease with Organoids. Cell 165, 1586-1597
(2016). https://doi.org:10.1016/j.cell.2016.05.082
11 Li, M. & Izpisua Belmonte, J. C. Organoids - Preclinical Models of Human Disease. N Engl J
Med 380, 569-579 (2019). https://doi.org:10.1056/NEJMc1903253
12 Fatehullah, A., Tan, S. H. & Barker, N. Organoids as an in vitro model of human
development and disease. Nat Cell Biol 18, 246-254 (2016).
https://doi.org:10.1038/ncb3312
61
13 Taguchi, A. et al. Redefining the in vivo origin of metanephric nephron progenitors
enables generation of complex kidney structures from pluripotent stem cells. Cell Stem
Cell 14, 53-67 (2014). https://doi.org:10.1016/j.stem.2013.11.010
14 Takasato, M. et al. Kidney organoids from human iPS cells contain multiple lineages and
model human nephrogenesis. Nature 526, 564-568 (2015).
https://doi.org:10.1038/nature15695
15 Morizane, R. et al. Nephron organoids derived from human pluripotent stem cells model
kidney development and injury. Nat Biotechnol 33, 1193-1200 (2015).
https://doi.org:10.1038/nbt.3392
16 Freedman, B. S. et al. Modelling kidney disease with CRISPR-mutant kidney organoids
derived from human pluripotent epiblast spheroids. Nat Commun 6, 8715 (2015).
https://doi.org:10.1038/ncomms9715
17 Vanslambrouck, J. M. et al. Enhanced metanephric specification to functional proximal
tubule enables toxicity screening and infectious disease modelling in kidney organoids.
Nature Communications 13 (2022). https://doi.org:10.1038/s41467-022-33623-z
18 Brown, A. C., Muthukrishnan, S. D. & Oxburgh, L. A synthetic niche for nephron progenitor
cells. Dev Cell 34, 229-241 (2015). https://doi.org:10.1016/j.devcel.2015.06.021
19 Tanigawa, S., Taguchi, A., Sharma, N., Perantoni, A. O. & Nishinakamura, R. Selective In
Vitro Propagation of Nephron Progenitors Derived from Embryos and Pluripotent Stem
Cells. Cell Rep 15, 801-813 (2016). https://doi.org:10.1016/j.celrep.2016.03.076
20 Li, Z. et al. 3D Culture Supports Long-Term Expansion of Mouse and Human Nephrogenic
Progenitors. Cell Stem Cell 19, 516-529 (2016).
https://doi.org:10.1016/j.stem.2016.07.016
21 Little, M. H. & Combes, A. N. Kidney organoids: accurate models or fortunate accidents.
Genes Dev 33, 1319-1345 (2019). https://doi.org:10.1101/gad.329573.119
22 Tran, T. et al. A scalable organoid model of human autosomal dominant polycystic kidney
disease for disease mechanism and drug discovery. Cell Stem Cell 29, 1083-1101 e1087
(2022). https://doi.org:10.1016/j.stem.2022.06.005
23 Xu, Y. et al. Adult human kidney organoids originate from CD24(+) cells and represent an
advanced model for adult polycystic kidney disease. Nat Genet 54, 1690-1701 (2022).
https://doi.org:10.1038/s41588-022-01202-z
24 Qiao, J., Sakurai, H. & Nigam, S. K. Branching morphogenesis independent of
mesenchymal-epithelial contact in the developing kidney. Proc Natl Acad Sci U S A 96,
7330-7335 (1999). https://doi.org:10.1073/pnas.96.13.7330
62
25 Qiao, J. et al. Multiple fibroblast growth factors support growth of the ureteric bud but
have different effects on branching morphogenesis. Mech Dev 109, 123-135 (2001).
https://doi.org:10.1016/s0925-4773(01)00592-5
26 Bush, K. T. et al. TGF-beta superfamily members modulate growth, branching, shaping,
and patterning of the ureteric bud. Dev Biol 266, 285-298 (2004).
https://doi.org:10.1016/j.ydbio.2003.10.023
27 Rosines, E. et al. Staged in vitro reconstitution and implantation of engineered rat kidney
tissue. Proc Natl Acad Sci U S A 104, 20938-20943 (2007).
https://doi.org:10.1073/pnas.0710428105
28 Yuri, S., Nishikawa, M., Yanagawa, N., Jo, O. D. & Yanagawa, N. In Vitro Propagation and
Branching Morphogenesis from Single Ureteric Bud Cells. Stem Cell Reports 8, 401-416
(2017). https://doi.org:10.1016/j.stemcr.2016.12.011
29 Tan, Z., Rak-Raszewska, A., Skovorodkin, I. & Vainio, S. J. Mouse Embryonic Stem Cell-
Derived Ureteric Bud Progenitors Induce Nephrogenesis. Cells 9 (2020).
https://doi.org:10.3390/cells9020329
30 Chow, T. et al. Recapitulating kidney development in vitro by priming and differentiating
mouse embryonic stem cells in monolayers. npj Regenerative Medicine 5 (2020).
https://doi.org:10.1038/s41536-020-0092-5
31 Taguchi, A. & Nishinakamura, R. Higher-Order Kidney Organogenesis from Pluripotent
Stem Cells. Cell Stem Cell 21, 730-746 e736 (2017).
https://doi.org:10.1016/j.stem.2017.10.011
32 Xia, Y. et al. Directed differentiation of human pluripotent cells to ureteric bud kidney
progenitor-like cells. Nat Cell Biol 15, 1507-1515 (2013). https://doi.org:10.1038/ncb2872
33 Mae, S. I. et al. Generation of branching ureteric bud tissues from human pluripotent stem
cells. Biochem Biophys Res Commun 495, 954-961 (2018).
https://doi.org:10.1016/j.bbrc.2017.11.105
34 Tsujimoto, H. et al. A Modular Differentiation System Maps Multiple Human Kidney
Lineages from Pluripotent Stem Cells. Cell Rep 31, 107476 (2020).
https://doi.org:10.1016/j.celrep.2020.03.040
35 Mae, S. I. et al. Expansion of Human iPSC-Derived Ureteric Bud Organoids with Repeated
Branching Potential. Cell Rep 32, 107963 (2020).
https://doi.org:10.1016/j.celrep.2020.107963
36 Uchimura, K., Wu, H., Yoshimura, Y. & Humphreys, B. D. Human Pluripotent Stem Cell-
Derived Kidney Organoids with Improved Collecting Duct Maturation and Injury Modeling.
Cell Rep 33, 108514 (2020). https://doi.org:10.1016/j.celrep.2020.108514
63
37 Howden, S. E. et al. Plasticity of distal nephron epithelia from human kidney organoids
enables the induction of ureteric tip and stalk. Cell Stem Cell 28, 671-684 e676 (2021).
https://doi.org:10.1016/j.stem.2020.12.001
38 Shi, M. et al. Human ureteric bud organoids recapitulate branching morphogenesis and
differentiate into functional collecting duct cell types. Nat Biotechnol 41, 252-261 (2022).
https://doi.org:10.1038/s41587-022-01429-5
39 Zeng, Z. et al. Generation of patterned kidney organoids that recapitulate the adult kidney
collecting duct system from expandable ureteric bud progenitors. Nat Commun 12, 3641
(2021). https://doi.org:10.1038/s41467-021-23911-5
40 Majumdar, A., Vainio, S., Kispert, A., McMahon, J. & McMahon, A. P. Wnt11 and Ret/Gdnf
pathways cooperate in regulating ureteric branching during metanephric kidney
development. Development 130, 3175-3185 (2003). https://doi.org:10.1242/dev.00520
41 Lu, B. C. et al. Etv4 and Etv5 are required downstream of GDNF and Ret for kidney
branching morphogenesis. Nat Genet 41, 1295-1302 (2009).
https://doi.org:10.1038/ng.476
42 Li, Z., Araoka, T. & Belmonte, J. C. I. Gene Editing in 3D Cultured Nephron Progenitor Cell
Lines. Methods Mol Biol 1926, 151-159 (2019). https://doi.org:10.1007/978-1-4939-
9021-4_13
43 Pachnis, V., Mankoo, B. & Costantini, F. Expression of the c-ret proto-oncogene during
mouse embryogenesis. Development 119, 1005-1017 (1993).
https://doi.org:10.1242/dev.119.4.1005
44 Schuchardt, A., D'Agati, V., Larsson-Blomberg, L., Costantini, F. & Pachnis, V. Defects in
the kidney and enteric nervous system of mice lacking the tyrosine kinase receptor Ret.
Nature 367, 380-383 (1994). https://doi.org:10.1038/367380a0
45 Harding, S. D. et al. The GUDMAP database--an online resource for genitourinary research.
Development 138, 2845-2853 (2011). https://doi.org:10.1242/dev.063594
46 Rutledge, E. A., Benazet, J. D. & McMahon, A. P. Cellular heterogeneity in the ureteric
progenitor niche and distinct profiles of branching morphogenesis in organ development.
Development 144, 3177-3188 (2017). https://doi.org:10.1242/dev.149112
47 Lindstrom, N. O. et al. Conserved and Divergent Features of Mesenchymal Progenitor Cell
Types within the Cortical Nephrogenic Niche of the Human and Mouse Kidney. J Am Soc
Nephrol 29, 806-824 (2018). https://doi.org:10.1681/ASN.2017080890
48 Nakamura, Y. et al. Wwp2 is essential for palatogenesis mediated by the interaction
between Sox9 and mediator subunit 25. Nature Communications 2, 251 (2011).
https://doi.org:10.1038/ncomms1242
64
49 Platt, R. J. et al. CRISPR-Cas9 knockin mice for genome editing and cancer modeling. Cell
159, 440-455 (2014). https://doi.org:10.1016/j.cell.2014.09.014
50 Sato, T. et al. Single Lgr5 stem cells build crypt-villus structures in vitro without a
mesenchymal niche. Nature 459, 262-265 (2009). https://doi.org:10.1038/nature07935
51 Lindstrom, N. O. et al. Conserved and Divergent Features of Human and Mouse Kidney
Organogenesis. J Am Soc Nephrol 29, 785-805 (2018).
https://doi.org:10.1681/ASN.2017080887
52 Combes, A. N., Zappia, L., Er, P. X., Oshlack, A. & Little, M. H. Single-cell analysis reveals
congruence between kidney organoids and human fetal kidney. Genome Med 11, 3 (2019).
https://doi.org:10.1186/s13073-019-0615-0
53 Li, Z. et al. BMP4 Signaling Acts via dual-specificity phosphatase 9 to control ERK activity
in mouse embryonic stem cells. Cell Stem Cell 10, 171-182 (2012).
https://doi.org:10.1016/j.stem.2011.12.016
54 Zuo, E. et al. One-step generation of complete gene knockout mice and monkeys by
CRISPR/Cas9-mediated gene editing with multiple sgRNAs. Cell Res 27, 933-945 (2017).
https://doi.org:10.1038/cr.2017.81
55 Menon, R. et al. Single-cell analysis of progenitor cell dynamics and lineage specification
in the human fetal kidney. Development 145 (2018). https://doi.org:10.1242/dev.164038
56 Umeda, K. et al. Long-term expandable SOX9+ chondrogenic ectomesenchymal cells from
human pluripotent stem cells. Stem Cell Reports 4, 712-726 (2015).
https://doi.org:10.1016/j.stemcr.2015.02.012
57 Skinner, M. A., Safford, S. D., Reeves, J. G., Jackson, M. E. & Freemerman, A. J. Renal
Aplasia in Humans Is Associated with RET Mutations. The American Journal of Human
Genetics 82, 344-351 (2008). https://doi.org:10.1016/j.ajhg.2007.10.008
58 Arora, V. et al. Biallelic Pathogenic GFRA1 Variants Cause Autosomal Recessive Bilateral
Renal Agenesis. J Am Soc Nephrol 32, 223-228 (2021).
https://doi.org:10.1681/ASN.2020040478
59 Pichel, J. G. et al. Defects in enteric innervation and kidney development in mice lacking
GDNF. Nature 382, 73-76 (1996). https://doi.org:10.1038/382073a0
60 Sanchez, M. P. et al. Renal agenesis and the absence of enteric neurons in mice lacking
GDNF. Nature 382, 70-73 (1996). https://doi.org:10.1038/382070a0
61 Michos, O. et al. Kidney development in the absence of Gdnf and Spry1 requires Fgf10.
PLoS Genet 6, e1000809 (2010). https://doi.org:10.1371/journal.pgen.1000809
65
62 Mendelsohn, C., Batourina, E., Fung, S., Gilbert, T. & Dodd, J. Stromal cells mediate
retinoid-dependent functions essential for renal development. Development 126, 1139-
1148 (1999). https://doi.org:10.1242/dev.126.6.1139
63 Batourina, E. et al. Vitamin A controls epithelial/mesenchymal interactions through Ret
expression. Nat Genet 27, 74-78 (2001). https://doi.org:10.1038/83792
64 Rosselot, C. et al. Non-cell-autonomous retinoid signaling is crucial for renal development.
Development 137, 283-292 (2010). https://doi.org:10.1242/dev.040287
65 Marose, T. D., Merkel, C. E., McMahon, A. P. & Carroll, T. J. Beta-catenin is necessary to
keep cells of ureteric bud/Wolffian duct epithelium in a precursor state. Dev Biol 314, 112-
126 (2008). https://doi.org:10.1016/j.ydbio.2007.11.016
66 Bridgewater, D. et al. Canonical WNT/beta-catenin signaling is required for ureteric
branching. Dev Biol 317, 83-94 (2008). https://doi.org:10.1016/j.ydbio.2008.02.010
67 Tan, Z., Shan, J., Rak-Raszewska, A. & Vainio, S. J. Embryonic Stem Cells Derived Kidney
Organoids as Faithful Models to Target Programmed Nephrogenesis. Scientific Reports 8
(2018). https://doi.org:10.1038/s41598-018-34995-3
68 Vanslambrouck, J. M. et al. Direct reprogramming to human nephron progenitor-like cells
using inducible piggyBac transposon expression of SNAI2-EYA1-SIX1. Kidney Int 95, 1153-
1166 (2019). https://doi.org:10.1016/j.kint.2018.11.041
69 Kim, J., Kim, Y. H., Cha, J. H., Tisher, C. C. & Madsen, K. M. Intercalated cell subtypes in
connecting tubule and cortical collecting duct of rat and mouse. J Am Soc Nephrol 10, 1-
12 (1999). https://doi.org:10.1681/ASN.V1011
70 Ransick, A. et al. Single-Cell Profiling Reveals Sex, Lineage, and Regional Diversity in the
Mouse Kidney. Dev Cell 51, 399-413 e397 (2019).
https://doi.org:10.1016/j.devcel.2019.10.005
71 Nielsen, S. et al. Aquaporins in the kidney: from molecules to medicine. Physiol Rev 82,
205-244 (2002). https://doi.org:10.1152/physrev.00024.2001
72 Blomqvist, S. R. et al. Distal renal tubular acidosis in mice that lack the forkhead
transcription factor Foxi1. J Clin Invest 113, 1560-1570 (2004).
https://doi.org:10.1172/JCI20665
73 Wang, X. et al. Cloning and variation of ground state intestinal stem cells. Nature 522,
173-178 (2015). https://doi.org:10.1038/nature14484
74 Wagner, C. A. et al. Renal vacuolar H+-ATPase. Physiol Rev 84, 1263-1314 (2004).
https://doi.org:10.1152/physrev.00045.2003
66
75 Alper, S. L., Natale, J., Gluck, S., Lodish, H. F. & Brown, D. Subtypes of intercalated cells in
rat kidney collecting duct defined by antibodies against erythroid band 3 and renal
vacuolar H+-ATPase. Proc Natl Acad Sci U S A 86, 5429-5433 (1989).
https://doi.org:10.1073/pnas.86.14.5429
76 Royaux, I. E. et al. Pendrin, encoded by the Pendred syndrome gene, resides in the apical
region of renal intercalated cells and mediates bicarbonate secretion. Proc Natl Acad Sci
U S A 98, 4221-4226 (2001). https://doi.org:10.1073/pnas.071516798
77 Werth, M. et al. Transcription factor TFCP2L1 patterns cells in the mouse kidney collecting
ducts. Elife 6 (2017). https://doi.org:10.7554/eLife.24265
78 Chen, L. et al. Transcriptomes of major renal collecting duct cell types in mouse identified
by single-cell RNA-seq. Proc Natl Acad Sci U S A 114, E9989-E9998 (2017).
https://doi.org:10.1073/pnas.1710964114
79 Park, J. et al. Single-cell transcriptomics of the mouse kidney reveals potential cellular
targets of kidney disease. Science 360, 758-763 (2018).
https://doi.org:10.1126/science.aar2131
80 Mukherjee, M. et al. Endogenous Notch Signaling in Adult Kidneys Maintains Segment-
Specific Epithelial Cell Types of the Distal Tubules and Collecting Ducts to Ensure Water
Homeostasis. J Am Soc Nephrol 30, 110-126 (2019).
https://doi.org:10.1681/ASN.2018040440
81 Kurita, M. et al. In vivo reprogramming of wound-resident cells generates skin epithelial
tissue. Nature 561, 243-247 (2018). https://doi.org:10.1038/s41586-018-0477-4

67
Appendix

Supplementary Methods

Screening for optimal UB culture condition
We systematically screened the optimal UB culture condition in four different stages
(Stages I-IV, Supplementary Fig. 1a). In Stage I, we tested the most updated UB culture condition
from literature, Yuri et al., 2017
21
, in which FGF1, retinoic acid (RA), CHIR99021, and GDNF were
used to allow the growth of isolated UBs in vitro. By repeating this condition, we confirmed that
this medium supported very well the growth and branching of the UB in the first 4-5 days. However,
after that, UB growth slowed down significantly, and more importantly, Wnt11-RFP expression
was dramatically decreased (Supplementary Fig. 1b), indicating that further optimization was
needed to selectively expand the Wnt11+ UB progenitor cells.
In Stage II, we optimized the individual medium components employed by Yuri et al. (FGF1,
RA, CHIR99021, and GDNF). We first asked if each individual factor was necessary in the
medium. For this, the factors were withdrawn from the medium individually, and the results clearly
showed that all these factors were essential for maintaining the branching of UB (Supplementary
Fig. 1c). Then we further optimized these factors. RA is known to be unstable in tissue culture, so
we replaced it with another widely used small molecule RA substitute TTNPB. CHIR99021 was
used at 1μM by Yuri et al., but based on our own experiences and literature, different doses of
CHIR99021 often have different biological effects. So we titrated it from 1μM, 3μM, to 6μM, from
which we identified 3μM to be the optimal concentration (Supplementary Fig. 1d, f). To optimize
for FGF1, we tested different members from the FGF family, including FGF2, FGF4, FGF7, FGF8,
FGF9, FGF10, and FGF20, from which we identified FGF9 to be superior in supporting Wnt11-
RFP expression than FGF1 (Supplementary Fig. 1e, f). GDNF was unchanged in the medium,
considering its essential role in maintaining the UB progenitor population in vitro and in vivo.
68
In Stage III, based on the optimized recipe consisting of FGF9, TTNPB, CHIR99021 (3μM)
and GDNF, we preformed our 1
st
round of screening of growth factors and small molecules
targeting major developmental pathways (e.g. TGF-β, BMP, Wnt, FGF, Hedgehog and Notch)
and others. The branching morphogenesis, growth rate and Wnt11-RFP were recorded as
readouts (Supplementary Fig. 1f, g). From this, we identified several hits that improved either the
UB growth rate or Wnt11-RFP, or both. Representative images were shown for LDN193189
(Supplementary Fig. 1h), A83-01 (Supplementary Fig. 1i), and R-Spondin 1 (Supplementary Fig.
1j). These individual hits were then subjected to a 2
nd
round of screening to test their effects in
various combinations, eventually leading to the identification of the optimal UB progenitor culture
medium UBCM consisting of FGF9, TTNPB, CHIR99021 (3μM), GDNF, LDN193189, A83-01, R-
Spondin1, JAKI, and SB202190. Representative images were shown for the combinatorial effect
of JAKI and SB2020190 (Supplementary Fig. 1k), for which only marginal effects were observed
when used individually (Supplementary Fig. 1f, g).
Lastly, in Stage IV, we asked whether each of the components in UBCM was essential.
The factors were removed from the UBCM individually and the results indicated that all of them
were necessary to achieve optimal UB organoid branching and to sustain Wnt11-RFP expression
(Supplementary Fig. 1l).

WNT11-GFP/PAX2-mCherry dual reporter hESC line generation
CRISPR-Cas9 based genome editing was used to insert 2A-EGFP-FRT-PGK-Neo-FRT
or 2A-mCherry-loxP-PGK-Neo-loxP cassette downstream of the stop codon (removed) of
endogenous WNT11 or PAX2 gene, respectively. DNA sequences ~1Kb upstream and ~1Kb
downstream of endogenous WNT11 (upstream F: CCGGAATTCGACGTAATCATTCCACTGACC;
upstream R: TACGAGCTCCTTGCAGACATAGCGCTCCAC; downstream F: CGCGTCGAC-
GGCCCTGCCCTACGCCCCA; downstream R: CCCAAGCTTTGCCTGGAAACTGGA-
GAGCTCCCTC) and PAX2 (upstream F: GAAGTCGACTTTCCACCCATTAGGGGCCA; up-
69
stream R: TATGCTAGCGTGGCGGTCATAGGCAGCGG; downstream F: TATACGCGTTTAC-
CGCGGGGACCACATCA; downstream R: GACGGTACCAGTAACTGCTGGAGGAAGAC) stop
codon were cloned upstream and downstream of 2A-EGFP-FRT-PGK-Neo-FRT or 2A-mCherry-
loxP-PGK-Neo-loxP cassette respectively to facilitate homologous recombination. 2A-EGFP
fragment was cloned from pCAS9_GFP (Addgene #44719) and the FRT-PGK-Neo-FRT cassette
was cloned from pZero-FRT-Neo3R (kindly provided by Dr. Keiichiro Suzuki). 2A-mCherry-loxP-
Neo-loxP fragment was cloned from Nanog-2A-mCherry plasmid (Addgene #59995). The
different fragments were then cloned to pUC19 plasmid to make the complete donor plasmids for
both knockin experiments. gRNA oligos for WNT11 (F: CACCGGTCCTCGCTCCTGCGTGGGG;
R: AAACCCCCACGCAGGAGCGAGGACC) and PAX2 (F:
CACCGATGACCGCCACTAGTTACCG; R: AAACCGGTAACTAGTGGCGGTCATC) were
synthesized and cloned into the lentiCRISPRv2 plasmid (Addgene # 52961). First, both donor
and gRNA plasmids for PAX2 reporter KI were transfected into the H1 hESCs using the
Lipofectamine 3000 Transfection Reagent (Invitrogen, Cat. No. L3000015). Neomycin-resistant
single cell colonies were picked up manually and genotyping was performed based on PCR. PCR
primers see Supplementary Table 4, results see Supplementary Fig. 4. Clones with biallelic
knockin of PAX2-mCherry were chosen for second round screen where plasmid encoding Cre
was delivered to allow the transient expression Cre, whose activities excise the loxP-flanked PGK-
Neo cassette from the knockin alleles. PCR was performed to identify single cell clones in which
PGK-Neo cassettes were excised from both alleles. Then the same strategy was used to knock
in WNT11 reporter based on the successful biallelic PAX2-mCherry knockin clones.

Gene editing in UB organoids
Gene over-expression:
Lentiviral infection was used to overexpress GFP in E11.5 mUB cells. Lentivirus was first
concentrated 100x using Lenti-X Concentrator kit from Takara (Cat # 631231). Concentrated
70
lentivirus was aliquoted and stored in -80°C before use. The lentivirus was used at 1x final
concentration together with 10 μM Polybrene (Sigma-Aldrich, Cat. No. TR-1003-G) diluted in
mUBCM (with 10 μM Y27632). 100 μL virus-UBCM mixture was added to the U-bottom 96-well
low-attachment plate well with single cells suspension prepared from 8-10 E11.5 mUBs. The UBs
and virus were centrifuged together at 800g for 30 minutes for spinfection at room temperature.
After the spinfection, the virus-UBCM mixture was removed and the infected UB cells were
washed three times with PBS, then aggregated overnight and embedded in Matrigel and cultured
in mUBCM in 37°C incubator following standard UB organoid culture procedures described above.
200 μg/mL G-418 (Invitrogen, Cat. # 10131027) was added to the culture to select for UB cells
that have been successfully infected. The resulting UB aggregate self-organized into typical
branching organoid 4-5 days after infection.

GFP knockout:
An E11.5 mUB single cell suspension from the Rosa26-Cas9/GFP background was used
and lentiviral vectors were constructed using the lentiGuide-puro vector system (Addgene #52963)
following standard protocol to make lentiviruses expressing three different gRNAs targeting GFP
(gRNA sequences: F1: CACCGAAGGGCGAGGAGCTGTTCAC, R1:
AAACGTGAACAGCTCCTCGCCCTTC; F2: CACCGCTGAAGTTCATCTGCACCAC, R2
AAACGTGGTGCAGATGAACTTCAC; F3: CACCGGGAGCGCACCATCTTCTTCA, R3:
AAACTGAAGAAGATGGTGCGCTCCC) with the Cas9 cutting site 100-150bp apart, or three non-
targeting gRNAs as control. The 100x concentrated lentivirus were used at 5x together with 10
μM Polybrene diluted in mUBCM (with 10 μM Y27632). 100 μL virus-UBCM mixture was added
to the U-bottom 96-well low-attachment plate well to combine them with 10 E11.5 mUBs that have
been dissociated into single cells. The UB cells and virus were centrifuged at 800 x g for 30
minutes for spin-infection. After the spin, virus-UBCM mixture was removed and fresh virus-
UBCM mixture was added into the same well and the UB cells were spin-infected for another 30
71
minutes at 800g. Then, virus-UBCM mixture was removed and the infected UBs were washed
three times with PBS, then aggregated overnight and embedded in Matrigel and cultured in
mUBCM in 37°C incubator following standard UB organoid culture procedures described above.
0.2 μg/mL puromycin was added to the medium to select for UB cells that have been successfully
infected. The UB aggregate self-organized into typical branching organoid by 4-5 days post-
infection.

Ret/RET knockout in mouse/human UB organoid:
Day 5 cultured mUB organoids (wildtype or any background) or stably expanded hUB
organoids were dissociated into single cells following the method described above. gRNA oligos
targeting mouse or human Ret/RET were synthesized and cloned into the lentiCRISPRv2 plasmid
(Addgene # 52961) (mRet gRNA: F1: CACCGGAAGCTCGGCACTTCTCCAG; R1: AAACCTG-
GAGAAGTGCCGAGCTTCC; F2: CACCGCTGTATGTAGACCAGCCAGC; R2: AAAC-
GCTGGCTGGTCTACATACAGC. hRET gRNA: F1: CACCGGTAGAGGCCCAATGCCACTG; R1:
AAACCAGTGGCATTGGGCCTCTACC; F2: CACCGAAGCATCCCTCGAGAAGTAG; R2:
AAACCTACTTCTCGAGGGATGCTTC). The same gRNA oligos cloned into the lentiGuide-puro
vector system (Addgene #52963) that don’t express the Cas9 enzyme were used as negative
control. Lentiviruses with these vectors were generated following standard protocol. The 100x
concentrated lentivirus were used at 2x together with 10 μM Polybrene diluted in m/hUBCM (with
10 μM Y27632). 100 μL virus-UBCM mixture was added to the U-bottom 96-well low-attachment
plate well to combine them with 15,000-20,000 m/hUB single cells. The UB cells and virus were
centrifuged at 800 x g for 15 minutes for spin-infection. Then, virus-UBCM mixture was removed
and the infected UBs were washed three times with PBS, then aggregated overnight and
embedded in Matrigel and cultured in m/hUBCM in 37°C incubator following standard UB
organoid culture procedures described above. Puromycin (0.2 μg/mL for mouse and 0.3 μg/mL
for human) was added to the medium two-days post-infection to select for UB cells that have been
72
successfully infected. The UB aggregate self-organized into typical branching organoid by 2-6
days post-infection. Mouse organoids were harvested 6 days post-infection and human organoids
were harvested 10-12 days post-infection for further analysis.

Human UB organoid cryopreservation
Human UB organoid were cultured until they reached the size ready for passaging. It was
transferred onto 100 mm Petri dish lid and Matrigel was removed following the method described
above. Organoid was then cut into 4-6 pieces using sterile needles and transferred into an
Eppendorf tube. Extra medium in the tube was removed and replaced with 200 μL hUBCM with
10 μM Y27632 supplemented with DMSO at 10%. The medium and organoid pieces were then
split into two cryogenic tubes for cryopreservation. To revive the organoid, the frozen cryogenic
tube was thawed in 37°C water bath. Medium in the tube was removed and replaced with 50-100
μL fresh hUBCM with 10 μM Y27632. Each organoid piece was then embedded into 8 μL Matrigel
and cultured in hUBCM (with 10 μM Y27632 for the first 24h) following the method described
before.

Mouse Engineered Kidney Generation
The day before generating the mouse engineered kidney, 50-60k 3D cultured mNPCs was
seeded per 96-well to aggregate overnight. A small piece (with 6-10 branching tips) of Day 7-10
cultured mUB organoid was manually dissected out using sterile needles (similar to passaging
UB organoid as small tips mentioned above) and inserted into a microdissected hole on a 3D
cultured mNPC aggregate (first, a fine dissecting tweezer was used to hold/stabilized the mNPC
aggregate sphere from one side, and a sterile needle was used to pierce a hole in the center of
mNPC aggregate sphere from the other side; the small piece of mUB organoid was then carefully
pushed into the hole using the needle; the NPC aggregate would then slowly wrap around the
inserted mUB organoid autonomously overnight; all these procedures were done in a drop (80-
73
100 μL) of kidney reconstruction medium (APEL2 + 0.1 μM TTNPB) with 10 μM Y27632 on an
inverted 100 mm plastic petri dish cap, to ensure minimal movement of the aggregate/organoid
during the procedures) in kidney reconstruction medium with 10 μM Y27632 to generate a
engineered kidney precursor. This engineered kidney precursor was then carefully transferred
into a well of a U-bottom 96-well low-attachment plate with 100 μL kidney reconstruction medium
with 10 μM Y27632, using a P200 pipette with the top 0.5-1 cm of the tip cut, and cultured in 37°C
incubator (day 0). After 24 h (day 1), dead cells surrounding the precursor were removed by gently
pipetting several times in the well using a P200 pipette with wide tip. The engineered kidney
precursor was then transferred onto a 6-well transwell insert membrane using a wide-tip
P200/P1000 pipette (depends on the size). Then 0.8-1 mL kidney reconstruction medium was
added in the lower chamber of the transwell. The medium was changed every two days for a total
of 7-10 days while the engineered kidney precursor maturation progressed. Then the engineered
kidney was processed for further analyses.

FACS
Cells were dissociated/prepared as described above. FACS sorting was performed on a
BD FACS ARIA lllu cell sorter. Sorted cells were collected in a 1.5 mL Eppendorf tube with 500
μL 10% FBS on ice.

RNA isolation, reverse transcription, and quantitative PCR
Samples were dissolved in 100 μL TRIzol (Invitrogen, Cat. No. 15596018) and kept in -
80°C freezer. RNA isolation was performed using the Direct-zol RNA MicroPrep Kit (Zymo
Research, Cat. No. R2062) according to the manufacturer’s instructions. Reverse transcription
was performed using the iScript Reverse Transcription Supermix (Bio-Rad, Cat. No. 1708841)
following the manufacturer’s instructions. qRT-PCR was performed using the Applied Biosystems
PowerUp SYBR Green Master Mix (Thermo Fisher, Cat. No. A25777) and carried out on an
74
Applied Biosystems Vii 7 RT-PCR system (Life Technologies). Validated gene-specific primers
can be found in Supplementary Table 5. Fold change was calculated using the comparative CT
Method (ΔΔCT Method) and Gapdh/GAPDH as housekeeping gene.

Immunofluorescence
Whole-mount staining:
Samples were fixed in 4% PFA for 45 minutes at 4°C in Eppendorf tubes with gentle
shaking (UB/CD organoid, 200 μL PFA) or 10 minutes at room temperature on transwell insert
membrane (kidney reconstruct, 1 mL total PFA on and below the membrane). They were then
washed four times in 0.8-1 mL 1X PBS (Corning, Cat. No. 21-040-CV) for total 30 minutes at 4°C
or room temperature (after the washes, kidney reconstructs on transwell membrane were cut out
and transferred into Eppendorf tubes). After the washes, samples were blocked in blocking
solution (0.1% PBST containing 3% BSA) for 1-2 hours at 4°C with gentle sharking, followed by
primary antibody staining (primary antibodies were diluted in blocking solution) at 4°C overnight.
On the second day, samples were washed three times with 800 μL 0.1% PBST for total 3 hours
at 4°C with gentle sharking. Secondary antibodies diluted in blocking solution were added and
samples were incubated at 4°C overnight. On the third day, samples were washed three times
with 800 μL PBST for total 3 hours at 4°C with gentle sharking. Lastly, samples were mounted in
mounting medium onto glass slides for imaging.

Cryo-section staining:
Samples were fixed and washed as described above. They were then transferred into a
plastic mold and embedded in OCT Compound (Scigen, Cat. No. 4586K1) and froze in -80°C for
24 hours to make a cryo-block. The cryo-blocks were sectioned using Leica CM1800 Cryostat.
The sectioned slides were then blocked for 30 minutes at room temperature followed by one hour
of primary antibodies staining at room temperature. The slides were then washed four times with
75
PBST for five minutes, and then secondary staining for 30 minutes. After the secondary staining,
the slides were washed four times with PBST for five minutes and mounted with mounting medium.

Image quantification for UB/CD marker gene expression in the UB/CD organoids (Figure
1e, 2h, 5f, 6d, 6h, supplementary figure 6h)
Whole-mount immunostaining images for mouse UB organoids, mouse CD organoids, or
human UB organoids were used for the quantification of various marker gene expression. ImageJ
software (version 1.52a) was used to count positive cells. 3 different fields of view per organoid
were randomly selected to count the number of positively stained cell numbers (positive for
marker genes) and total cell numbers (DAPI+). Percentage was calculated by the number of cells
that are positive for different UB/CD marker genes divided by the total DAPI+ cell numbers. At
least 500 cells in total were counted. Error bars represent standard derivation between different
field views.

76
Supplementary Tables 3. Key Resources
Antibodies

Chemicals, Peptides, and Recombinant Proteins

77
Biological Samples

Critical Commercial Assays

78
Deposited Data

Experimental Models: Cell lines

Experimental Models: Organisms/Strains

79
Supplementary table 4. Genotyping primer sequences for detecting PAX2-mCherry and
WNT11-GFP knockin (KI).

Supplementary table 5. qRT-PCR Primer sequences.

80
Supplementary table 6. Medium recipe of mUBCM (mouse UB culture medium).
Supplements:

81
Supplementary table 7. Medium recipe of hUBCM (Human UB culture medium).
Supplements:
hUBCM-v1

hUBCM-v2

82
Supplementary table 8. Medium recipe of CDDM (CD differentiation medium).
Supplements:
mCDDM (mouse):

hCDDM (human):

83
Supplementary table 9. Medium recipe of stepwise directed differentiation to ND cells (D1
to D7 from hPSC).
Supplements:

84
Additional Acknowledgements
I would like to thank Jeffrey Boyd and Bernadette Masinsin of the USC Flow Cytometry
Facility for FACS; Seth Ruffins of the USC Optical Imaging Facility for help with microscopy;
Dejerianne Ostrow and David Ruble of the Children’s Hospital Los Angeles Molecular Pathology
Genomics Core for RNA-seq; Meng Li, Yibu Chen, and Eddie Loh of the USC Norris Medical
Library Bioinformatics Service for help with the RNA-seq computational analysis; Haruhiko
Akiyama and Juan Carlos Izpisua Belmonte for sharing the Sox9-GFP mice, Naoki Nakayama
for sharing the SOX9-GFP hiPSC line; Dr Melissa L. Wilson (Department of Preventive
Medicine, University of Southern California) and Family Planning Associates for coordinating
fetal tissue collection; and Cristy Lytal for helping with editing the manuscript. This work is
supported by departmental startup funding and USC/UKRO Kidney Research Center funding.

Competing interests
Patent application has been filed on April 27, 2020 by University of Southern California
on behalf of inventors Zipeng Zeng, Biao Huang, Andrew P. McMahon, and Zhongwei Li for the
mouse and human UB organoid generation, expansion, and CD organoid differentiation
systems described in this study. The other authors declare no competing interests.

Abstract (if available)

Abstract Human kidney is responsible for filtering and processing blood and maintaining body fluid homeostasis. Normal kidney function is critical for supporting human life. However, there are still many knowledge gaps in the study of kidney development and diseases. One of the reasons is the lack of high-quality model that represent authentic human kidney for modeling development and disease pathogenesis. In the past decade, organoid systems leveraging the power of cell self-organization and 3-dementional cell culture have been adopted and became powerful tools for kidney research. Although various nephron organoid models have been reported, high-quality organoid system modeling the kidney collecting duct (CD) system, or it’s precursor, ureteric bud (UB) is still lacking. Here in this dissertation, in Chapter 1, I report the generation of expandable, 3D branching ureteric bud (UB) organoid culture model that can be derived from primary UB progenitors from mouse and human fetal kidneys, or generated de novo from human pluripotent stem cells. In Chapter 2, I report the generation of mouse and human CD organoids from these expandable UB organoids that recapitulate the adult kidney collecting duct system. These platforms will facilitate an enhanced understanding of development, regeneration and diseases of the mammalian collecting duct system.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

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

PDF

β-catenin couples self-renewal, induction and epithelial morphogenesis and patterning at the outset of mammalian nephrogenesis

PDF

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

PDF

Dissecting the heterogeneity of mouse hematopoietic stem cells in vivo

PDF

Molecular programs in epithelial morphogenesis in mammalian organ systems

PDF

Exploring the function of distal nephron enhancers in zebrafish

PDF

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

PDF

The autism-associated gene SYNGAP1 regulates human cortical neurogenesis

PDF

Deciphering retinoblastoma multi-step tumorigenesis with RB1 knockout cone reporter retinal organoids

PDF

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

PDF

Molecular role of EZH2 overexpression in Colorectal Cancer progression

PDF

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

PDF

Understand the distinct patterns of selection in auto-immune diseases with ancient DNA data by the S-LDSC model

PDF

Deciphering heterogeneity of preleukemic clonal expansion

PDF

Deconstructing the lung: a novel assay to evaluate lung stem/progenitor cell potential in vivo

PDF

Identification of therapeutic targets for neurons and microglia in amyotrophic lateral sclerosis

PDF

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

PDF

Characterization of metabolically distinct muscle resident fibro-adipogenic subpopulations reveals a potentially exploitable mechanism of skeletal muscle aging and disease

PDF

Mechanisms of airway multiciliated cell differentiation

PDF

The role of ERK1/2 in mouse embryonic stem cell fate control

Asset Metadata

Creator Zeng, Zipeng (author)

Core Title Generation of patterned and functional kidney organoids that recapitulate the adult kidney collecting duct system

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Development, Stem Cells and Regenerative Medicine

Degree Conferral Date 2023-05

Publication Date 04/03/2025

Defense Date 03/22/2023

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

Tag collecting duct,kidney,OAI-PMH Harvest,organoid,progenitor cell,ureteric bud

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Ying, Qi-Long (committee chair), Jadhav, Unmesh (committee member), Li, Zhongwei (committee member), McMahon, Andrew (committee member), Pastor-Soler, Nuria (committee member)

Creator Email zengzipeng123@gmail.com,zipengze@usc.edu

Unique identifier UC112932409

Identifier etd-ZengZipeng-11565.pdf (filename)

Legacy Identifier etd-ZengZipeng-11565

Document Type Dissertation

Format theses (aat)

Rights Zeng, Zipeng

Internet Media Type application/pdf

Type texts

Source 20230405-usctheses-batch-1016 (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. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

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