Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
The role of ERK1/2 in mouse embryonic stem cell fate control
(USC Thesis Other)
The role of ERK1/2 in mouse embryonic stem cell fate control
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
The Role of ERK1/2 in Mouse Embryonic Stem Cell Fate Control
By
Liang Hu
__________________________________________________________
A dissertation presented to the
FACULTY OF THE GRADUATE SCHOOL of
UNIVERSITY OF SOUTHERN CALIFORNIA in
partial fulfillment of the requirements for the degree of
Doctor of Philosophy
Development, Stem Cell, and Regenerative Medicine
December 2018
i
Dedication
To my parents and young brother, for their unselfish support and devotion
To my teachers, for their valuable guidance and advice
To my friends, for their continuous encouragement
ii
ACKNOWLEDGMENTS
I express my deepest appreciation to my mentor, Dr. Qilong Ying, for his continual patience,
advice, support, and encouragement. He is not only a good role model for me in the field of science,
but also an experienced advisor for my life.
I wish to thank my thesis committee members, Dr. Ichida Justin, Dr. Qilong Ying, and Dr. Wei Li
for their supervision, advice, and contribution regarding this dissertation work. From them, I have
learned a lot about how to excel at science and become a good story-teller. In addition, I greatly
appreciate my qualifying exam committee members, Dr. Robert Maxson, Dr. Cheng-Ming Chuong,
Dr. Ruchi Bajpai, and Dr. Michael Bonaguidi for their advice and ideas on my research project.
I wish to thank the members of Dr. Ying’s lab: Meng Zou, Sankalp Srivastava, Edward Trope, Tao
Zhou, Yongming Wu, Jing Xu, Xiong Xiao, Weilu Chen, Xingliang Zhou, Xi Chen, Chang Liu,
Yue Shi, Haeyoung Park, and Wesley Huang for their help, friendship, and ideas. Your
contributions have made my work and life in Dr. Ying’s lab become much easier. I would also like
to thank Peihen Gan, Hua Shen and Michaela Patterson in Dr. Sucov’s lab for their help.
I truly appreciate the collaboration and help from Nancy Wu and Zhen You at the
Transgenic/Knockout Core Facility at USC Norris Comprehensive Cancer Center for assisting
with the blastocyst injection and Erk1/2 knock-in mice generation. Special thanks should be given
to Dr. Chao Zhang for his expertise in chemical biology and valuable advice on my project of
studying Erk isoforms.
iii
Last but not the least, I wish to thank the PIBBS office and Department of Stem Cell Biology and
Regenerative Medicine offices for making everything run smoothly, especially Bami Andrada and
Ite A. Offringa.
iv
Table of Contents
Part I Deciphering the Roles of Erk1/2 Isoforms in Mouse Embryonic Stem Cell Fate Control
......................................................................................................................................................... 1
Abstract ....................................................................................................................................................1
Introduction .............................................................................................................................................2
1. The Origin and Application of Mouse Embryonic Stem Cells ....................................................2
2. Mouse Embryonic Stem Cells: Culture Conditions and Self-Renewal Signaling Pathways .......4
3. The MAPK/ERK Pathway and Its Inhibitors .............................................................................10
4. The Role of MAPK/ERK Pathway in ESC Self-Renewal .........................................................13
5. The Role of MAPK/ERK Pathway in ESC Differentiation .......................................................14
6. ERK1 and ERK2-Similarities and Differences ..........................................................................15
7. Chemical Genetics: A Useful Tool to Decipher Kinase Isoform-Specific Roles ......................21
8. Rationale of Current Study .........................................................................................................25
Experimental Results ............................................................................................................................28
1. Establishment and Characterization of ERK1/2 Knockout ESCs ..............................................28
2. Deletion or Chemical Inhibition of ERK1/2 is Sufficient for ESC Self-renewal in the presence
of GSK3 Inhibitor ...............................................................................................................................35
3. Deletion of either ERK1 or ERK2 Is Insufficient for ESC Self-Renewal in the Presence of GSK3
Inhibitor ...............................................................................................................................................39
4. Selective Chemical Inhibition of Individual ERK Isoforms Is Achieved through an Inhibitor-
Resistant Approach .............................................................................................................................42
v
5. Selective Inhibition of either ERK1 or ERK2 Is Not Sufficient to Maintain ESC Self-Renewal
in the Presence of GSK3 Inhibitor ......................................................................................................52
6. Blocking ERK1 or ERK2 Kinase Function via a Kinase-Dead Approach Cannot Maintain ESC
Self-Renewal in the Presence of GSK3 Inhibitor ...............................................................................57
7. Selective Inhibition of either ERK1 or ERK2 Cannot Block ESC Differentiation ....................61
8. Selective inhibition of ERK1 but Not ERK2 Promotes the Mesendoderm Differentiation of ESCs
.............................................................................................................................................................66
Discussion ...............................................................................................................................................70
Perspective and Future Directions .......................................................................................................74
Materials and Methods .........................................................................................................................76
Cell Culture ....................................................................................................................................................76
Genome Editing in Mouse ESCs ....................................................................................................................78
Establishment of Erk Transgenic Embryonic Stem Cell Lines ......................................................................82
Blastocyst Injection and Chimera Formation .................................................................................................84
Self-Renewal Assay and AP staining .............................................................................................................85
Embryoid Body-Based ESC Differentiation ..................................................................................................86
In Vitro Monolayer Differentiation of ESCs ..................................................................................................86
Western Blot ...................................................................................................................................................87
Immunostaining ..............................................................................................................................................89
Quantitative Real-Time PCR (qRT-PCR) ......................................................................................................90
Statistical Analysis .........................................................................................................................................91
Bibliography ...........................................................................................................................................93
Part II The Effect of Medium Osmolality on Neural Differentiation of Mouse Embryonic
Stem Cell .................................................................................................................................... 105
vi
Abstract ................................................................................................................................................105
Introduction .........................................................................................................................................106
1. Neural Induction in vivo ...........................................................................................................106
2. Neural induction in vitro: ESC neural differentiation ..............................................................109
3. Signaling Pathways Implicated in ESC Neural Differentiation ...............................................114
4. Implication of Physical Factors in Neural Differentiation .......................................................118
5. Rationale of This Study ............................................................................................................119
Experimental Results ..........................................................................................................................121
1. Addition of NaCl Promotes Neural Differentiation of ESCs Cultured in Neurobasal Medium
with N-2 and B-27
TM
Supplements ...................................................................................................121
2. Addition of Inorganic or Organic Osmolytes Promotes Neural Differentiation of ESCs in
NeuroBasal Medium with N-2 and B-27
TM
Supplements .................................................................124
3. Activation of MAPK/ERK Signaling Is Involved in the Neural Differentiation of ESCs
Promoted by Increased Medium Osmolality .....................................................................................127
4. ERK-Independent Signaling Is Involved in the Neural Differentiation of ESCs Promoted by
Increased Medium Osmolality ..........................................................................................................132
Discussion .............................................................................................................................................135
Perspectives and Future Directions ...................................................................................................138
Materials and Methods .......................................................................................................................139
Cell Culture ..................................................................................................................................................139
Monolayer Neural Differentiation ................................................................................................................140
vii
Chemical Inhibitors and Cytokines ..............................................................................................................140
Osmolality Control and Measurement ..........................................................................................................141
Western Blotting ...........................................................................................................................................141
Flow Cytometry Analysis .............................................................................................................................143
Immunofluorescence Staining ......................................................................................................................144
Statistical Analysis .......................................................................................................................................144
Bibliography .........................................................................................................................................145
viii
List of Figures and Tables
Figure 1.1 The origin and specific markers of mouse ESCs. .......................................................... 3
Figure 1.2 The roles of LIF and BMP signaling in mouse ESCs. .................................................. 6
Figure 1.3 Current understanding of 2i in mouse ESCs. ................................................................ 9
Figure 1.4 Canonical mitogen-activated protein kinase pathways. .............................................. 11
Figure 1.5 Kinase inhibitors for blocking MAPK/ERK signaling. ............................................... 12
Figure 1.6 Comparison between ERK1 and ERK2 in mouse. ...................................................... 16
Figure 1.7 Chemical genetics ........................................................................................................ 22
Figure 1.8 Application of chemical genetics in studying kinase isoform-specific roles. ............. 24
Figure 1.9 Confirmation of the deletion of ERK1/2 in Erk DKO ESCs by immunoblotting. ...... 28
Figure 1.10 Confirmation of the deletion of ERK1/2 in Erk DKO ESCs by DNA sequencing ... 29
Figure 1.11 Erk DKO ESCs can survive in the presence of LIF and serum in the long run. ....... 29
Figure 1.12 Deletion of ERK1/2 does not affect the genome stability of ESCs. .......................... 30
Figure 1.13 Deletion of ERK1/2 enhances ESC self-renewal in the LIF + serum medium. ........ 30
Figure 1.14 Deletion of ERK1/2 delays the expression of lineage-specific markers in ESCs. .... 32
Figure 1.15 Deletion of ERK1/2 impairs mesendoderm differentiation of ESCs. ....................... 33
Figure 1.16 Erk2 WT transgenic ESCs can differentiate in vitro and contribute to chimera
formation in vivo. .......................................................................................................................... 34
Figure 1.17 Erk DKO ESCs can be maintained undifferentiated in the presence of GSK3 inhibitor.
....................................................................................................................................................... 36
Figure 1.19 The combined use of ERK and GSK3 inhibitors maintains ESC self-renewal. ........ 37
Figure 1.18 Erk DKO ESCs retain the self-renewal in the presence of GSK3 inhibitor in the long
run. ................................................................................................................................................ 37
ix
Figure 1.20 The combined use of ERK and GSK3 inhibitors maintains ESC self-renewal in the
long term. ...................................................................................................................................... 38
Figure 1.21 Confirmation of Erk1
-/-
ESCs and Erk2
-/-
ESCs. ....................................................... 40
Figure 1.22 Erk1
-/-
ESCs and Erk2
-/-
ESCs cannot be kept undifferentiated under the CHIR alone
condition. ...................................................................................................................................... 41
Figure 1.23 GSK3 inhibitor alone cannot retain the self-renewal of Erk1
-/-
ESCs and Erk2
-/-
ESCs.
....................................................................................................................................................... 41
Figure 1.25 Illustration of the inhibitor-resistant strategy to achieve individual inhibition of highly
homologous ERK1/2 isoforms. ..................................................................................................... 44
Figure 1.24 The ERK1 AS (ERK1 Q123G) mutant loses the kinase activity in ESCs. ............... 44
Figure 1.26 Confirmation of Erk DKO ESCs expressing ERK mutants. ..................................... 45
Figure 1.27 The Egr1 level in transgenic ESCs is a good readout to evaluate the kinase activity
and drug response of ERK proteins. ............................................................................................. 46
Figure 1.28 The screening on ERK1 IR mutants by Egr1-based qPCR. ...................................... 47
Figure 1.29 The screening on ERK2 IR mutants by Egr1-based qPCR. ...................................... 48
Figure 1.30 The level of pRSK in transgenic ESCs is a good readout to evaluate the relative kinase
activity and drug response of ERK proteins. ................................................................................ 49
Figure 1.31 pRSK-based confirmation of the kinase activities and inhibition responses of ERK1
G55A and ERK2 G35S mutants in ESCs. .................................................................................... 50
Figure 1.32 Erk DKO ESCs is a valuable tool to evaluate the kinase activity of ERK mutants. . 51
Figure 1.33 Erk1 G55A transgenic ESCs and Erk2 G35S transgenic ESCs differentiate under the
Vx11e +CHIR and CHIR alone conditions. ................................................................................. 52
Figure 1.34 Confirmation of Erk IR knock-in ESC lines by DNA sequencing and immunoblotting.
....................................................................................................................................................... 53
x
Figure 1.35 Selective inhibition of ERK1 or ERK2 cannot maintain ESC self-renewal in the
presence of GSK3 inhibitor. ......................................................................................................... 54
Figure 1.36 Confirmation of the expression of ERK proteins in transgenic ESCs established from
the Erk1
-/-
ESCs and the Erk2
-/-
ESCs. .......................................................................................... 55
Figure 1.38 Specific inhibition of ERK1 in the Erk2 IR transgenic ESCs cannot maintain the self-
renewal in the presence of GSK3 inhibitor. .................................................................................. 55
Figure 1.37 Specific inhibition of ERK2 in the Erk1 IR transgenic ESCs cannot maintain the self-
renewal in the presence of GSK3 inhibitor. .................................................................................. 56
Figure 1.38 Specific inhibition of ERK1 in the Erk2 IR transgenic ESCs cannot maintain the self-
renewal in the presence of GSK3 inhibitor. .................................................................................. 56
Figure 1.39 Functional confirmation of ERK kinase-dead mutants in ESCs. .............................. 59
Figure 1.40 Confirmation of the expression of ERK proteins in Erk1
-/-
+ Erk1 KD ESCs and Erk2
-
/-
+ Erk2 KD ESCs. ....................................................................................................................... 60
Figure 1.41 Individual inhibition of ERK1 or ERK2 by the kinase-dead method cannot maintain
ESC self-renewal in the presence of GSK3 inhibitor. .................................................................. 61
Figure 1.42 Individual inhibition of ERK1 or ERK2 does not block neuroectoderm differentiation.
....................................................................................................................................................... 62
Figure 1.43 Selective inhibition of ERK1 or ERK2 does not block mesoderm differentiation. .. 63
Figure 1.44 Selective inhibition of ERK1 or ERK2 does not block endoderm differentiation. ... 64
Figure 1.45 Individual inhibition of ERK1 or ERK2 by a kinase-dead approach does not block the
commitment of ESCs to neuroectoderm and mesendoderm. ........................................................ 65
Figure 1.46 ERK1 and ERK2 play redundant roles in neural differentiation. .............................. 67
Figure 1.47 Individual inhibition of ERK1 but not ERK2 promotes mesoderm differentiation. . 68
Figure 1.48 Individual inhibition of ERK1 but not ERK2 promotes endoderm differentiation. .. 69
Table 1.1 Basal medium composition for mouse ESC culture ..................................................... 76
xi
Table 1.2 Composition of N2B27 medium ................................................................................... 77
Table 1.3 DNA oligonucleotides for genome editing with CRISPR/Cas9 technique .................. 79
Table 1.4 The sequences of primers used in this study. ................................................................ 91
Figure 2.1 Default model of neural induction. ............................................................................ 107
Figure 2.2 Protocols for embryonic stem cell (ESC) neural differentiation. .............................. 110
Figure 2.3 Addition of NaCl accelerates the emergence of Sox1-GFP+ NSCs in the differentiating
ESCs cultured in NB medium. .................................................................................................... 122
Figure 2.5 NaCl dose-dependently promotes the production of Sox1-GFP
+
NSCs from ESCs
cultured in NB medium. .............................................................................................................. 123
Figure 2.4 Addition of NaCl increases the percentages of Sox1-GFP
+
NSCs among the
differentiating ESCs cultured in NB medium. ............................................................................ 123
Figure 2.6 Addition of different salts promotes the differentiation of ESCs into Sox1-GFP
+
NSCs
among the differentiating ESCs cultured in NB medium. .......................................................... 125
Figure 2.7 The addition of different osmolytes increases the percentages of Sox1-GFP
+
NSCs
among the differentiating ESCs cultured in NB medium. .......................................................... 126
Figure 2.8 Addition of glucose promotes the differentiation of ESCs into Sox1-GFP
+
NSCs among
the differentiating ESCs cultured in NB medium. ...................................................................... 127
Figure 2.9 The MAPK/ERK pathway inhibitors suppress the formation of Sox1-GFP
+
NSCs from
differentiating ESCs cultured in NB medium supplemented with NaCl. ................................... 128
Figure 2.10 The MAPK/ERK pathway inhibitors suppress the promoting effect of NaCl-addition
on neural differentiation of ESCs cultured in NB medium. ........................................................ 129
Figure 2.12 Addition of FGF4 promotes the differentiation of ESCs into Sox1-GFP
+
NSCs among
the differentiating ESCs cultured in NB medium. ...................................................................... 131
Figure 2.11 NaCl activates the MAPK/ERK pathway in ESCs cultured in NB medium. .......... 131
xii
Figure 2.13 Increased osmolality by NaCl promotes neural differentiation of ESC cultured in NB
medium via an ERK-independent pathway. ............................................................................... 133
Figure 2.14 The enhancement of neural differentiation by increased osmolality by NaCl is
dependent on the inhibition of BMP signaling. .......................................................................... 134
1
Part I Deciphering the Roles of Erk1/2 Isoforms in Mouse
Embryonic Stem Cell Fate Control
Abstract
Mouse embryonic stem cells (ESCs) are pluripotent stem cells derived from the inner cell mass of
mouse blastocyst. Mouse ESCs can self-renew indefinitely while retaining the ability to
differentiate into different cell types in the body. Mouse ESCs can be maintained and propagated
in vitro by dual inhibition of GSK3 and MEK. MEK has two direct substrates, extracellular signal
regulated kinase 1 and 2 (ERK1/2), which are effector kinases of the MAPK/ERK pathway.
However, the roles of ERK1/2 in the self-renewal and differentiation of mouse ESCs remain ill-
defined. Here, we show that ERK1/2 are dispensable for ESC survival and self-renewal, and that
chemical inhibition or genetic deletion of ERK1/2 is sufficient to maintain ESC self-renewal when
GSK3 is simultaneously inhibited. Genetic deletion of either ERK1 or ERK2, however, is not
sufficient to maintain ESC self-renewal in the presence of GSK3 inhibitor. Similarly, selective
inhibition of individual ERK isoforms with a novel chemical genetic approach cannot maintain
ESC self-renewal in the presence of GSK3 inhibitor. During ESC differentiation, inhibition of
ERK1/2 blocks mesendoderm differentiation but not neuroectoderm differentiation. Selective
inhibition of either ERK1 or ERK2, however, cannot block the lineage commitment of ESCs into
different germ layers. In addition, selective inhibition of ERK1 but not ERK2 is able to promote
ESC differentiation towards mesendoderm. These results suggest that the two ERK isoforms have
both overlapping and non-overlapping roles in regulating ESC self-renewal and differentiation.
This study opens new research avenues for dissecting the roles of individual ERK isoforms in
different cell types and during various cellular processes.
2
Introduction
1. The Origin and Application of Mouse Embryonic Stem Cells
All mammals develop from a single cell, the fertilized egg, which can give rise to any cell type in
the developing embryo, including cells in the placenta (Baker and Pera, 2018). With the continuous
cell division, the fertilized egg can produce the blastocyst containing the inner cell mass (ICM)
and the trophectoderm (TE), marking the first cell lineage specification in the embryo (Niakan et
al., 2013). The ICM and the TE give rise to the embryo and the extraembryonic tissue, respectively.
During implantation of the embryo into endometrium at embryonic day 4.5 (E4.5), ICM cells
undergo the second cell fate specification to generate the hypoblast (primitive endoderm) and
epiblast layers. The hypoblast contributes to the extraembryonic mesoderm and yolk sac, whereas
the epiblast differentiates into the definitive germ layers of endoderm, mesoderm, and ectoderm,
and generates the ultimate embryo.
Mouse embryonic stem cells (ESCs) are derived from the ICM of mouse preimplantation
blastocyst (Nichols and Smith, 2009) (Figure 1.1). Mouse ESCs not only have the ability to self-
renew but also retain the capacity to differentiate into any cell type of the three germ layers. When
transplanted subcutaneously into immunodeficient mice, mouse ESCs can form teratomas in
xenograft models. Mouse ESCs express a group of core pluripotency makers, including Oct4 and
Nanog. Additionally, mouse ESCs express an array of naïve pluripotency markers, including Rex1,
Esrrb, Nr0b1, and FGF4, distinguishing them from another type of pluripotent stem cells, called
epiblast stem cells (EpiSCs) (Nichols and Smith, 2009; Weinberger et al., 2016). ESCs have two
active X chromosomes in female cells and can produce chimeras when injected into a blastocyst.
3
The application of ESCs in regenerative medicine and disease modeling underscores the
importance of studying the self-renewal and differentiation of ESCs (Fox et al., 2014; Lancaster
and Knoblich, 2014). As early as the 1980s, hematopoietic stem cells differentiated from mouse
ESCs have been experimentally used to restore bone marrow cells in lethally irradiated mice
(Hollands, 1988). The potential of ESCs to be infinitely expanded in vitro (Ying et al., 2008) and
the invention of protocols for directed differentiation of ESCs into specific somatic cells (Murry
and Keller, 2008) have sparked tremendous interest in ESC research for the replacement of lost or
abnormal tissues (Fox et al., 2014). This interest has been further cultivated by the successful
reprogramming of mouse somatic cells into induced pluripotent stem cells (iPSCs), the functional
equivalent counterpart of mouse ESCs (Takahashi and Yamanaka, 2006). With the ease of genetic
Figure 1.1 The origin and specific markers of mouse ESCs.
Mouse ESCs are derived from inner cell mass (ICM) in preimplantation blastocyst. Lineage-specific maker(s) is
noted below each cell type. E: embryonic day. Scale bars: 50µm. Information compiled from (Niakan et al., 2013;
Nichols and Smith, 2009).
4
manipulation in vitro and the competency of germline transmission upon blastocyst injection,
ESCs have been used to generate off-springs carrying genetic mutations. These mouse models are
very useful for investigating the function of gene(s), understanding the pathogenesis of diseases,
and carrying out drug screening for therapy development (Kauppila et al., 2016; Zuberi and Lutz,
2016). In particular, ESCs have been utilized to generate animals as surrogate organ donors via
interspecies blastocyst complementation (Wu et al., 2016; Wu et al., 2017), highlighting the
importance and usefulness of ESC research. The expansion of ESCs in vitro provides an
inexhaustible cell source for pathogenesis exploration, large-scale drug screening, and therapeutic
usage in regenerative medicine.
2. Mouse Embryonic Stem Cells: Culture Conditions and Self-Renewal Signaling
Pathways
To fulfill the enormous potential of mouse ESCs in regenerative medicine, we need to establish
proper conditions to robustly expand mouse ESCs without impairing their capacity to differentiate
into different cell types.
In 1981, embryonic stem (ES) cell lines were firstly established from mouse embryos by two
independent groups (Evans and Kaufman, 1981; Martin, 1981). The conditions they used were
similar, as both groups maintained and propagated ESCs on a feeder layer of irradiated murine
embryonic fibroblasts (MEF) in serum medium. Under the condition of serum plus a feeder layer,
mouse ESCs retained a normal euploid karyotype, as well as the capacity to differentiate into
multiple cell types (Doetschman et al., 1985). The pluripotency of these mouse ESC lines was
further confirmed by demonstrating their expression of a group of pluripotency markers (Mitsui et
al., 2003; Rogers et al., 1991; Saijoh et al., 1996) and their ability to contribute to chimera
5
formation after reintroduction of ESCs into blastocysts (Palmiter and Brinster, 1986; Williams et
al., 1988). Therefore, the derived mouse ESC lines are considered to be identical to cells in the
ICM of the embryo.
The endeavor of developing feeder-free culture conditions for ESCs led to the identification of the
cytokine leukemia inhibitory factor (LIF) secreted by feeder cells. LIF that can be used for the
isolation and maintenance of mouse ESCs from the blastocyst in the presence of serum (Smith et
al., 1988; Williams et al., 1988). Serum can also be replaced by bone morphogenetic proteins
(BMPs), so that mouse ESCs can be derived and expanded in a chemically defined medium
supplemented with LIF and BMPs (Ying et al., 2003a). The combination of LIF and BMPs can
preserve ESC differentiation potentials, chimera colonization, and germline transmission
properties. The LIF (or feeder cells) plus serum (or BMPs) conditions have been successfully used
to derive and propagate ESCs from a few mouse strains including the inbred 129 strain (Brook and
Gardner, 1997); however, most mouse strains are recalcitrant to ESC derivation using these
empirical culture conditions (Buehr and Smith, 2003; Ying and Smith, 2017). Additionally, the
LIF and serum (or BMPs) condition has no success in deriving ESCs from species other than the
mouse (Buehr et al., 2008; Li et al., 2008; Thomson et al., 1998).
The molecular mechanisms underlying the roles of LIF and BMP in ESC self-renewal and
maintenance are relatively clear (Figure 1.2). LIF binds to a heterodimeric receptor complex
consisting of the LIF receptor subunit (LIF-R) and the signal transducer gp130, resulting in the
activation of the Janus-associated kinase (JAK). JAK then phosphorylates Stat3, which
subsequently translocates into the nucleus and activate its target genes (Niwa et al., 1998).
Activation of JAK-STAT3 pathway in ESCs by LIF suppresses mesendoderm differentiation.
BMPs bind to the BMP receptor and activate the Smad1/5 dependent pathways. The key
6
downstream targets of Smad1/5 include Id1 and dual-specificity phosphatase 9 (DUSP9), which
can both act to suppress neural commitment (Li et al., 2012; Ying et al., 2003a). Morikawa and
colleagues demonstrated that Smad1/5 knockout ESCs could still be maintained in LIF plus BMP4
(or serum) condition. They found that the MEK5/ERK5 pathway mediates the role of BMP in
retaining pluripotency by inducing the expression of Klf4 (Morikawa et al., 2016). Furthermore,
the inhibitory role of the MEK5/ERK5 pathway in neuroectoderm differentiation has been recently
revealed (Williams et al., 2016), affirming the negative role of BMPs on ESC neural differentiation.
The combination of LIF and BMP block the lineage commitment of ESCs, thus retaining ESC
self-renewal.
Figure 1.2 The roles of LIF and BMP signaling in mouse ESCs.
In mouse ESCs, LIF binds to its receptor and activates Janus-associated kinase (JAK), which subsequently
phosphorylates and activates STAT3. The active STAT3 inhibits the differentiation of ESCs into mesendoderm.
BMPs activate both Smad and MEK5/ERK5 pathways, which suppress neuroectodermal differentiation. The Smad
pathway activates Id1 and dual-specificity phosphatase 9 (DUSP9). DUSP9 can inhibit the activation of ERK
signaling. The synergistic effects of LIF and BMP impede the lineage commitment of ESCs, thus supporting ESC
self-renewal and propagation. ERK5: extracellular regulated kinase 5; LIF: leukemia inhibitory factor; BMP: bone
morphogenetic proteins.
7
The existence of an alternative pathway(s) independent of LIF in the maintenance of pluripotency
has been supported by several lines of evidence. Firstly, the blastocysts with the deletion of LIF,
LIF-R, or gp130 are viable and can develop through gastrulation when the differentiation of three
germ layers take place (Stewart et al., 1992; Ware et al., 1995; Yoshida et al., 1996). In addition,
knock-out of members of the IL-6 cytokine family, including LIF-R and gp130, in embryos did
not impair the derivation and maintenance of ESCs (Dani et al., 1998). In search for the alternative
route(s) for ESC derivation, Burdon et al. found that suppression of ERK signaling, which is
activated by LIF-STAT3 signaling, promotes ESC self-renewal (Burdon et al., 1999). However,
inhibition of ERK signaling alone is not sufficient for long-term propagation of ESCs without the
loss of pluripotency (Wray et al., 2010). The addition of CHIR99021 (CHIR), a GSK3 inhibitor,
supports the clonal expansion of ESCs, and the combination of CHIR with a MEK inhibitor
(PD0325901) is sufficient for the derivation and propagation of germline-competent ESCs (Wray
et al., 2010; Ying et al., 2008). This new culture condition comprising two inhibitors is termed as
“2i”.
The 2i condition effectively maintains mouse ESCs as a homogenous and undifferentiated
population. Most importantly, the 2i condition has been successfully used for deriving germline-
competent ESCs from multiple mouse strains which are recalcitrant to ESC derivation using LIF
plus serum/BMPs conditions (Nichols et al., 2009; Reinholdt et al., 2012) and even from a different
species, the rat (Buehr et al., 2008; Li et al., 2008). Recently, the 2i-based inhibitor cocktail has
been employed in deriving naïve human ESCs, a functional counterpart to mouse ESCs, from
human embryos (Chan et al., 2013; Duggal et al., 2015; Gafni et al., 2013; Theunissen et al., 2014;
Valamehr et al., 2014; Ware et al., 2014), suggesting that the fundamental mechanism underlying
ESC self-renewal might be shared among different species. Therefore, a better understanding of
8
the molecular basis of 2i in mouse ESC maintenance and expansion may assist the derivation of
germ-line competent ESCs from species other than rodents and humans.
The role of GSK3 inhibition in ESC maintenance has been well-defined. CHIR, one of the
components of 2i, is an inhibitor of GSK3 kinases (Ring et al., 2003). GSK3 in mammals has two
highly homologous isoforms, GSK3a and GSK3b (Chen et al., 2017; Doble and Woodgett, 2003).
GSK3 is constitutively active in cells, and generally acts as a negative regulator in signaling
transductions, such as in the Wnt/b-catenin and insulin signaling pathways (Doble and Woodgett,
2003; McManus et al., 2005). Inhibition of GSK3 exhibits both pro-proliferation and anti-
differentiation roles in mouse ESCs. Inhibition of GSK3 by CHIR promotes anabolism in ESCs
through b-catenin independent pathways (Doble and Woodgett, 2003). CHIR has a further role in
inhibiting ESC differentiation by blocking the degradation of cytoplasmic b-catenin, a process
mediated by a complex consisting of GSK3, Axin, and APC (adenomatous polyposis complex)
(Wray et al., 2011). The stabilized b-catenin is accumulated and translocated into the nucleus.
Nuclear b-catenin interacts with Tcf3, abolishing its role in repressing the expression of multiple
transcription factors, including Esrrb, Oct4, and Nanog (Kelly et al., 2011; Martello et al., 2012;
Pereira et al., 2006) (Figure 1.3). Therefore, CHIR promotes mouse ESC self-renewal by
promoting proliferation via enhancement of anabolic processes and by shutting down
differentiation signals mediated by the GSK3/b-Catenin/Tcf3 axis and b-Catenin-independent
pathway(s).
The molecular basis of mouse ESC self-renewal mediated by MEK has not been well-defined.
MEK inhibitor PD0325901 used in 2i inhibits the kinase activity of MEK and decreases the levels
of phosphorylated ERK1 and ERK2, which are the downstream substrates of MEK kinases (Bain
et al., 2007). In general, the blocking of MAPK/ERK signaling (discussed in the next section) by
9
the MEK inhibitor or by genetic modification impedes the transition of ESCs from self-renewal to
differentiation (Kunath et al., 2007). The potential mechanism is that MEK inhibitor induces a
transcriptionally poised state of developmental genes by regulating RNA polymerase II, putting a
brake on ESC differentiation priming (Tee et al., 2014). Additionally, MAPK/ERK signaling
inhibition increases the expression of core pluripotency factors, including Nanog, Klf4, and Klf2,
Figure 1.3 Current understanding of 2i in mouse ESCs.
2i, a combination of GSK3 inhibitor (GSK3i) and MEK inhibitor (MEKi), robustly maintains and propagates
mouse ESCs by preventing the collapse of the pluripotency network. The GSK3 inhibitor blocks the kinase activity
of GSK3 and diminishes the degradation of b-catenin, thereby inducing the accumulation of cytoplasmic b-catenin
and activating b-catenin dependent signaling. b-catenin interacts with Tcf3 and abolishes its suppressive role on
multiple genes in the pluripotency network. The MEK inhibitor inhibits the kinase activity of MEK and declines
the activation of MAPK/ERK signaling, which plays a negative role on pluripotency genes. It was believed that
ERK1/2 were the only substrates of MEK before the discovery of a new MEK substrate, heat shock factor 1
(HSF1). Information compiled from (Tang et al., 2015; Wray et al., 2011).
10
by reducing their degradation, thus promoting ESC self-renewal (Hamazaki et al., 2006; Kim et
al., 2012; Kim et al., 2014; Yeo et al., 2014). However, the critical downstream targets of
MAPK/ERK signaling that mediate mouse ESC self-renewal have not been identified yet.
3. The MAPK/ERK Pathway and Its Inhibitors
Mitogen-activated protein kinase (MAPK) pathways are evolutionarily conserved among
mammals. They are involved in signal transduction from cell surface receptors to the nucleus and
numerous organelles (Morrison, 2012) (Figure 1.4). The MAPK/ERK pathway (also known as the
RAS/RAF/MEK/ERK pathway) is one of the four canonical MAPK pathways that consist of a
core three-tier kinase cascade: MAP3K, MAP2K, and MAPK. The phosphorylation status of the
kinases acts as on-off signals for MAPK signaling transduction, which ultimately controls different
transcriptional and translational events in cells and affects various cellular behaviors.
To turn on the MAPK/ERK pathway, various extracellular stimuli first bind to receptor tyrosine
kinase (RTK) family members (such as the receptor for basic fibroblast growth factor-bFGFR) and
promote their intrinsic tyrosine kinase activity (Dhillon et al., 2007; Pritchard and Hayward, 2013).
Adaptor proteins are recruited to the intracellular domain of RTKs and activate small G proteins,
known as RAS proteins. Active RAS, in turn, activates downstream RAF kinase in the
MAPK/ERK pathway. Next, RAF kinases phosphorylate and activate MEK, which subsequently
activates ERK by phosphorylating the serine/threonine residue in its kinase domain. ERK proteins,
as the effectors of the MAPK/ERK pathway, execute an array of diverse functions by
phosphorylating downstream substrates (e.g., RSK and Elk1) and regulating their expression.
The MAPK/ERK pathway plays a central role in a variety of fundamental cellular processes,
including proliferation, differentiation, migration, survival, apoptosis, etc. (Kim and Choi, 2010;
11
Li and Johnson, 2006; Pritchard and Hayward, 2013). Normal functioning of MAPK/ERK
signaling is critical for embryogenesis and immune responses (Fischer et al., 2005; Hatano et al.,
2003; Pages et al., 1999; Yao et al., 2003). Aberrant activity of MAPK/ERK signaling, either
hyperactivation or loss of function, is involved in many diseases, including congenital defects,
growth retardation, Alzheimer’s disease, immunodeficiency, leukemia and solid tumors, etc.
(Dhillon et al., 2007; Kim and Choi, 2010). Astonishingly, around one-third of cancer patients
Figure 1.4 Canonical mitogen-activated protein kinase pathways.
The four classical mitogen-activated protein kinase (MAPK) pathways respond to various stimuli and mediate the
transmission of an external signal to its target effector proteins via kinase cascades. The MAPK pathways regulate
diverse cellular behaviors. Of note, the signaling pathways are not purely linear, as there are many cross-talks,
both between different MAPK pathways and with other receptor tyrosine kinase (RTK)-stimulated pathways.
Adapted from (Pritchard and Hayward, 2013).
12
have a mutation in the components of the MAPK/ERK pathway. For instance, K-RAS with a G12V
mutation is frequently expressed in pancreatic cancer, and B-RAF with a V600E mutation is
prevalent in melanoma (Dhillon et al., 2007). These mutations in the MAPK/ERK pathway result
in its constitutive activation, facilitating hyper-proliferation, metastasis, and drug resistance of
tumor cells.
Figure 1.5 Kinase inhibitors for blocking MAPK/ERK signaling.
The kinase inhibitors in the figure above target the MAPK/ERK pathway at different tiers and are classified into
three types based on their targets: RAF inhibitors, MEK inhibitors, and ERK inhibitors. All inhibitors can bind to
the targeted kinases and suppress the intrinsic kinase activity, thereby inhibiting the phosphorylation of downstream
targets. Additionally, some kinase inhibitors (such as SCH772984, GDC-0623, and RO5126766) can also induce
the conformational changes in their targets and prevent the phosphorylation of the kinase itself by the upstream
kinases. Of note, the negative feedback from ERK1/2 on RAS and RAF auto-regulates the upstream kinase activity.
Therefore, the enhanced phosphorylation of upstream kinases in the presence of a kinase inhibitor is commonly seen
in immunoblotting. Compiled information from (Samatar and Poulikakos, 2014).
13
To treat diseases caused by abnormal activity, hyperactivation in most cases, of the MAPK/ERK
pathway, a series of chemical inhibitors have been developed to target the MAPK/ERK pathway
(Bain et al., 2007; Samatar and Poulikakos, 2014) (Figure 1.5). These kinase inhibitors include
RAF inhibitors, MEK inhibitors, and ERK inhibitors, and have been used to treat diseases such as
cancers. However, due to the high heterogeneity among tumor cells, the tumor clone resistant to
the kinase inhibitors will gain growth advantage and lead to relapse (Pritchard and Hayward, 2013).
Of note, in the kinase cascades of the MAPK/ERK pathway, each type of kinase has at least two
highly homologous isoforms. For instance, RAF proteins include A-RAF B-RAF, and C-RAF, and
ERK proteins are expressed in the forms of ERK1 and ERK2 (Pritchard and Hayward, 2013).
There is high homology in the kinase domains of different kinase isoforms from the same family,
and as most inhibitors target the kinase domain, most chemical inhibitors block the kinase activity
of these isoforms indiscriminately (Bain et al., 2007).
4. The Role of MAPK/ERK Pathway in ESC Self-Renewal
Pharmacological inhibition and genetic deletion approaches have been used to explore the role of
MAPK/ERK signaling in ESCs. Traditionally, the MAPK/ERK pathway has been considered to
be dispensable for ESC maintenance. Blocking MEK/ERK signaling with MEK inhibitors
(PD184352 and PD098051) does not impair the proliferation and propagation of wild-type and
Erk2 knock-out (Erk2
-/-
) mouse ESCs (Burdon et al., 1999; Yao et al., 2003). Additionally, the
inhibition of MAPK/ERK signaling with a MEK inhibitor (PD98059) or via the upregulation of
DUSP9 promotes ESC self-renewal (Buehr and Smith, 2003; Li et al., 2012). As such, MEK
inhibition has been employed in enhancing iPSC reprogramming and converting mouse EpiSCs to
mouse ESCs (Huh et al., 2018; Vrbsky et al., 2015).
14
Recently, two research groups manipulated MAPK/ERK signaling by either knocking down Raf-
A/B/C or knocking out ERK1/2 and revealed essential roles of the MAPK/ERK pathway in
promoting mouse ESC proliferation and survival (Chen et al., 2015; Guo et al., 2013). The above
conclusion was further confirmed by applying a different MEK inhibitor, U-0126, in the ESC
culture medium. Of note, Chen et al. reported that deletion of both ERK1 and ERK2 in mouse
ESCs impeded cell cycle progression, compromised the expression of pluripotency factors, and
led to genomic instability (Chen et al., 2015). As a result, ESC lines with depletion of both ERK1
and ERK2 cannot be maintained. The inconsistency with regards to the role of the MAPK/ERK
pathway in ESCs among different reports may be related to 1) the different genetic backgrounds
of ESCs used; 2) the different culture conditions; 3) variation in the levels of MAPK/ERK
signaling manipulated by chemical inhibitors and gene knock-down or knock-out; 4) the intrinsic
complexity of the MAPK/ERK pathway; and 5) differences in the duration of observation.
5. The Role of MAPK/ERK Pathway in ESC Differentiation
The role of MAPK/ERK signaling in ESC differentiation is complicated, as it is stage-specific and
cell-type dependent. Specifically, MAPK/ERK signaling plays distinct roles in ESC differentiation
into epiblast and primitive endoderm and in ESC differentiation into the three germ layers (Greber
et al., 2010; Hamilton and Brickman, 2014; Kunath et al., 2007).
To begin with, the activation of the MAPK/ERK pathway is essential for specification and
segregation of ICM cells into epiblast and hypoblast (primitive endoderm) (Bessonnard et al.,
2014). In vivo, FGF4 is the primary determinant and activator of MAPK/ERK signaling during
embryo development. Knock-out of Fgf4 or Grb2 (an adaptor in the MAPK/ERK pathway) in
embryos leads to the failure of Gata6 expression and of primitive endoderm formation (Chazaud
et al., 2006; Kang et al., 2013); in contrast, Nanog is expressed in all ICM cells with the deletion
15
of FGF4 and Grb2 at the time of implantation (E4.5). In vitro, MEK inhibition completely
abrogates the induction of primitive endoderm markers (e.g., Gata6, Sox17, and Pdgrfa) in ESCs
and the differentiation of ESCs into primitive endoderm (Hamilton and Brickman, 2014).
Conversely, the activation of MAPK/ERK signaling in ESCs suppresses their ability to self-renew,
promotes primitive endoderm priming, and compromises epiblast formation.
MAPK/ERK signaling also plays a crucial role in triggering the transition of ESCs from self-
renewal to differentiation of ESCs into cells of the ectoderm, mesoderm, and endoderm (Kunath
et al., 2007). FGF-dependent MAPK/ERK signaling is required for ESCs to differentiate beyond
the epiblast stage. Knock-out of Fgf4 (Fgf4
-/-
) in ESCs forces them to stay undifferentiated after
LIF withdrawal and experience difficulty while differentiating into neural and mesodermal
lineages (Kunath et al., 2007). In addition, MAPK/ERK signaling is necessary for mesendoderm
differentiation from ESCs. Erk2
-/-
mice have no visible mesoderm at E6.5 (Yao et al., 2003),
though Erk2
-/-
ESCs express brachyury (a mesodermal marker) and FoxA2 (an endodermal marker)
during differentiation (Hamilton et al., 2013; Yao et al., 2003). However, the application of a MEK
inhibitor to the differentiation medium abolishes the expression of brachyury and FoxA2 in
differentiating ESCs, underscoring the importance of MAPK/ERK signaling in mesendoderm
differentiation (Hamilton et al., 2013; Yao et al., 2003).
6. ERK1 and ERK2-Similarities and Differences
ERK1 and ERK2 lie at the bottom tier of the kinase cascades in the MAPK pathway and act as
effector kinases of MAPK/ERK signaling (Morrison, 2012; Pritchard and Hayward, 2013;
Roskoski, 2012). ERK1 and ERK2 exert their diverse functions by phosphorylating different
cytosolic and nuclear substrates. In fact, the substrates phosphorylated by ERK1 and ERK2 are the
16
ones most responsible for the specificity of the cellular responses to extracellular stimuli via the
MAPK/ERK pathway (Barsyte-Lovejoy et al., 2002; Murphy and Blenis, 2006).
In mammals, ERK1 and ERK2 are the evolutionarily conserved products of two different genes,
Erk1 (Mapk3) and Erk2 (Mapk1), located at two different chromosomes (Busca et al., 2015). It is
believed that the emergence of two different ERK proteins, ERK1 and ERK2, is related to specific
whole genome duplication (WGD) in the vertebrates. ERK1 and ERK2 are expressed in all
Figure 1.6 Comparison between ERK1 and ERK2 in mouse.
(A) Schematic representation of the structures of ERK1 and ERK2 in mouse. (B) Primary amino acid sequence
comparison of ERK1 and ERK2 in mouse (with ClustalX v2.0). Amino acid numbering is based on the ERK1
sequence. The ERK2 sequence appears below the ERK1 sequence with shared amino acids denoted by asterisks
(*). The colon (:) and period (.) indicate the conservation of amino acids with high and low similar properties,
respectively. The colored letters (red, blue, and green) are the mutation sites used in this study. The letters in
lavender denote the activation loop. MEK1/2 phosphorylate the threonine (T) and tyrosine (Y) (in sky-blue) in
ERK1/2 and activate ERK kinase activity.
17
vertebrates except frog and avian species, which only have ERK2 (Busca et al., 2015). The
genome size of Erk2 is much larger than that of Erk1, which is much densely packed. Despite the
large difference in genome size, mouse ERK1and mouse ERK2 are more than 80% identical in
their overall amino acid sequences (Figure 1.6) (Roskoski, 2012) and are similar in the molecular
weight: ERK1 is 44 kDa, and ERK2 is 42 kDa.
Similar to other kinases, ERK1 and ERK2 consist of a catalytic kinase domain wrapped by
regulatory stretches at the N-terminal and C-terminal and also have a unique insertion within the
kinase domain (Roskoski, 2012). As mentioned in the section of the MAPK pathway, ERK1 and
ERK2 are activated by the MAP2Ks (MEK1 and MEK2) (Morrison, 2012; Roskoski, 2012), which
carry out dual phosphorylation on the Thr and Tyr residues of ERK proteins located within the
Thr-Xaa-Tyr motif, also known as the activation loop (Figure 1.6B). Full activation of ERK
proteins relies on the protein-protein interactions between MEK and ERK mediated by different
domains, such as the common docking (CD) domain in the C-terminal of ERK proteins and the D
domain (three basic and two hydrophobic residues) in the N-terminus of MEK proteins. The
protein-protein interactions between MEK and ERK induce the conformational changes in ERK
proteins, exposing the Tyr and Thr residues on the activation loop for their phosphorylation and
activation. Activated ERK proteins become potent Ser/Thr kinases, which phosphorylate
substrates on their consensus sequences, Pro-Xaa-Ser/Thr-Pro. Many ERK substrates interact with
ERK proteins via particular docking domains, such as the D domain in MEK. Another well-known
domain mediating the interaction with ERK is a DEF (docking site for ERK, FXF) site that
interacts with a DEF recruiting site (DRS) in ERK proteins (Barsyte-Lovejoy et al., 2002;
Roskoski, 2012).
18
Because of the high homology between ERK1 and ERK2, especially in their activation loops, CD
domains, and DRS sites, it is believed that ERK1 and ERK2 are functionally redundant and termed
as ERK1/2 (Busca et al., 2016; Roskoski, 2012; Saba-El-Leil et al., 2016). In fact, all known
extracellular stimuli activating the MAPK/ERK pathway lead to the parallel activation of both
ERK isoforms (Robbins et al., 1993). Furthermore, ERK1/2 have the identical substrate specificity
in vitro and share the very similar 3D structures (Busca et al., 2015; Lefloch et al., 2008; Robbins
et al., 1993). No ERK isoform-specific interacting partners and substrates have been reported yet
(Busca et al., 2015; Shin et al., 2018; von Kriegsheim et al., 2009), which may be due to their
nearly indistinguishable substrate interacting sites where 22 out of the 23 residues are identical
(Busca et al., 2015).
Despite the potential functional overlapping between ERK1 and ERK2, massive efforts have been
invested in defining the reason for the existence of two different ERK isoforms in mammals. The
different outcomes of Erk1
-/-
mice (viable) and Erk2
-/-
mice (lethal at E8.5) have sparked interest
in the search of non-redundant roles of ERK1/2 isoforms (Hatano et al., 2003; Pages et al., 1999;
Saba-El-Leil et al., 2003; Yao et al., 2003). The functional differences in the isoforms of MAP3K
and MAP2K provide the clues for establishing the differences in ERK1/2 isoforms. For instance,
at the MAP3K level, B-RAF has the strongest binding affinity to Ras compared with A-Raf and
C-Raf, which underscores its high frequency of mutation seen in different types of tumors
(Desideri et al., 2015). Additionally, MEK1 rather than MEK2 can be specifically phosphorylated
by ERK proteins at Thr292, the residue absent in MEK2, and can negatively regulate the amplitude
and duration of MAPK/ERK signaling (Catalanotti et al., 2009).
In light of these differences in Raf and MEK proteins, Marchi et al. revealed that compared with
ERK2, ERK1 has slower nucleocytoplasmic shuttling kinetics due to an extra 20-amino acid
19
insertion in its N-terminal (Marchi et al., 2010). By swapping the N-terminals of ERK1/2 isoforms,
the roles of ERK1 and ERK2 in cell transformation are exchanged (Marchi et al., 2008),
highlighting the sequence basis of their functional differences. The differences in the activation
kinetics of ERK1/2 isoforms may account for their functional differences as well. In support of the
above hypothesis, for example, thrombin specifically induces the activation of ERK2 rather than
ERK1 in platelets (Papkoff et al., 1994), and IGF-1 preferentially activates ERK2-dependent
differentiation in myoblasts (Sarbassov et al., 1997).
To better identify the functional differences between the two ERK isoforms, their respective
expression levels have been manipulated genetically by morpholinos, shRNA/RNA interference
(RNAi), or knock-out approaches, and the phenotypes of interest have been characterized in
different genetically engineered models. Non-redundant roles of ERK1/2 isoforms have been
revealed in embryogenesis and different types of somatic cells.
Morpholinos (MO), a type of short antisense oligonucleotides interfering with the transcription by
binding to transcription start sites or pre-mRNA splicing sites, have been used in knocking down
genes in zebrafishes and frogs (Corey and Abrams, 2001). By employing morpholinos of Erk1 and
Erk2 in zebrafish embryos, Krens et. al. found that ERK1 and ERK2 regulate different sets of
genes in embryogenesis involved in BMP, FGF, Nodal, and Wnt pathways, and identified a group
of ERK isoform-specific downstream targets (Krens et al., 2008a). Notably, genes oppositely
regulated by the two ERK isoforms, such as cfos and Vegf, have been uncovered. The differences
in the downstream targets and regulatory effects between ERK1 and ERK2 may account for the
functional differences between ERK1/2 isoforms in zebrafish gastrulation (Krens et al., 2008b).
The re-expression of ERK2 into Erk1 MO zebrafish repaired the defects in convergence migration,
whereas the re-expression of ERK1 into Erk2 MO zebrafish could not compensate for the defects
20
in anterior-posterior extension during gastrulation and epiboly initiation. These results
unequivocally demonstrate the functional differences between ERK1 and ERK2 in the zebrafish.
The RNAi and knock-out approaches have also been utilized to manipulate the expression of ERK
isoforms in cell lines and mice to dissect the functions of ERK1/2 isoforms. It has been shown that
ERK1 plays specific roles in adipogenesis, T cell maturation and Th2 differentiation, splenic
erythropoiesis, cisplatin-induced apoptosis of hepatocellular cancer cells, survival of hepatocyte,
and memory gain (Bost et al., 2005; Fremin et al., 2012; Goplen et al., 2012; Guegan et al., 2013;
Guihard et al., 2010; Mazzucchelli et al., 2002; Pages et al., 1999). In contrast, ERK2 plays
essential roles in placenta development, mesoderm differentiation, protection from myocardium
infarction, epithelial-mesenchymal transition and migration of tumor cells, and nociception
(Hatano et al., 2003; Lips et al., 2004; Otsubo et al., 2012; Saba-El-Leil et al., 2003; Shin et al.,
2010; von Thun et al., 2012; Yao et al., 2003). Interestingly, ERK1 and ERK2 have opposite roles
in the regulation of RAS-dependent cell proliferation and transformation (Vantaggiato et al., 2006):
ERK1 inhibits proliferation and transformation by suppressing overall MAPK/ERK signaling;
ERK2 boosts the output of the MAPK/ERK pathway and positively contributes to cell growth and
tumor formation. These results have demonstrated that ERK1/2 isoforms have non-overlapping
roles in different cellular processes and behaviors.
However, caution should be exercised when explaining the functional differences between ERK1
and ERK2, as these differences could possibly arise from their differential expression levels rather
than inherent differences in the proteins themselves. In fact, Lefloch et al. showed that the overall
expression levels of ERK1 and ERK2 are what determine cell proliferation rate and mouse survival
(Lefloch et al., 2008). The functional differences between the ERK1/2 isoforms caused by their
differential expression levels are further bolstered by the findings that embryonic lethal Erk2
-/-
21
mice can be rescued when a very high level of ERK1 is re-expressed in Erk2
-/-
embryos (Fremin
et al., 2015). It would thus be more advisable to utilize an add-back strategy to see if one isoform
can compensate for the role of the other while keeping the expression of the two ERK isoforms at
similar levels. Furthermore, other regulatory elements surrounding the genomic regions of ERK1
and ERK2 may account for the functional differences. To address this, we may need to generate
knock-in mice in which the expression of ERK1 is driven by the endogenous Erk2 promoter and
vice versa. Last but not least, RNAi or other gene disruption methods eliminate both protein
scaffold and catalytic activity of kinases at the same time, causing the kinase-independent
functions of ERK1/2 proteins to be overlooked. This is important for ERK1/2, as their kinase-
independent roles have been revealed (Hong et al., 2009; Rauch et al., 2011).
7. Chemical Genetics: A Useful Tool to Decipher Kinase Isoform-Specific Roles
Chemical genetics is a powerful toolkit for probing the function of proteins and signal transduction
pathways in cells and organisms by coupling conventional genetics with chemical compounds
(Kawasumi and Nghiem, 2007). In chemical genetics, small molecule libraries are screened on a
model system, and the phenotype of interest is analyzed (Runcie et al., 2016). Based on the intent
of the chemical screening, chemical genetics can be divided into two types (Figure 1.7): forward
chemical genetics and reverse chemical genetics. In forward chemical genetics, diverse libraries
of small molecules are screened on a model system to identify the compounds eliciting the
phenotype of interest and to subsequently define the proteins targeted by the chemicals.
Alternatively, in reverse chemical genetics, a protein is first chosen, and the chemical library is
then screened to identify potent and selective inhibitors, followed by uncovering of the function(s)
of the protein based on the phenotypic observation. Of note, in chemical genetics, the change in
22
the phenotype of interest is caused by the exposure to chemical compounds rather than by the
introduced mutations.
Chemical genetics is advantageous in several aspects in comparison with conventional genetics,
which relies on mutagenesis and knock-down or knock-out approaches (Alaimo et al., 2001;
Runcie et al., 2016). Firstly, the removal of proteins by knock-out or knock-down may lead to the
lethality in developing embryos, resulting in inability to study the phenotype. This problem may
be obviated by a conditional knock-out approach with the help of inducible systems based on drugs
or heat. However, the utilization of drugs or heat may interfere with the phenotype of interest,
Figure 1.7 Chemical genetics
The combination of chemical drugs with genetics for studying the phenotype(s) in a model system defines chemical
genetics. Similar to conventional genetics, chemical genetics can be classified into two types: forward and reverse
approaches. In the forward approach, a drug screening is performed in a model system first, and then the drug
target(s) (proteins in most cases) can be uncovered by identifying the drug binding partners. In the reverse approach,
a target and its agonist/antagonist are chosen first, and then the function of the target can be determined by exploring
the phenotype in a model system (Modified based on (Runcie et al., 2016)).
23
confounding data analysis. Additionally, the phenotype of interest may be masked due to
compensation from other proteins that share the high degree of homology with our targeted protein.
In contrast, the application of chemical drugs to manipulate protein activity through a chemical
genetics approach will preserve protein stoichiometry and thus clearly define the function of the
targeted protein. Furthermore, the ease of administering drugs provides the possibility of revealing
the roles of targeted proteins in a stage-specific and dose-dependent manner in multiple model
systems.
Chemical genetics has achieved great success in dissecting the functions of proteins and signal
pathways (Chen et al., 2017; Chen et al., 2005). To inhibit a kinase using chemical genetics, a
point mutation (either Glycine or Alanine) is first introduced at the “gate-keeper” position of the
kinase, which enlarges the pocket of accommodating ATPs (Figure 1.8A and 1.8B). The expanded
ATP binding pocket, rarely seen in wild-type kinases in vivo, allows the engineered kinase mutant
to be specifically inhibited by an analog inhibitor, the modified derivative of pyrazolo [3,5-d]
pyrimidine-based kinase inhibitor (PP1) (Bishop et al., 2000). This approach is also termed as the
“bump-hole” approach because of the match between the inhibitor with a bulky side group (called
bump inhibitor or analog inhibitor) and the kinase mutant with a smaller gatekeeper residue
(named analog-sensitive mutant, AS mutant). Owing to the evolutionary conservation of the “gate-
keeper” position in kinases, the bump-hole approach can be widely used for the selective inhibition
of kinases in model systems, either cells or organisms, and the determination of their kinase-
dependent functions.
Importantly, chemical genetics gives a unique competitive edge in deciphering the individual roles
of highly homologous kinases, given that conventional chemical inhibitors designed for targeting
the kinase domains in most cases, if not all, cannot discriminate between the paralogs with high
24
levels of homology (Alaimo et al., 2001; Kawasumi and Nghiem, 2007). Different paralogs are
individually engineered and expressed in multiple model systems, and the application of an analog
inhibitor will individually inhibit one paralog at a time and help define its role in the phenotype of
interest. Of note, one of the preconditions for using the bump-hole approach to decipher the
catalytic role of a kinase is that the kinase activity is still well preserved after site-directed
mutagenesis at the gate-keeper residue. However, some kinases (~30%) are not able to tolerate this
modification and their mutants lose most of kinase activity (Garske et al., 2011). To address the
loss of kinase activity in these kinase mutants, a second-site suppressor strategy may be employed
to restore its catalytic activity (Zhang et al., 2005), which can still be repressed by the analog
inhibitor. With the help of this new strategy, the catalytic role of intolerable kinases can be
determined. Another potential problem associated with the bump-hole approach is a possible off-
Figure 1.8 Application of chemical genetics in studying kinase isoform-specific roles.
(A) Highly homologous isoforms are simultaneously inhibited by conventional inhibitors, rendering it impossible
to decipher the role of each isoform. (B) The bump-hole approach for selectively inhibiting one isoform while
sparing the others. An analog-sensitive (AS) mutant is produced by mutagenesis at the gate-keeper position of a
kinase to expanding its kinase domain, and a bump inhibitor is used to inhibit the AS mutant. (C) The Ele-Cys
approach for the selective inhibition of one kinase isoform. An electrophile-sensitive (ES) mutant is generated by
cysteine replacement, and an electrophilic inhibitor is employed for its repression. Information compiled from
(Kung et al., 2017; Runcie et al., 2016).
25
target effect of the analog inhibitors, especially when their working concentration for complete
inhibition is high (Bain et al., 2007).
To address the off-target effect and other issues of the bump-hole approach, an alternative
chemical-genetic approach has been developed (Garske et al., 2011). In the new chemical genetic
strategy, an electrophile-sensitive (ES) mutant is created by introducing a cysteine point mutation
at the gate-keeper position, and an electrophilic inhibitor is used for inhibiting the kinase activity
of the ES mutant (Figure 1.8C). Because of the combination of an electrophilic inhibitor and a
cysteine mutation in the ES mutant, this new method is termed as the Ele-Cys approach (Kung et
al., 2017). The Ele-Cys approach is better than the bump-hole method in that the ES mutant with
a cysteine point mutation can better maintain the geometry of its ATP-binding site and thereby
kinase activity, and an electrophilic inhibitor has much fewer off-targets than the PP1 inhibitors
used in the bump-hole approach (Garske et al., 2011). Therefore, the Ele-Cys approach greatly
expands the scope of chemical genetics in dissecting the role of kinases. Furthermore, a
combination of the bump-hole method and Ele-Cys approach together in the same model system
allows for the orthogonal inhibition of two different kinases, thus avoiding the laborious task to
establish two different model systems. It also enables the study of the interactions between two
distinct signal pathways in the model system.
8. Rationale of Current Study
The inhibition of MAPK/ERK signaling by a MEK inhibitor, PD0325901, promotes ESC self-
renewal and is sufficient for long-term ESC maintenance when combined with a GSK3 inhibitor.
Conversely, the activation of MAPK/ERK signaling is critical for ESC differentiation. ERK1 and
ERK2 have been considered as the only substrates of MEK kinases, a claim challenged by a report
asserting that heat shock factor 1 is also a direct substrate of MEK. The newly identified substrate
26
of MEK suggests that MEK may have other unknown substrates and exert ERK-independent roles
in ESCs. In 2i medium, the inhibitor for suppressing the MAPK/ERK pathway is a MEK inhibitor,
which blocks the action of active MEK on downstream targets. Whether the inhibition (gene
disruption or protein inhibition) of ERK proteins is sufficient for ESC maintenance in the presence
of GSK3 inhibitor is unknown. In addition, the role of individual ERK1 and ERK2 in ESC
differentiation is ill-defined.
Moreover, the ERK proteins in the MAPK/ERK pathway exist in two isoforms, ERK1 and ERK2,
sharing 85% identical in their overall amino acid sequences. Despite their high levels of homology,
ERK1 and ERK2 possess distinct cellular functions in embryogenesis, as evidenced by the
different outcomes in Erk1
-/-
mice and Erk2
-/-
mice. These results suggest that ERK1 and ERK2
may have different roles in the self-renewal and differentiation of mouse ESCs. One difficulty in
dissecting the ERK isoform-specific functions is that there is no chemical inhibitor for specifically
inhibiting one of the ERK1/2 isoforms. In a recent paper from our lab, we took advantage of the
bump-hole approach and achieved isoform-specific inhibition of glycogen synthase kinase
3a (GSK3a) or GSK3b (Chen et al., 2017). We revealed that high-homologous GSK3a and
GSK3b had distinct roles in ESC self-renewal and neural differentiation. Furthermore, we found
that knock-out of and chemical inhibition of individual GSK3a/b isoforms in ESCs exhibited
different phenotypes, which was related to the compensation between GSK3a and GSK3b in the
knock-out approach (Chen et al., 2017). Therefore, it is necessary to study in the individual role
of ERK1 and ERK2 in mouse ESCs using both the knock-out approach and the chemical-genetic
method.
Our study on ERK1/2 in ESCs will not only reveal the individual role of ERK1 and ERK2 in ESC
fate control, but also open the avenue to explore the ERK isoform-specific functions in different
27
cell types and various cellular processes. Furthermore, a better understanding of ERK isoform-
specific functions will enrich our understanding of MAPK/ERK signaling in diseases and provide
the rationale for the development of ERK isoform-specific drugs to treat diseases.
28
Experimental Results
1. Establishment and Characterization of ERK1/2 Knockout ESCs
To determine the roles of ERK1 and ERK2 (ERK1/2) in ESC fate control, I generated ESCs in
which ERK1/2 were deleted (Erk DKO ESCs) using CRISPR/Cas9 technology. The embryonic
stem (ESC) line I used was the “46c” cell line, derived from the 129 strain of mouse (Ying et al.,
2003b). A total of 33 clones were picked in the pool of candidate Erk DKO ESCs, and seven of
them showed complete deletion of ERK1/2 as confirmed by immunoblotting results. I randomly
selected three clones (c1, c2, and c3) for our further study (Figure 1.9). DNA sequencing of the
three clones on Erk1/2 genomic regions confirmed the knock-out of both genes (Figure 1.10).
I first characterized the survival and genome stability of these Erk DKO ESCs under the condition
of LIF with serum (LIF + serum). All three clones of Erk DKO ESCs were alive and flat-shaped,
which were similar to wild-type (WT) ESCs. However, the Erk DKO ESCs were more uniform
with fewer differentiated cells of irregular sizes when compared with WT ESCs, indicating their
more undifferentiated state (Figure 1.11A). To test whether Erk DKO ESCs can survive in the long
run, I continuously passaged the cells and cultured them in LIF + serum medium for ten passages.
Figure 1.9 Confirmation of the deletion of ERK1/2 in Erk DKO ESCs by immunoblotting.
Western blot analysis of ERK1 and ERK2 in the three chosen clones of Erk DKO ESCs. WT: wild-type; DKO: knock-
out of both Erk1 and Erk2. ERK1: 44 kDa, Erk2: 42 kDa.
29
These cells were able to stay alive and keep their normal cell morphology (Figure 1.11B). We also
Figure 1.10 Confirmation of the deletion of ERK1/2 in Erk DKO ESCs by DNA sequencing
Two guide RNAs (gRNA) targeted the genomic regions of Erk1 and Erk2, respectively. Genotyping PCR was
performed to amplify the Erk1/2 genomic regions covering the two guide RNAs targeted DNA sequences. The
premature stop codons are in red; the insertion is denoted in blue; the deletion of bases is labeled with the hyphen
(-).
Figure 1.11 Erk DKO ESCs can survive in the presence of LIF and serum in the long run.
(A) Representative phase contrast images of WT ESCs and the Erk DKO ESCs (three clones) (2
nd
passage, p2)
cultured under the condition of LIF plus serum. Scale bar, 100µm. (B) Representative phase contrast images of the
Erk DKO ESCs (c1) cultured under the condition of LIF plus serum at p5 and p10. The other two clones of the Erk
DKO ESCs could also survive after ten passages as well. Scale bar, 100µm.
30
performed karyotyping on these long-termly cultured Erk DKO ESCs. Results showed that the
Figure 1.12 Deletion of ERK1/2 does not affect the genome stability of ESCs.
(A) The number of cells with a normal karyotyping (chromosome No.=40) among the 50 chosen cells of each ESC
line. The 50 cells were randomly selected under a microscope and the numbers of chromosomes were counted. (B)
Representative images of chromosome staining results of WT and two clones of Erk DKO ESCs (p12). Scale bar,
200µm.
Figure 1.13 Deletion of ERK1/2 enhances ESC self-renewal in the LIF + serum medium.
(A) qPCR analysis of the expression of pluripotency markers (Rex1, Oct4, and Nanog) in different ESC lines under
the condition of LIF plus serum. WT ESCs cultured in the LIF + serum medium was used as the control. The data
represent three independent experiments and are shown as mean ± SEM. *: p< 0.05; ***: p<0.001. (B)
Representative phase contrast and immunofluorescent staining images for pluripotency markers (Rex1, Oct4, and
Nanog) in WT and the Erk DKO ESC lines maintained under the condition of LIF plus serum. The percentage of
Nanog negative (Nanog-) cells were quantified with CellProfiler (details in Materials and Method). The group of
WT ESCs was used as the control for two tailed Student’s t-tests. The data represent two independent experiments
and are shown as mean ± SEM. ***: p<0.001.Scale bar, 100µm.
31
majority of the Erk DKO ESCs had a normal chromosome number, suggesting that their genome
remained stable in the absence of ERK1/2 (Figure 1.12).
It is well-known that inhibition of MAPK/ERK signaling promotes ESC self-renewal and that
activation of MAPK/ERK signaling primes ESC differentiation (Kunath et al., 2007; Ying et al.,
2008). Next, I investigated the effects of ERK1/2 ablation on ESC self-renewal and differentiation,
two fundamental features of pluripotent stem cells. In terms of self-renewal ability, I profiled the
naive pluripotency markers (Rex1, Nanog, and Oct4) by real-time quantitative PCR (qPCR) in Erk
DKO ESCs. Erk DKO ESCs expressed a similar level of Oct4 but higher levels of Rex1 and Nanog,
compared with WT ESCs (Figure 1.13A). Moreover, the expressional levels of Rex1 and Nanog
in Erk DKO ESCs were comparable to those in WT ESCs cultured in the presence of the MEK
inhibitor PD0325901 (PD), suggesting that suppression of ERK signaling promotes ESC self-
renewal. In addition, I found that Erk DKO ESCs had significantly lower percentages of Nanog
negative cells by immunostaining, indicating an enhanced self-renewal ability (Figure 1.13B).
To investigate the role of ERK1/2 in ESC differentiation, we carried out embryoid body (EB)
differentiation assays in Erk DKO ESCs. We used a 4-/4+ retinoic acid-based EB differentiation
protocol and collected the EB samples on day 4 and day 8 over the course of EB formation. qPCR
analysis was performed on one naïve pluripotency marker (Rex1), and three lineage-specific
markers (Sox1 for neuroectoderm, Brachyury for early mesendoderm, and FoxA2 for endoderm)
to assess differentiation status of cells in the EBs. On both day 4 and day 8, Erk DKO ESCs had
significantly higher levels of Rex1 but significantly lower levels of the three lineage markers, when
compared with WT ESCs (Figure 1.14). However, these high levels of Rex1 and low levels of Sox1
or Brachyury could not be maintained during the course of EB formation in the Erk DKO ESCs.
32
These results suggested that Erk DKO ESCs might preserve the ability to differentiate, but their
differentiation took place at a much slower rate.
To further confirm the differentiation potentials of Erk DKO ESCs, we collected the day 8 EBs
and re-plated them to induce their differentiation into mature cell types of different germ layers. I
found that WT ESCs readily differentiated into mature cells of both neuroectoderm and
mesendoderm, as revealed by the strongly positive staining of lineage-specific markers (Figure
Figure 1.14 Deletion of ERK1/2 delays the expression of lineage-specific markers in ESCs.
qPCR analysis of the expression of a pluripotency maker (Rex1) and three lineage-specific markers (Sox1 for
neuroectoderm, Brachyury for mesoderm, FoxA2 for endoderm) in the embryoid bodies of the indicated cell lines
collected on day 4 and day 8. The data represent two independent experiments and are shown as mean ± SEM.
The group of WT ESCs was used as the control for student’s t tests. *: p< 0.05; ***: p<0.001.
33
1.15). In contrast, although Erk DKO ESCs could differentiate into TUJ1+ (neuronal lineage
marker) cells, there were no cells positive for mesendodermal markers (Myosin or GATA4) in the
differentiating Erk DKO ESCs. These results indicate that Erk DKO ESCs preserve the ability of
differentiating into neuroectoderm, but lose their ability to differentiate into cells of mesendoderm
(Figure 1.15).
We next investigated whether the inability of Erk DKO ESCs to differentiate into mesendoderm
was reversible. We re-expressed ERK2 WT proteins in the Erk DKO ESCs by a transgenic method.
The differentiation potentials of the transgenic ESCs were assessed both in vitro and in vivo. Since
we wanted to perform the chimera formation assay and ERK2 WT was necessary for the survival
of embryo (Hatano et al., 2003; Saba-El-Leil et al., 2003), we chose to re-express ERK2 WT rather
than ERK1 WT in the Erk DKO ESCs. Erk DKO + Erk2 WT transgenic ESCs were generated by
Figure 1.15 Deletion of ERK1/2 impairs mesendoderm differentiation of ESCs.
Representative phase contrast and immunofluorescent staining images for TUJ1, Myosin and GATA4 in the
differentiated WT ESCs and Erk DKO ESCs using an embryoid body differentiation protocol. Data are representative
of at least three independent experiments. For mesodermal differentiation, cells in the whole well were scanned. There
were no cells positive for mesendodermal markers. TUJ1: neuronal lineage marker; Myosin: mesoderm marker; GATA4:
endoderm marker. Scale bar, 100µm.
34
transfecting the Erk DKO ESCs with a construct expressing ERK2 WT and red fluorescence
protein (RFP) linked with a P2A self-cleaving peptide. The ERK2 WT and RFP proteins would be
separated through the cleavage of P2A peptide after they were translated in the Erk2 WT transgenic
ESCs. The RFP in the Erk2 WT transgenic ESCs facilitates the tracing of the expression of ERK2
WT in vitro and the transgenic ESCs in the chimeras. We showed that in the EB differentiation
assays, the Erk2 WT transgenic ESCs readily differentiated into cells of the neuroectoderm (TUJ1+
neurons) and mesendoderm (Myosin+ cardiomyocytes) lineages (Figure 1.16A). Furthermore, we
performed a chimera formation assay to test whether the Erk2 WT transgenic ESCs could be
incorporated into the developing embryo, a defining feature of authentic ESCs (Ying and Smith,
2017). Indeed, these Erk2 WT transgenic ESCs could contribute to the formation of chimeras after
the injection of them into E3.5 blastocysts (Figure 1.16B). These data together suggested that the
Figure 1.16 Erk2 WT transgenic ESCs can differentiate in vitro and contribute to chimera formation in vivo.
(A) Representative immunofluorescent images for TUJ1 and Myosin in the in vitro differentiated Erk2 WT
transgenic ESCs (Erk DKO + Erk2 WT ESCs) using an embryoid body differentiation protocol. TUJ1: neuronal
lineage marker; Myosin: mesoderm marker. Scale bar, 100µm. (B) Representative images of an E12.5 embryo
harvested from pseudopregnant C57BL/6 mice, which were transplanted with the blastocysts injected with the Erk2
WT transgenic ESCs (Erk DKO + Erk2 WT ESCs). Three embryos out of the 19 embryos (3/19) collected from two
pseudopregnant mice were RFP+ (chimera with the Erk2 WT transgenic ESCs). TUJ1: neuronal lineage marker;
Myosin: mesoderm marker; Scale bars, 1000µm.
35
inability of Erk DKO ESCs to differentiate was reversible and that Erk DKO ESC were genuine
ESCs.
Collectively, these findings suggest that ERK1/2 are dispensable for ESC survival. The deletion
of ERK1/2 in ESCs enhances their self-renewal, but impairs their differentiation into mesodermal
and endodermal cells.
2. Deletion or Chemical Inhibition of ERK1/2 Is Sufficient for ESC Self-renewal in
the presence of GSK3 Inhibitor
Dual inhibition of GSK3 and MEK (2i) can maintain ESCs undifferentiated in the long term (Ying
and Smith, 2017; Ying et al., 2008). To determine whether the deletion of ERK1/2 in ESCs could
mimic the effects of MEK inhibition by PD, we cultured WT ESCs and Erk DKO ESCs in serum-
free N2B27 medium supplemented with PD + CHIR (2i) or CHIR alone. WT ESCs remained
undifferentiated and formed compact colonies with sharp boundaries in the 2i medium; however,
the majority of WT ESCs differentiated and became flat in the CHIR alone medium (Figure 1.17A).
In contrast, Erk DKO ESCs could formed uniform compact colonies under both the 2i and the
CHIR alone conditions (Figure 1.17A).
To assess the self-renewal status of WT ESCs and Erk DKO ESCs maintained in 2i or the CHIR
alone condition, we performed alkaline phosphatase (AP) staining assays. In the AP staining assay,
undifferentiated ESCs are stained red (termed AP+); in contrast, differentiated ESCs cannot be
stained (termed AP-) due to the loss of alkaline phosphatase over the course of differentiation
(Figure 1.16B). The dome-shaped colonies consisting of AP+ cells are classified as self-renewing
colonies, and the colonies with AP- cells and rough boundaries are termed as differentiated
36
colonies. For WT ESCs, the percentage of self-renewing colonies in the CHIR alone condition
was significantly lower than that in the 2i condition, suggesting that WT ESCs underwent
differentiation under the CHIR alone condition. In contrast, for the Erk DKO ESCs, percentages
of self-renewing colonies under the 2i condition were similar to that under the CHIR alone
condition (Figure 1.17C). This result suggests that Erk DKO ESCs may be able to self-renew under
the CHIR alone condition for long-term. To test this hypothesis, we continuously passaged Erk
DKO ESCs for more than ten passages. We found that Erk DKO ESCs remained undifferentiated
Figure 1.17 Erk DKO ESCs can be maintained undifferentiated in the presence of GSK3 inhibitor.
(A) Representative phase contrast images of WT ESCs and the different clones of Erk DKO ESCs cultured under the
indicated conditions in serum-free N2B27 medium. Scales bar, 100µm. (B) Alkaline phosphatase (AP) staining
results of the colonies formed from the indicated ESCs cultured under PD + CHIR condition or CHIR alone condition
for 7 days. All treatments were done in triplicate (also all the AP staining in this study). Representative images of the
whole well (top) and a colony in the indicated well (bottom) were shown. The Erk DKO ESCs c1 was selected for
this assay. Scale bar, 100µm. (C) Quantification of the percentages of self-renewing colonies (dome-shaped colonies
consisting of AP+ cells) formed in WT ESCs and Erk DKO ESCs. The data represent three independent experiments
and are shown as mean ± SEM. ***: p < 0.001.
37
in the presence of CHIR alone based on the colony morphology and the expression of pluripotency
markers Rex1 and Oct4 (Figure 1.18).
Figure 1.18 Erk DKO ESCs retain the self-renewal in the presence of GSK3 inhibitor in the long run.
(A) Representative phase contrast images of the Erk DKO ESCs continuously passaged (p4 and p10) and cultured
under indicated conditions in N2B27 medium. Scale bar, 100µm. (B) Representative phase contrast and
immunofluorescent images for pluripotency markers (Rex1 and Oct4) in the Erk DKO ESCs continuously passaged
(p20) under the indicated conditions in N2B27 medium. Scale bar, 100µm.
Figure 1.19 The combined use of ERK and GSK3 inhibitors maintains ESC self-renewal.
(A) Representative phase contrast images of the WT ESCs and the Erk DKO ESCs (c1) cultured in N2B27 medium
supplemented with PD + CHIR or Vx11e + CHR. The other two clones of Erk ESCs were also kept undifferentiated
under the 2i and Vx11e + CHR conditions. Scale bar, 100µm. (B) AP staining-based quantification of the
percentages of self-renewing colonies formed in WT ESCs and the Erk DKO ESCs (c1) cultured in N2B27 medium
supplemented with PD + CHIR or Vx11e + CHIR. One of the three clones of Erk DKO ESCs (Erk DKO c1) was
selected. The data represent two independent experiments and are shown as mean ± SEM.
38
To further confirm the role of ERK1/2 in ESC self-renewal and differentiation, we tested an
ERK1/2 inhibitor (Vx11e) (Chaikuad et al., 2014) on ESCs to see if it could replace the role of PD
in the 2i condition to maintain ESC self-renewal. We combined Vx11e with CHIR and tested this
new combination in both WT ESCs and Erk DKO ESCs. These cells exhibited as compact colonies
under both the 2i and the Vx11e + CHIR conditions (Figure 1.19A). Consistent with the above
results, in the AP staining assay, both WT ESCs and Erk DKO ESCs cultured in the Vx11e +
CHIR medium had very high percentages of self-renewing colonies, comparable to those
percentages in these two types of ESCs cultured in the 2i medium (Figure 1.19B). To further
evaluate the effect of Vx11e + CHIR on ESC self-renewal in the long term, we continuously
Figure 1.20 The combined use of ERK and GSK3 inhibitors maintains ESC self-renewal in the long term.
(A) Representative phase contrast images of WT ESCs (p12 and p20) continuously passaged and cultured under the
indicated conditions. Scale bar, 100µm. (B) Representative immunofluorescent images for pluripotency markers
(Rex1 and Oct4) in the WT ESCs (p20) cultured under the indicated conditions in N2B27 medium. Scale bar, 100µm.
39
passaged WT ESCs and cultured them under this condition for over 20 passages. The results
showed that WT ESCs could always form uniform compact colonies under both the 2i and the
Vx11e + CHIR conditions (Figure 1.20A). In addition, these WT ESCs propagated with the Vx11e
+ CHIR condition strongly expressed pluripotency markers, Rex1 and Oct4 (Figure 1.20B). The
above results indicate that the inhibition of kinase activity of ERK1/2 by Vx11e can fully
recapitulate the effect of PD in maintaining ESCs in the presence of CHIR.
In summary, our data suggest that inhibition of ERK1/2 by either knock-out or the use of chemical
inhibitors is sufficient to maintain ESC self-renewal in the presence of CHIR.
3. Deletion of either ERK1 or ERK2 Is Insufficient for ESC Self-Renewal in the
Presence of GSK3 Inhibitor
To determine the roles of individual ERK isoforms in ESCs, we generated Erk1 knock-out ESCs
(Erk1
-/-
ESCs) and Erk2 knock-out ESCs (Erk2
-/-
ESCs) with CRIPSR/Cas9 technology. Screening
by the immunoblotting, nine out of the twenty picked clones in the pool of candidate Erk1
-/-
ESCs
showed the deletion of ERK1, and eight out of the twenty-two picked clones in the pool of
candidate Erk2
-/-
ESCs showed the deletion of ERK2. Three clones of both the Erk1
-/-
ESCs (c1,
c2, and c3) and the Erk2
-/-
ESCs (c1, c2, and c3) were randomly selected for our further study
(Figure 1.21A). DNA sequencing on Erk genomic regions confirmed the knock-out of Erk1 and
Erk2 in Erk1
-/-
ESCs and Erk2
-/-
ESCs, respectively (Figure 1.21B).
To explore the effects of individual ERK isoform deletion on ESC self-renewal, we cultured Erk1
-
/-
ESCs and Erk2
-/-
ESCs under the 2i and CHIR alone conditions. All three clones of Erk1
-/-
ESCs
under the 2i condition formed dome-shaped colonies (Figure 1.22A), and were positive for Rex1
40
and Oct4 (Figure 1.22B). However, they differentiated and became flat under the CHIR alone
condition, and became Rex1 and Oct4 negative (Figure 1.22A and 1.22B). Similar results could
also be obtained from the three clones of Erk2
-/-
ESCs (Figure 1.22C and 1.22D). These results
suggest that inhibition of ERK1 in Erk2
-/-
ESCs and the inhibition of ERK2 in Erk1
-/-
ESCs by
MEK inhibitor (PD) are necessary to maintain their undifferentiated status under the CHIR alone
condition.
To further confirm the self-renewal status of these ESCs under the CHIR alone condition or 2i
condition, we performed the AP staining assays and quantified the percentages of self-renewing
Figure 1.21 Confirmation of Erk1
-/-
ESCs and Erk2
-/-
ESCs.
(A) Western blot analysis of the expression of ERK1/2 and GAPDH in three chosen clones of both Erk1
-/-
ESCs and
Erk2
-/-
ESCs. WT ESCs were used as the control. (B) The DNA sequencing results on Erk genomic regions in
different clones of Erk1
-/-
ESCs and Erk2
-/-
ESCs.
41
colonies in Erk1
-/-
ESCs and Erk2
-/-
ESCs. It was revealed that in either Erk1
-/-
ESCs or the Erk2
-
Figure 1.22 Erk1
-/-
ESCs and Erk2
-/-
ESCs cannot be kept undifferentiated under the CHIR alone condition.
(A, C) Representative phase contrast images of different clones of Erk1
-/-
ESCs (A) and Erk2
-/-
ESCs (C) cultured
under the indicated conditions in serum-free N2B27 medium. Scale bar, 100µm. (B, D) Representative phase
contrast and immunofluorescent images for Rex1 and Oct4 in the Erk1
-/-
ESCs c1 (B) and the Erk2
-/-
ESCs c1 (D)
cultured under the indicated conditions in serum-free N2B27 medium. The other clones of Erk1
-/-
ESCs and Erk2
-/-
ESCs had similar staining results under 2i and the CHIR alone conditions (data not shown). Scale bar, 100µm.
Figure 1.23 GSK3 inhibitor alone cannot retain the self-renewal of Erk1
-/-
ESCs and Erk2
-/-
ESCs.
AP staining-based quantification of the percentages of self-renewing colonies in Erk1
-/-
ESCs and Erk2
-/-
ESCs
cultured in N2B27 medium supplemented with PD + CHIR or CHIR alone. The Erk1
-/-
ESCs c1 and the Erk2
-/-
ESCs c1 were selected for the tests. The data represent three independent experiments and are shown as mean ±
SEM. ***: p < 0.001.
42
/-
ESCs, the percentage of self-renewing colonies under the CHIR alone condition was significantly
lower than that under the 2i condition (Figure 1.23).
These results suggest that deletion of one of the ERK1/2 isoforms is not sufficient for ESC self-
renewal in the presence of GSK3 inhibitor.
4. Selective Chemical Inhibition of Individual ERK Isoforms Is Achieved through
an Inhibitor-Resistant Approach
It is well known that pharmacological inhibition and genetic deletion of kinases can result in very
different phenotypes (Bi et al., 2002; Chen et al., 2017; Papa et al., 2003). As such, I next
investigated the effect of chemically inhibiting one ERK isoform on ESC self-renewal when both
ERK1/2 are present. To achieve individual inhibition of one of the ERK1/2 isoforms, an ERK-
isoform specific inhibitor should be used. However, currently there are no small molecules that
can selectively inhibit one of the two highly homologous ERK1/2 isoforms while sparing the other.
To address this problem, we first tried a bump-hole approach on ERK1/2 to achieve selective
inhibition of either ERK1 or ERK2 (Figure 1.8). Successfully utilization of the bump-hole
approach for studying the function of individual kinase isoforms requires the kinase mutant to
preserve its kinase activity. A system was needed to assess the kinase activity of ERK analog-
sensitive (AS) mutants in ESCs. Erk DKO ESCs represent a valuable tool to evaluate the kinase
activity of ERK mutants by expressing the mutants in Erk DKO ESCs via a transgenic method: if
the ERK mutant retains kinase activity, Erk DKO ESCs expressing this ERK mutant will
differentiate under the CHIR alone condition, similar to the phenotype of ESCs with one functional
ERK isoform (e.g. Erk1
-/-
ESCs or Erk2
-/-
ESCs) under the same condition; if the ERK mutant
43
loses its kinase activity, Erk DKO ESCs expressing this ERK mutant should self-renew in the
presence of CHIR, similar to the phenotype of Erk DKO ESCs under the same condition.
The ERK AS mutant was generated by replacing the “gate-keeper” residue in ERK1/2 with a
smaller amino acid residue to expand the kinase domain so that the AS mutant can be specifically
inhibited by a bump inhibitor (e.g. 3MB-PP1) (Carlson et al., 2011; Emrick et al., 2006). To
generate the ERK AS mutant, I introduced the ERK1 Q123G mutation and the ERK2 Q103G
mutation into ERK1 and ERK2, respectively. To evaluate whether these two mutants preserve
their kinase activity in ESCs, I first established Erk1 AS transgenic ESC lines (Erk DKO + Erk1
AS ESCs) and Erk2 AS transgenic ESC lines (Erk DKO + Erk2 AS ESCs) introducing the Erk AS
transgenes into Erk DKO ESCs. The isogenic transgenic ESC lines expressing ERK1 WT proteins
(Erk DKO + Erk1 WT ESCs) or ERK2 WT proteins (Erk DKO + Erk2 WT ESCs) were generated
and used as controls. I plated these four transgenic ESC lines under 2i and the CHIR alone
conditions, and checked their differentiation status. Both the Erk2 WT transgenic ESCs and the
Erk2 AS (Erk2 Q103G) ESCs differentiated under the CHIR alone condition, implying the well-
preserved kinase activity in the ERK2 AS mutant (data not shown). However, the Erk1 AS (Erk1
Q123G) transgenic ESCs could not differentiate and formed uniform dome-shaped colonies under
the CHIR alone condition, in comparison to the flat morphology of the Erk1 WT transgenic ESCs
under the same condition (Figure 1.24). This suggested that the ERK1 AS (ERK1 Q123G) mutant
lost its kinase activity. Mouse ERK1 Q123G mutant lost the majority of kinase activity in an in
vitro kinase assay (Endo et al., 2006), which is consistent with the above findings. I further tried a
second-site suppressor strategy (Zhang et al., 2005) on the ERK1 AS mutant , but this strategy
failed to rescue its kinase activity (data not shown). Taking into consideration of the self-renewal-
promoting effect of 3MB-PP1 (a bump inhibitor) on ESCs (Figure 1.24), I abandoned this
44
approach and tried to develop an alternative approach for the selective inhibition of ERK1/2
isoforms.
I devised an inhibitor-resistant (IR) strategy based on Vx11e to specifically inhibit one of the
ERK1/2 isoforms (Figure 1.25). With the introduction of an isoform specific IR mutation, the
mutated ERK isoform will no longer be inhibited by Vx11e. For instance, in Erk1 IR knock-in
cells, Vx11e can inhibit ERK2 WT proteins but not ERK1 IR mutant. For the successful
Figure 1.24 The ERK1 AS (ERK1 Q123G) mutant loses the kinase activity in ESCs.
Representative phase contrast images of Erk1 WT transgenic ESCs (Erk DKO + Erk1 WT) and Erk1 AS ESCs (Erk
DKO + Erk1 AS) cultured under the indicated conditions in serum-free N2B27 medium. 3-MB: 3MB-PP1, a bump
inhibitor. The red arrowhead indicates a dome-shaped colony. AS: analog sensitive. Scale bar, 100µm.
Figure 1.25 Illustration of the inhibitor-resistant strategy to achieve individual inhibition of highly homologous
ERK1/2 isoforms.
(A) ERK1 IR mutants are expressed in the engineered cells. ERK2 WT proteins are specifically inhibited when Vx11e
is added since ERK1 IR mutants are resistant to Vx11e. (B) ERK2 IR mutants are expressed in the engineered cells
and ERK1 WT proteins are selectively inhibited when Vx11e is supplemented. The kinase activity of two ERK1/2
isoforms can be inhibited in both two types of cells by PD (MEK inhibitor). IR: inhibitor (Vx11e)-resistant.
45
application of the IR strategy to ERK1/2, an ideal ERK IR mutant should have at least the
following features: a) well-preserved kinase activity; b) resistance to Vx11e but not to PD; c) no
changes in the studied phenotype(s) when the IR mutant is expressed. Human ERK mutations
conferring drug resistance to Vx11e had been revealed via the drug screening in human cancer
cell lines (Goetz et al., 2014). There were three homologous mutations between ERK1 and ERK2.
It was shown that human ERK1/2 with the three homologous mutations matched the three
characteristics for generating an ideal ERK IR mutant (Goetz et al., 2014). As such, we utilized
these three homologous mutations to generate ERK IR mutants in mouse.
Transgenic ESC lines individually expressing one of the three either ERK1 (Y54H, G55A, P76L)
or ERK2 (Y34N, G35S, P56L) homologous mutants were established based on the Erk DKO ESCs.
Figure 1.26 Confirmation of Erk DKO ESCs expressing ERK mutants.
(A) Schematic illustration of the construct for establishing Erk mutant transgenic ESCs. RFP: red fluorescence
protein. (B) Western blot analysis of the expression of ERK1/2, RFP, and GAPDH in the Erk mutant transgenic
ESCs established by transfecting the Erk DKO ESCs with the construct shown in (A). The RFP transgenic ESC
line (Erk DKO + RFP ESCs) was established as control. For each type of Erk mutant transgenic ESCs, we had
two different clones and one of the two clones was shown.
46
In these Erk IR transgenic ESCs (Erk DKO + Erk IR ESCs), ERK mutants were co-expressed with
RFP linked by a P2A self-cleaving peptide to facilitate the tracking of their expression (Figure
1.26A). Because of this, I also generated an isogenic transgenic ESC line expressing RFP (Erk
DKO + RFP ESCs) as the control. The expression of these ERK IR mutants in transgenic ESCs
was confirmed by immunoblotting (Figure 1.26B).
To identify an optimal ERK IR mutant(s), I needed a system to evaluate the kinase activity of these
ERK mutants. Egr1 is a downstream target of ERK1/2 and can be induced in both Erk1
-/-
cells and
Erk2
-/-
cells when the MAPK/ERK pathway is activated (Fischer et al., 2005; Lee et al., 2014; Ye
et al., 2013). We then chose Egr1 and tested whether its transcriptional levels could represent ERK
kinase activity in ESCs. The levels of Egr1 were significantly increased in the Erk WT (Erk1 WT
or Erk2 WT) transgenic ESCs but not in the RFP transgenic ESCs when these cells were stimulated
with serum (Figure 1.27). In addition, the up-regulation of Egr1 was abolished in the Erk WT
transgenic ESCs when either PD or Vx11e was supplemented together with serum (Figure 1.27).
Figure 1.27 The Egr1 level in transgenic ESCs is a good readout to evaluate the kinase activity and drug
response of ERK proteins.
qPCR analysis of Egr1 expression in the RFP transgenic ESC lines (Erk DKO + RFP ESCs) and Erk WT transgenic
ESCs (Erk DKO + Erk1 WT ESCs, Erk DKO + Erk2 WT ESCs) under indicated conditions. Cells were starved in
serum free medium overnight and then treated with serum, serum + 1µM PD03, serum + 1µM Vx11e for 3 hours
before qPCR analysis. The data represent two independent experiments and are shown as mean ± SEM. ***:
p<0.001. NT: non-treatment.
47
These results indicated that the Egr1 level in the transgenic ESCs is a good readout for evaluating
the kinase activity and drug response of ERK1/2.
I evaluated all six ERK IR mutants by qPCR analysis of the Egr1 levels in their transgenic ESCs
under different conditions. Among the three ERK1 IR mutants, all could significantly elevate the
transcriptional levels of Egr1 in the Erk1 IR transgenic ESCs upon the addition of serum. On the
other hand, the induction of Egr1 in these transgenic ESCs was significantly repressed when PD
was present in the medium (Figure 1.28A). However, Egr1 expression could still be induced in
Erk DKO + ERK1 IR ESCs when Vx11e was supplemented (Figure 1.28A), suggesting that
MAPK/ERK signaling in these Erk1 IR transgenic ESCs was activated under the condition of
serum plus Vx11e. My results revealed that all of the three Erk1 mutants were genuine Vx11e-
Figure 1.28 The screening on ERK1 IR mutants by Egr1-based qPCR.
(A) qPCR analyses of Egr1 levels in the Erk1 IR transgenic ESC lines (Erk DKO + Erk1 Y54H ESCs, Erk DKO
+ Erk1 G55A ESCs, and Erk DKO + Erk1 P76L ESCs) under the indicated conditions. The data represent two
independent experiments and are shown as mean ± SEM. ***: p<0.001. (B) The relative kinase activity for the
ERK1 IR mutants under each condition was calculated. For each Erk1 IR transgenic cell line, the NT group was
used as a baseline, and the FBS alone group was set as 100%. The relative kinase activities of the PD and Vx11e
groups were then calculated. FBS: fetal bovine serum.
48
resistant mutants. To better compare the three ERK1 IR mutants, their relative kinase activities
under different conditions were quantified based on the Egr1 levels. For each transgenic cell line,
the kinase activity in the serum-starved group was used as a baseline, and that in the serum-alone
group was defined as 100%. After quantification, it was revealed that the ERK1 G55A mutant had
a higher kinase activity under the serum + Vx11e condition than under the serum-alone condition
(Figure 1.28B). Likewise, all the three ERK2 IR mutants could significantly induce Egr1
transcription in their transgenic ESCs when stimulated with serum, and the up-regulation of Egr1
levels could be inhibited by PD but not Vx11e (Figure 1.29A). Additionally, similar to the ERK1
G55A mutant, the homologous ERK2 G35S mutant increased its kinase activity when Vx11e was
supplemented (Figure 1.29B). Based on the levels of relative kinase activity under the serum +
Figure 1.29 The screening on ERK2 IR mutants by Egr1-based qPCR.
(A) qPCR analyses of Egr1 levels in the Erk2 IR transgenic ESC lines (Erk DKO + Erk2 Y34N ESCs, Erk DKO +
Erk2 G35S ESCs, and Erk DKO + Erk2 P56L ESCs) under the indicated conditions. The data represent two
independent experiments and are shown as mean ± SEM. ***: p<0.001. (B) The relative kinase activity for the
ERK2 IR mutants under each condition was calculated. For each Erk2 IR transgenic cell line, the NT group was
used as a baseline, and the FBS alone group was set as 100%. The relative kinase activities of the PD and Vx11e
groups were then calculated. FBS: fetal bovine serum.
49
Vx11e condition, the ERK1 G55A mutant and its homologous ERK2 G35S mutant were the
optimal IR mutants for ERK1 and ERK2, respectively.
To further verify the kinase activities and drug response of the two ERK mutants, we measured
the phosphorylation levels of p90-ribosomal protein S6 kinase (RSK), a direct and common
substrate of ERK1/2 (Goetz et al., 2014). The levels of phosphorylated RSK (pRSK) were
significantly increased in the Erk WT transgenic ESCs but not in the RFP transgenic ESCs when
these cells were stimulated with basic fibroblast growth factor (bFGF) (Figure 1.30). The
supplementation of either PD or Vx11e reduced the pRSK levels to the basal level observed in
serum-starved groups of the Erk WT transgenic ESCs (Figure 1.30). These findings suggest that
the changes in the levels of pRSK in transgenic ESCs can represent the relative kinase activity and
inhibition response of ERK proteins.
Figure 1.30 The level of pRSK in transgenic ESCs is a good readout to evaluate the relative kinase activity
and drug response of ERK proteins.
Western blot analysis of the expressions of phosphorylated RSK (pRSK), total RSK (tRSK), phosphorylated ERK
(pERK), total ERK (tERK), and GAPDH in RFP transgenic ESCs (Erk DKO + RFP ESCs) and Erk WT transgenic
ESCs (Erk DKO + Erk1 WT ESC, Erk DKO + Erk2 WT ESCs) under the indicated conditions. Cells were treated
with the indicated combination of bFGF, PD03, and Vx11e for 10-15 min after overnight serum starvation. bFGF:
basic fibroblast growth factor. +: the addition; -: the absence.
50
Next, the kinase activities and drug responses of the ERK1 G55A mutant and the ERK2 G35S
mutant were evaluated in ESCs by the pRSK-based system. Both the Erk1 G55A transgenic ESCs
and the Erk2 G35S transgenic ESCs had higher levels of pRSK when cells were stimulated with
bFGF, suggesting that both the ERK1 G55A and ERK2 G35S mutants retained their kinase activity
in ESCs (Figure 1.31). The addition of PD reduced the levels of pRSK in these two transgenic
ESCs; however, the addition of Vx11e did not decrease the pRSK levels (Figure 1.31). The
increased pRSK levels in the Erk IR (Erk1 G55A or Erk2 G35S) transgenic ESCs after the
stimulation with Vx11e and bFGF suggested the activation of MAPK/Erk signaling in these cells
even in the presence of Vx11e. Based on these results together with the findings in the Egr1-based
tests, we concluded that the ERK1 G55A and ERK2 G35S mutants preserved their kinase activity
and that these two ERK IR mutants conferred drug resistance to Vx11e but still remained sensitive
to PD in ESCs.
Figure 1.31 pRSK-based confirmation of the kinase activities and inhibition responses of ERK1 G55A and
ERK2 G35S mutants in ESCs.
Western blot analysis of the expression of pRSK, tRSK, pERK, tERK, and GAPDH in Erk1 G55A transgenic ESCs
(Erk DKO + Erk1 G55A ESCs) and Erk2 G35S transgenic ESCs (Erk DKO + Erk2 G35S ESCs) under the indicated
conditions. Cells were treated with the indicated combination of bFGF, PD03 and Vx11e for 10-15 min after overnight
serum starvation. bFGF: basic fibroblast growth factor. +: the addition; -: the absence.
51
Finally, I examined the function of ERK1 G55A and ERK2 G35S mutants in ESCs. Erk DKO
ESCs provide a valuable tool to evaluate the kinase activity of ERK mutants based on cell fates of
the Erk mutant transgenic ESCs under the CHIR alone condition. The RFP transgenic ESCs
without the expression of ERK1/2, similar to Erk DKO ESCs, retained the self-renewal under the
CHIR alone condition (Figure 1.32). The Erk WT transgenic ESCs differentiated in the presence
of CHIR alone (Figure 1.32). Similar to the Erk WT transgenic ESCs, the Erk IR transgenic ESCs
differentiated in the CHIR alone medium (Figure 1.33), suggesting that the ERK1 G55A and
ERK2 G35S mutant preserved their kinase activity in ESCs. In addition, both the Erk WT
transgenic ESCs and Erk IR transgenic ESCs remained undifferentiated under the 2i condition.
Likewise, the WT transgenic ESCs were undifferentiated under the Vx11e + CHIR condition;
however, the transgenic ESCs expressing ERK1 G55A or ERK2 G35S differentiated under the
Figure 1.32 Erk DKO ESCs is a valuable tool to evaluate the kinase activity of ERK mutants.
Representative phase contrast images of RFP transgenic ESCs (Erk DKO + RFP ESCs) and Erk WT transgenic
ESCs (Erk DKO + Erk1 WT ESC, Erk DKO + Erk2 WT ESCs) cultured under the indicated conditions in N2B27
medium for 7 days. Scale bars, 100µm.
52
same condition (Figure 1.33). These data indicated that both ERK1 G55A and ERK2 G35S
mutants had good kinase activity in ESCs, and could induce ESC differentiation under the CHIR
alone condition.
Collectively, these results suggest that ERK1 G55A and ERK2 G35S mutants in ESCs retain
kinase activity, are resistant to Vx11e but responsive to PD, and do not alter the phenotype of our
interest. We also conclude that the ERK1 G55A mutation and the ERK2 G35S mutation can be
used to establish engineered ESC lines in which individual ERK isoforms can be selectively
inhibited by Vx11e.
5. Selective Inhibition of either ERK1 or ERK2 Is Not Sufficient to Maintain ESC
Self-Renewal in the Presence of GSK3 Inhibitor
Having demonstrated that the individual ERK1/2 isoforms can be specifically inhibited by this
inhibitor-resistant strategy, I next investigated the resulting phenotypes with regard to ESC self-
renewal. To this end, I first generated Erk1 G35S knock-in (KI) ESCs and Erk2 G35S KI ESCs
Figure 1.33 Erk1 G55A transgenic ESCs and Erk2 G35S transgenic ESCs differentiate under the Vx11e +CHIR
and CHIR alone conditions.
Representative phase contrast images of Erk1 G55A transgenic ESCs (Erk DKO + Erk1 G55A ESCs) and Erk2 G35S
transgenic ESCs (Erk DKO + Erk2 G35S ESCs) cultured under the indicated conditions in N2B27 medium for 7 days.
Scale bars, 100µm.
53
via CRISPR/Cas9-mediated homologous recombination. A new restrictive enzyme digestion site,
which could not be found in the genotyping products of wild-type alleles, was introduced into each
donor to facilitate the identification of the clones with the mutated allele(s) via a restriction
fragment length polymorphism (RFLP) assay. In the RFLP assays, five out of the 83 picked clones
were homozygous (both alleles with Erk1 G55A mutation) in the pool of candidate Erk1 G55A KI
ESCs, and six out of the 93 picked clones were homozygous in the pool of candidate Erk2 G35S
KI ESCs (data not shown). Two clones of both the Erk1 G55A KI ESCs (c1 and c2) and the Erk2
G35S KI ESCs (c1 and c2) were selected and DNA sequencing on Erk genomic regions confirmed
the homozygous intended mutations in them (Figure 1.34A and 1.34B). I also confirmed the
expression of ERK1/2 in these Erk IR KI ESCs by immunoblotting (Figure 1.34C). In both the
two clones of Erk1 G55A KI ESCs and Erk2 G35S KI ESCs, the levels of ERK1/2 were similar to
those in WT ESCs (Figure 1.34C). These results revealed that the successful establishment of the
Erk1 IR KI ESCs (Erk1 G55A KI ESCs) and Erk2 IR KI ESCs (Erk2 G35S KI ESCs) expressing
ERK IR mutants.
Figure 1.34 Confirmation of Erk IR knock-in ESC lines by DNA sequencing and immunoblotting.
(A-B) Representative DNA sequencing results on Erk genomic regions of Erk1 G55A KI ESCs (A) and Erk2 G35S
KI ESCs (B). For each type of Erk IR (Erk1 G55A or Erk2 G35S) KI ESCs, we had two different clones and the
results of one clone of the two cell lines were shown. (C) Western blot analysis of the expression of ERK1/2 and
GAPDH in the two different clones of Erk IR KI ESCs. KI: knock-in. ERK1: 44 kDa, Erk2: 42 kDa. GAPDH: 37
kDa.
54
Using these genetically engineered ESCs, I next examined the effect of selective inhibition of
ERK1 or ERK2 on ESC self-renewal. In these Erk IR KI ESCs, both ERK1/2 were inhibited under
2i condition and were active under the CHIR alone condition. In the medium with Vx11e + CHIR,
ERK1 and ERK2 were selectively inhibited in Erk2 IR KI ESCs and Erk1 IR KI ESCs, respectively.
It was revealed that both Erk1 and Erk2 IR KI ESCs were undifferentiated under 2i condition and
differentiated under the CHIR alone condition (Figure 1.35A and 1.35B). In addition, the selective
inhibition of ERK2 in Erk1 IR KI ESCs or the selective inhibition of ERK1 in Erk2 IR KI ESCs
by Vx11e was not sufficient to maintain ESC self-renewal together with CHIR, leading to the
Figure 1.35 Selective inhibition of ERK1 or ERK2 cannot maintain ESC self-renewal in the presence of GSK3
inhibitor.
(A-B) Representative phase contrast images of the Erk1 IR KI ESCs (A) and Erk2 IR KI ESCs (B) under the indicated
conditions in N2B27 medium for 7 days. Two clones for each KI cell line. Scale bar, 100µm. (C) AP staining-based
quantification of the percentages of self-renewing colonies formed in the Erk1 IR KI ESCs and the Erk2 IR KI ESCs
cultured in N2B27 medium supplemented with PD + CHIR, Vx11e + CHIR, or CHIR. Data represent two independent
experiments and are shown as means ± SEM. ***: p < 0.001. KI: knock-in.
55
formation of a mixed population of undifferentiated and differentiated colonies (Figure 1.35A and
1.35B). Consistent with the above results, in the AP staining assay, the percentages of self-
renewing colonies in the Erk1 IR KI and Erk2 IR KI ESC populations cultured with the Vx11e +
CHIR medium were significantly lower than those cultured with the 2i medium (Figure 1.35C).
These data suggested that the individual inhibition of ERK1 or ERK2 could not maintain ESC self-
renewal in the presence of GSK3 inhibitor.
To rule out possible off-target effects of CRISPR/Cas9 based genetic modifications on ESC self-
renewal, we further confirmed our conclusion regarding the individual roles of ERK1/2 isoforms
in ESC self-renewal with an add-back strategy. To this end, I first established Erk1 IR transgenic
ESCs (Erk1
-/-
+ Erk1 IR) by transfecting Erk1
-/-
ESCs with the construct expressing the ERK1 IR
mutant and RFP together. I also generated isogenic RFP transgenic ESCs (Erk1
-/-
+ RFP) and Erk1
WT transgenic ESCs (Erk1
-/-
+ Erk1 WT) as controls. Similarly, I produced isogenic Erk2 IR, Erk2
WT, and RFP transgenic ESCs based on Erk2
-/-
ESCs with similar constructs. The expression of
Figure 1.36 Confirmation of the expression of ERK proteins in transgenic ESCs established from the Erk1
-
/-
ESCs and the Erk2
-/-
ESCs.
(A-B) Western blot analysis of the expression of ERK1/2, RFP, and GAPDH in the Erk1
-/-
ESCs based transgenic
ESC lines (Erk1
-/-
+ RFP ESCs, Erk1
-/-
+ Erk1 WT ESCs, and Erk1
-/-
+ Erk1 IR ESCs) (A) and in the Erk2
-/-
ESCs
based transgenic ESC lines (Erk2
-/-
+ RFP ESCs, Erk2
-/-
+ Erk2 WT ESCs, and Erk2
-/-
+ Erk2 IR ESCs) (B). WT
ESCs were used as the control. ERK1: 42 kDa; ERK2: 44 kDa; GAPDH: 37 kDa; RFP: 26 kDa.
56
the transgenes (RFP, ERK1or ERK2) in these different transgenic ESCs were confirmed with
Figure 1.37 Specific inhibition of ERK2 in the Erk1 IR transgenic ESCs cannot maintain the self-renewal in the
presence of GSK3 inhibitor.
(A) Representative phase contrast images of the Erk1
-/-
ESCs based transgenic ESC lines (Erk1
-/-
+ RFP ESCs, Erk1
-/-
+ Erk1 WT ESCs, and Erk1
-/-
+ Erk1 IR ESCs) under the indicated conditions in N2B27 medium for 7 days. Scale bars,
100µm. (B) AP staining-based quantification of the percentages of self-renewing colonies formed in the transgenic
ESC lines (Erk1
-/-
+ RFP ESCs, Erk1
-/-
+ Erk1 WT ESCs, and Erk1
-/-
+ Erk1 IR ESCs) cultured in N2B27 medium
supplemented with PD + CHIR or Vx11e + CHIR. Data represent two independent experiments and are shown as
means ± SEM. ***: p < 0.001.
Figure 1.38 Specific inhibition of ERK1 in the Erk2 IR transgenic ESCs cannot maintain the self-renewal in
the presence of GSK3 inhibitor.
(A) Representative phase contrast images of the Erk2
-/-
ESCs based transgenic ESC lines (Erk2
-/-
+ RFP ESCs, Erk2
-
/-
+ Erk2 WT ESCs, and Erk2
-/-
+ Erk2 IR ESCs) under the indicated conditions in N2B27 medium for 7 days. Scale
bars, 100µm. (B) AP staining-based quantification of the percentages of self-renewing colonies formed in the
transgenic ESC lines (Erk2
-/-
+ RFP ESCs, Erk2
-/-
+ Erk2 WT ESCs, and Erk2
-/-
+ Erk2 IR ESCs) cultured in N2B27
medium supplemented with PD + CHIR or Vx11e + CHIR. Data represent two independent experiments and are
shown as means ± SEM. ***: p < 0.001.
57
immunoblotting (Figure 1.36A and B).
I next determined the individual roles of each ERK isoform in ESC self-renewal using these Erk
IR transgenic ESCs. Erk1
-/-
+ RFP ESCs and Erk1
-/-
+ Erk1 WT ESCs remained undifferentiated
in the presence of either 2i or Vx11e + CHIR (Figure 1.37A). Similarly, Erk1
-/-
+ Erk1 IR ESCs
was undifferentiated in 2i medium; however, they differentiated in the Vx11e + CHIR medium
(Figure 1.37A). In the AP staining assays for both Erk1
-/-
+ RFP ESCs and Erk1
-/-
+ Erk1 WT
ESCs, the percentages of self-renewing colonies under the 2i condition were similar to those under
the Vx11e + CHIR condition (Figure 1.37B). In contrast, the Erk1
-/-
+ Erk1 IR ESCs had a
significantly lower percentage of self-renewing colonies under the Vx11e + CHIR condition than
under 2i condition. Additionally, both Erk2
-/-
+ RFP ESCs and Erk2
-/-
+ Erk2 WT ESCs could be
maintained in the Vx11e + CHIR medium (Figure 1.38A), and their percentages of self-renewing
colonies under the Vx11e + CHIR condition were comparable to those under 2i condition (Figure
1.38B). In contrast, the Erk2
-/-
+ Erk2 IR ESCs could not keep self-renewing under the Vx11e +
CHIR condition (Figure 1.38A), and the percentage of self-renewing colonies for them under the
same condition was significantly lower than that under 2i condition (Figure 1.38B).
Collectively, based on the findings from the Erk IR KI ESCs and the Erk IR transgenic ESCs, we
conclude that individual chemical inhibition of either ERK1 or ERK2 is not sufficient to maintain
ESC self-renewal in the presence of GSK3 inhibitor.
6. Blocking ERK1 or ERK2 Kinase Function via a Kinase-Dead Approach Cannot
Maintain ESC Self-Renewal in the Presence of GSK3 Inhibitor
To further validate our conclusion about ERK1/2 from the novel inhibitor-resistant strategy, we
intended to selectively eliminate kinase activity of one of the ERK1/2 isoforms with a kinase-dead
58
(KD) approach so that the kinase activity of ERK1 or ERK2 in ESCs could be specifically blocked
with another method. The KD mutant of a protein is generated by introducing a point mutation in
the kinase domain so that its phosphoryl transfer potential is lost; however, the dead kinase can
still act as a scaffold for binding substrates. To generate ERK1 KD mutants and ERK2 KD mutants,
mutagenesis was performed to introduce a K72R mutation and a K52R mutation into the coding
sequences of Erk1 WT and Erk2 WT, respectively (Goetz et al., 2014).
To examine whether the kinase activities of these two ERK KD mutants in ESCs are fully lost or
not, Erk1 KD transgenic ESCs (Erk DKO + Erk1 KD ESCs) and Erk2 KD transgenic ESCs (Erk
DKO + Erk2 KD ESCs) were established by transfecting the Erk DKO ESCs with constructs
expressing the ERK1 K72R mutants and the ERK2 K52R mutants, respectively. The expression
of ERK mutants in these transgenic ESCs was confirmed by immunoblotting (Figure 1.39A). Our
previously established systems, including qPCR analysis of Egr1, immunoblotting of pRSK, and
the phenotype evaluation of Erk transgenic ESCs cultured in the CHIR alone medium, were used
to evaluate the kinase activities of the ERK KD mutants. The Egr1 levels had relatively smaller
changes in either Erk1 KD or Erk2 KD transgenic ESCs when stimulated with serum (Figure
1.39B), in comparison to more than twenty-fold changes in the Egr1 levels in the Erk WT
transgenic ESCs after the stimulation with serum (Figure 1.27). Similarly, the pRSK levels in the
Erk KD transgenic ESCs had very small changes when the cells were stimulated by bFGF, or
treated with PD or Vx11e (Figure 1.39C). Furthermore, the Erk KD transgenic ESCs were
59
maintained undifferentiated and formed uniform colonies when they were cultured under the CHIR
alone condition, indistinguishable from those cultured under the 2i or Vx11e + CHIR condition
(Figure 1.39D). These findings suggested that ERK KD mutants had lost their kinase activities and
were feasible to achieve selective inhibition of ERK1 or ERK2 in ESCs.
Figure 1.39 Functional confirmation of ERK kinase-dead mutants in ESCs.
(A) Western blot analysis of the expression of ERK1/2 and GAPDH in WT ESCs, the RFP transgenic ESC line
(Erk DKO + RFP ESCs) and the Erk KD transgenic ESC lines (Erk DKO + Erk1 KD ESCs, Erk DKO + Erk2 KD
ESCs) under indicated conditions. (B) qPCR analysis of Egr1 transcription in the Erk KD transgenic ESC lines
under indicated conditions. Cells were starved in serum free medium overnight and then treated with serum, serum
+ 1µM PD03, serum + 1µM Vx11e for 3 hours before qPCR analysis. Data represent mean ± SEM of three
independent experiments. (C) Western blot analysis of the expression of pRSK, tRSK, pERK1/2, tERK1/2, and
GAPDH in the Erk KD transgenic ESC lines under indicated conditions. Cells were treated with the indicated
combination of bFGF, PD03 and Vx11e for 10-15 min after overnight serum starvation. +: the addition; -: the
absence. (D) Representative phase contrast images of the Erk KD transgenic ESC lines in N2B27 medium
supplemented with PD + CHIR, Vx11e + CHIR, or CHIR for 7 days. Scale bar, 100µm. KD: kinase dead.
60
I next utilized the KD approach to confirm the roles of each ERK1/2 isoform in ESC self-renewal.
To specifically inhibit ERK1 or ERK2 in ESCs, we generated Erk1
-/-
+ Erk1 KD ESCs and Erk2
-
/-
+ Erk2 KD ESCs by transfecting Erk1
-/-
ESCs and Erk2
-/-
ESCs with constructs expressing the
ERK1 KD and the ERK2 KD mutants, respectively. The expression of the ERK KD mutants in
these transgenic ESCs was confirmed by immunoblotting (Figure 1.40). Following this, I
characterized the cell fate of these Erk KD transgenic ESCs cultured under the 2i and CHIR alone
conditions. It was revealed that both the Erk1
-/-
+ Erk1 KD ESCs and the Erk2
-/-
+ Erk2 KD ESCs
retained their self-renewal under the 2i condition; in contrast, both ESC lines differentiated under
the CHIR alone condition (Figure 1.41A and 1.41B). In accordance with the above results, both
the Erk1
-/-
+ Erk1 KD ESCs and the Erk2
-/-
+ Erk2 KD ESCs had significantly lower percentages
of self-renewing colonies under the CHIR alone condition than under the 2i condition (Figure
1.41C). These results suggest that blocking the kinase activity of one of the ERK1/2 isoforms is
not sufficient to retain ESC self-renewal in the presence of GSK3 inhibitor.
Figure 1.40 Confirmation of the expression of ERK proteins in Erk1
-/-
+ Erk1 KD ESCs and Erk2
-/-
+ Erk2 KD
ESCs.
Western blot analysis of the expression of ERK1/2, RFP, and GAPDH in WT ESCs, the Erk1
-/-
ESCs expressing
RFP or ERK1 KD mutant, and the Erk2
-/-
ESCs expressing RFP or ERK2 KD mutant. ERK1:42 kDa; ERK2: 42kDa;
RFP: 26 kDa; GAPDH: 37 kDa.
61
Taken together, these results further demonstrate that selective inhibition of either ERK1 or ERK2
in ESCs is not sufficient to maintain their self-renewal in the presence of CHIR. The consistent
results between the IR approach and the KD approach used for the individual inhibition of one of
the ERK1/2 isoforms also indicate the feasibility of using the IR approach to study the individual
roles of ERK1/2 isoforms in ESC differentiation.
7. Selective Inhibition of either ERK1 or ERK2 Cannot Block ESC Differentiation
The activation of MAPK/ERK signaling is critical for the differentiation of ESCs into cells of three
germ layers (Kunath et al., 2007). Previous studies have demonstrated that ERK1 and ERK2 in
the zebrafish played non-redundant roles in the gastrulation, a key process for three germ layer
differentiation (Krens et al., 2008a; Krens et al., 2008b). In addition, Erk1
-/-
mice are viable and
fertile, however, Erk2
-/-
mice are embryonic lethal due to defective placenta formation (Hatano et
al., 2003; Pages et al., 1999; Saba-El-Leil et al., 2003; Yao et al., 2003). These findings in zebrafish
Figure 1.41 Individual inhibition of ERK1 or ERK2 by the kinase-dead method cannot maintain ESC self-
renewal in the presence of GSK3 inhibitor.
(A-B) Representative phase contrast images of two different clones of the Erk1
-/-
+ Erk1 KD ESCs (A) and the
Erk2
-/-
+ Erk2 KD ESCs (B) in N2B27 medium supplemented with PD + CHIR or CHIR for 7 days. Scale bar,
100µm. (C) AP staining-based quantification of the percentages of self-renewing colonies formed in the Erk1
-/-
+
Erk1 KD and the Erk2
-/-
+ Erk2 KD ESC populations cultured in N2B27 medium supplemented with PD + CHIR
or CHIR. Data represent two independent experiments and are shown as means ± SEM. ***: p < 0.001.
62
and in mouse led me to speculate that ERK1/2 may have non-overlapping roles in the lineage
commitment of ESCs into cells in the three germ layers represented by neuroectoderm, mesoderm
and endoderm.
I first investigated the individual roles of ERK1/2 isoforms in ESC differentiation using the
established Erk IR KI ESC lines. Considering the difficulty of drug diffusion into the core of
embryoid bodies and the large variation in the differentiation efficiency in embryoid body-based
differentiation protocols (Brickman and Serup, 2017), I used monolayer differentiation protocols.
For neural differentiation, an adherent monolayer neural differentiation protocol was employed
(Ying et al., 2003b). Since the ESCs used to generate the Erk IR KI ESCs carry a Sox1-GFP knock-
in reporter (Ying et al., 2003b), and Sox1 is the earliest known specific marker of neuroectoderm
in the mouse (Wood and Episkopou, 1999), I was able to detect the Sox1+ neural stem cells (NSCs)
Figure 1.42 Individual inhibition of ERK1 or ERK2 does not block neuroectoderm differentiation.
Representative phase-contrast and fluorescent images (Sox1-GFP) of the indicated ESCs (WT ESCs, the Erk1 IR
KI ESCs c1, and the Erk2 IR KI ESCs c1) subjected to monolayer neural differentiation (details in Materials and
Methods) in N2B27 medium supplemented with PD, Vx11e, or DMSO (as the NT group) for 5 days. The medium
was changed every other day. The Erk1 IR KI ESCs c2 and the Erk2 IR KI ESCs c2 were also subjected to neural
differentiation in N2B27 medium supplemented with PD, Vx11e, or DMSO (as the NT group), and had very similar
results (data not shown). NT: non-treatment with inhibitors. Sox1: the earliest known specific marker of
neuroectoderm in mouse. Scale bar, 100µm.
63
by GFP expression. Results showed that both WT ESCs and the Erk IR KI ESCs readily
differentiated into Sox1-GFP+ neural stem cells (NSCs) in the neural differentiation medium
(Figure 1.42). Inhibition of the MAPK/ERK signaling by PD could not block the formation of
Sox1-GFP+ NSCs in both WT and the Erk IR KI ESC populations when cultured in the neural
differentiation medium. Consistent with this, Sox1-GFP+ NSCs could be generated from WT
ESCs when ERK1/2 were inhibited by Vx11e in these cells. Additionally, both the Erk1 IR KI
ESCs and the Erk2 IR KI ESCs could still produce many Sox1-GFP+ NSCs in the neural
differentiation medium when one of the ERK1/2 isoforms in the Erk IR ESCs was inhibited with
Vx11e (Figure 1.42). These results suggested that inhibition of both ERK1/2 or selective inhibition
of one of the ERK1/2 isoforms could not block neural differentiation of ESCs.
Figure 1.43 Selective inhibition of ERK1 or ERK2 does not block mesoderm differentiation.
Representative immunofluorescent staining images for a mesoderm marker (Brachyury) in WT ESCs, the Erk1 IR
KI ESCs c1, and the Erk2 IR KI ESCs c1 subjected to monolayer mesoderm differentiation (details in Materials and
Methods) in the medium supplemented with PD, Vx11e, or DMSO (as the NT group) for 6 days. The Erk1 IR KI
ESCs c2 and the Erk2 IR KI ESCs c2 were also subjected to differentiation in the mesoderm differentiation medium
supplemented with PD, Vx11e, or DMSO, and had very similar results (data not shown). NT: no treatment with
inhibitors. Scale bar, 100µm.
64
I next investigated the roles of ERK1/2 during mesendoderm differentiation. In mesoderm
differentiation medium, both WT ESCs and the Erk IR KI ESCs produced considerable mesoderm
progenitors (Brachyury+) (Figure 1.43), and the numbers of mesoderm progenitors formed in these
ESCs dramatically decreased when MAPK/ERK signaling was inhibited by PD (Figure 1.43). In
line with the above results, direct inhibition of ERK1/2 by Vx11e in the WT ESCs suppressed their
differentiation into Brachyury+ mesoderm progenitors. In contrast, for the Erk IR KI ESCs, the
supplementation of Vx11e in the mesoderm differentiation medium did not block their
differentiation into Brachyury+ cells (Figure 1.43). These results indicate that specific inhibition
of one of the Erk1/2 isoforms in ESCs cannot block mesoderm differentiation.
As for endoderm differentiation, both WT ESCs and the Erk IR ESCs could differentiate and form
many endoderm progenitors (FoxA2 positive, FoxA2+) in the endoderm differentiation medium
Figure 1.44 Selective inhibition of ERK1 or ERK2 does not block endoderm differentiation.
Representative immunofluorescent staining images for an endoderm lineage marker (FoxA2) in WT ESCs, the Erk1
IR KI ESCs c1, and the Erk2 IR KI ESCs c1 subjected to the monolayer endoderm differentiation (details in
Materials and Methods) under indicated conditions (NT, +PD, +Vx11e) for 6 days. The Erk1 IR KI ESCs c2 and
the Erk2 IR KI ESCs c2 were also subjected to differentiation in the mesoderm differentiation medium supplemented
with PD, Vx11e, or DMSO (as the NT group), and had very similar results (data not shown). NT: no treatment with
inhibitors. Scale bar, 100µm.
65
(Figure 1.44), and the formation of endoderm progenitors were abolished in these ESCs when PD
was supplemented to the differentiation medium. Similarly, FoxA2+ endoderm cells could not be
generated from the WT ESCs in the endoderm differentiation medium supplemented with Vx11e
(Figure 1.44). However, under the same condition, we found many FoxA2+ endoderm progenitors
emerged from the populations of the Erk1 IR KI ESCs and the Erk2 IR KI ESCs (Figure 1.44).
These findings implied that the activation of either Erk1 or Erk2 in ESCs was sufficient to induce
endoderm differentiation.
To further confirm the individual roles of ERK1 and ERK2 in three germ layer differentiation, I
repeated the three germ layer differentiation assays using the Erk1
-/-
+ Erk1 KD ESCs and the
Erk2
-/-
+ Erk2 KD ESCs, in which one of the ERK1/2 isoforms was inhibited. Both of these two
types of transgenic ESCs could differentiate into progenitor cells in the three germ layers based on
immunostaining of different lineage-specific markers (Figure 1.45).
Figure 1.45 Individual inhibition of ERK1 or ERK2 by a kinase-dead approach does not block the commitment
of ESCs to neuroectoderm and mesendoderm.
Representative immunofluorescent staining images for neuroectoderm (Sox1), mesoderm (Brachyury), and endoderm
(FoxA2) lineage makers in Erk1
-/-
+ Erk1 KD ESCs and Erk2
-/-
+ Erk2 KD ESCs subjected to monolayer differentiation
to cells in three germ layers for 6 days. Scale bars, 100µm.
66
Taken together, our findings from the Erk IR KI ESCs demonstrate that selective inhibition of one
of the ERK1/2 isoforms in ESCs cannot block their differentiation into cells of the three germ
layers.
8. Selective Inhibition of ERK1 but Not ERK2 Promotes the Mesendoderm
Differentiation of ESCs
To better determine the roles of ERK1/2 in three germ layer differentiation, we quantified the
expression of lineage-specific markers in the three types of ESCs under different differentiation
conditions by qPCR.
Regarding neural differentiation, the inhibition of MAPK/ERK signaling in the WT ESCs (by PD
or Vx11e) and in the Erk IR KI ESCs (by PD) significantly reduced Sox1 transcription (Figure
1.46), underscoring the importance of activation of MAPK/ERK signaling during neural
differentiation. In addition, addition of PD could not completely inhibit the upregulation of Sox1
in the three types of ESCs, which was consistent with the emergence of Sox1-GFP+ cells when
the MAPK/ERK pathway was inhibited in these ESCs (Figure 1.46). These results suggested that
MAPK/ERK signaling was dispensable for neuroectoderm differentiation, in line with our findings
about the neural differentiation of Erk DKO ESCs using an embryoid body differentiation protocol
(Figure 1.15). Moreover, the selective inhibition of either ERK1 or ERK2 with Vx11e in Erk IR
KI ESCs did not significantly alter Sox1 levels during neural differentiation (Figure 1.46),
67
suggesting that selective inhibition of either ERK1 or ERK2 in ESCs had trivial effects on the
formation of Sox1+ NSCs.
In terms of mesendoderm differentiation, the inhibition of the MAPK/ERK pathway in the WT
ESCs (by PD or Vx11e) and in the Erk IR ESCs (by PD) significantly inhibited the induction of
mesoderm markers (Brachyury, Mixl1) (Figure 1.47A and 1.47B) and endoderm markers (FoxA2,
and Sox17) (Figure 1.48A and 1.48B). Interestingly, the selective inhibition of ERK2 in the Erk1
IR KI ESCs by Vx11e decreased the levels of Brachyury and Mixl1 (Figure 1.47A and 1.47B) but
had minor effects on the transcription of FoxA2 and Sox17 (Figure 1.48A and 1.48B). In contrast,
selective inhibition of ERK1 in the Erk2 IR KI ESCs with Vx11e significantly increased the levels
of mesoderm (Figure 1.47A and 1.47B) and endoderm markers (Figure 1.48A and 1.48B). These
results suggested that activation of ERK1 rather than ERK2 in ESCs negatively regulates
mesendoderm differentiation.
Figure 1.46 ERK1 and ERK2 play redundant roles in neural differentiation.
qPCR analysis of a neuroectoderm marker (Sox1) in WT ESCs, the Erk1 IR KI ESCs c1, and the Erk2 IR KI ESCs
c1 cultured in neural differentiation medium (N2B27 medium) under the indicated conditions for 4 days. The Erk1
IR KI ESCs c2 and the Erk2 IR KI ESCs c2 had similar results (data not shown). The ESCs cultured with 2i were
used as the controls. Data represent two independent experiments and are shown as means ± SEM. ***: p < 0.001.
68
Furthermore, I performed immunoblotting to confirm the roles of individual ERK isoform
inhibition in mesendoderm differentiation. In Erk1 IR KI ESCs, the expression of Brachyury
(Figure 1.47C and 1.47D) and FoxA2 (Figure 1.48C and 1.48D) were lower when ERK2 was
selectively inhibited by Vx11e than when both ERK1/2 were active. In contrast, in Erk2 IR KI
ESCs, the expression of Brachyury (Figure 1.47C and 1.47D), in particular, and FoxA2 (Figure
1.48C and 1.48D) were higher when ERK1 was selectively inhibited by Vx11e than when both
ERK1/2 were active. These results indicated that selective inhibition of ERK1 but not ERK2 in
ESCs promotes mesendoderm differentiation of ESCs.
Figure 1.47 Individual inhibition of ERK1 but not ERK2 promotes mesoderm differentiation.
(A-B) qPCR analysis of mesoderm markers, Brachyury (A) and Mixl1(B), in WT ESCs, the Erk1 IR KI ESCs
c1, and the Erk2 IR KI ESCs c1 cultured in mesoderm differentiation medium under the indicated conditions for
5 days. The Erk1 IR KI ESCs c2 and the Erk2 IR KI ESCs c2 had similar results (data not shown). The ESCs
cultured with 2i were used as the controls. Data represent two independent experiments and are shown as means
± SEM. *: p<0.05; ***: p < 0.001. (C) Western blot analysis of the expression of Brachyury and GAPDH in the
Erk1 IR KI ESCs c1, and the Erk2 IR KI ESCs c1 cultured in mesoderm differentiation medium under indicated
conditions for 6 days. (D) The relative Brachyury levels in the immunoblotting results shown in (C) were
quantified using ImageJ software and normalized to the levels of GAPDH under the same conditions.
69
Based on the above findings, we can conclude that ERK1/2 in ESCs are functionally redundant in
neural differentiation, but are non-redundant in mesendoderm differentiation.
Figure 1.48 Individual inhibition of ERK1 but not ERK2 promotes endoderm differentiation.
(A-B) qPCR analysis of endoderm markers, FoxA2 (A) and Sox17 (B), in WT ESCs, the Erk1 IR KI ESCs c1, and
the Erk2 IR KI ESCs c1 cultured in endoderm differentiation medium under the indicated conditions for 5 days.
The Erk1 IR KI ESCs c2 and the Erk2 IR KI ESCs c2 had similar results (data not shown). The ESCs cultured
with 2i were used as the controls. Data represent two independent experiments and are shown as means ± SEM.
***: p < 0.001. (C) Western blot analysis of the expression of FoxA2 and GAPDH in the Erk1 IR KI ESCs c1, and
the Erk2 IR KI ESCs c1 cultured in endoderm differentiation medium under the indicated conditions for 6 days.
(D) The relative FoxA2 levels in the immunoblotting results shown in (C) were quantified using ImageJ software
and normalized to the levels of GAPDH under the same conditions.
70
Discussion
In the present study, by combining genetic and chemical-genetic approaches with ESC-based
technologies, we demonstrated that ERK1/2 are dispensable for the survival and expansion of
ESCs and that inhibition of ERK1/2 is sufficient to mimic the effect of MEK inhibitor PD0325901
in promoting ESC self-renewal when GSK3 is inhibited. We have also developed a novel inhibitor-
resistant (IR) approach to achieve selective inhibition of individual ERK isoforms. Our results
suggest that the two ERK isoforms play both redundant and non-redundant roles in regulating ESC
fate. Genetic deletion or chemical inhibition of either ERK isoform is not sufficient to promote
ESC self-renewal or block neural differentiation of ESCs. For mesendoderm differentiation of
ESCs, however, selective inhibition of ERK1 plays different roles compared with selective
inhibition of ERK2. Our study has provided novel insights into the roles of ERK isoforms in
regulating cell fate.
A previous study claimed that ERK1/2 are indispensable for genomic stability and self-renewal of
mouse ESCs (Chen et al., 2015). However, our results are contrary to this claim. We demonstrated
that deletion of both ERK isoforms in mouse ESCs does not impact their survival and self-renewal
ability. In fact, mouse ESCs could be maintained by Vx11e + CHIR, further supporting the idea
that ERK1/2-dependent signaling is dispensable for mouse ESC self-renewal. We also showed that
ERK1/2 are dispensable for the genomic stability of mouse ESCs. The differences in the genetic
backgrounds of the mouse ESCs used may account for the discrepancy in the two studies. In the
future, we may screen different mouse ESC lines with the ERK inhibitor (e.g. Vx11e) first to better
define the dispensability of ERK1/2 in mouse ESCs.
Our findings that ERK1 and ERK2 play redundant and non-redundant roles in ESC self-renewal
and lineage differentiation are likely to have broad implications. Firstly, manipulation of the
71
MAPK/ERK pathway via individual chemical inhibition alters the efficiency of lineage
commitment without changing cell fates, at least for the formation of progenitor cells in the three
germ layers. This could be very useful in cell therapy in regenerative medicine, as the efficiency
of cell differentiation can be greatly enhanced via individual inhibition of one ERK isoform.
Secondly, the different roles of ERK isoforms in ESC self-renewal and differentiation highlight
the complexity of the MAPK/ERK pathway. Especially considering the critical roles of
MAPK/ERK signaling in the physiology of multiple biological processes and the pathogenesis of
diverse diseases, the specific roles of ERK isoforms in other biological processes warrant further
exploration (Kim and Choi, 2010; Roskoski, 2012). We anticipate that individual inhibition of
ERK1 and ERK2 will produce distinct changes in other cellular and developmental processes. This
is likely true since the deletion of either ERK1 or ERK2 has yielded different phenotypes in the
generation of various cell types including hepatocytes, adipocytes, cancer cells, and so on (Aceves-
Luquero et al., 2009; Bost et al., 2005; Cho et al., 2002; Fremin et al., 2009; Fremin et al., 2012;
Gusenbauer et al., 2015; Li and Johnson, 2006; Radtke et al., 2013; Shin et al., 2013; Shin et al.,
2010; von Thun et al., 2012). It would be interesting to see whether selective inhibition of one
ERK isoform could replicate the above results obtained from the knock-out approach. Thirdly, the
specific roles of each ERK isoform in different cell types and diseases could be readily explored
using ESCs or patient derived iPSCs since pluripotent stem cells preserve the ability to
differentiate into any cell type in the body. Defining ERK isoform-specific phenotypes will
provide rationale for developing inhibitors individually targeting each ERK isoform.
The use of the chemical genetic approach allows for deciphering the individual roles of highly
homologous kinases. To use the chemical-genetic approach in kinases, a kinase needs to be
engineered at the gatekeeper position so that the mutated kinase, termed as an analog-sensitive
72
(AS) kinase, is uniquely sensitive to an analog inhibitor whereas the wild-type kinase(s) does not
recognize the inhibitor. Given that the gatekeeper residue is located in the kinase domain and has
direct contact with the N6 group of ATP, the introduced mutation at this position may render the
complete loss, or at least reduction, of kinase catalytic activity (Zhang et al., 2005). The intolerant
kinases that have been reported include Cdc5, Mekk1, Grk2, Pto, and so on, with ERK1 as a new
member of this list of intolerant kinases (Endo et al., 2006). Though a second-site suppressor
strategy may work for some kinases to ameliorate kinase activity, changes in substrate specificity
cannot be overlooked. Another issue with the current chemical genetic method regards off-target
effects of bump inhibitors. The majority of bump inhibitors are developed based on PP1, which
was initially identified as a potent ATP-competitive inhibitor of Src family kinases (Zhang et al.,
2013). As a result, bump inhibitors may inhibit those kinases while targeting analog mutants. In
addition, it is possible that the off-target effects of bump inhibitors confound the phenotypes
observed, a likelihood that will increase if the studied phenotypes relate to Src family kinases (we
confronted this issue in ESC self-renewal study) (Shimizu et al., 2012). One way to solve this
problem is to develop a new class of inhibitors that match with a new form of protein mutants (e.g.
Ele-Cys approach) (Kung et al., 2017). Another solution is to use our novel IR strategy. The IR
approach has at least two advantages. Firstly, IR mutants are screened by a phenotype based assay.
Hence, it is likely that IR mutants preserve their kinase activity and substrate specificity. Secondly,
we do not need an extra inhibitor. Particularly since the original inhibitors are still used in their
normal ranges, any off-target effects can be minimized. Theoretically, the IR strategy can be
utilized to explore the roles of any other kinases but in particular, the highly homologous kinases.
Additionally, the combination of our IR strategy with traditional chemical genetic approaches (e.g.
bump-hole approach and Ele-Cys method) enables us to individually inhibit two highly
73
homologous kinase isoforms in one type of cell, circumventing the troublesome option of
establishing two separate cell lines for the study. One caveat of the IR approach would be that this
method requires more time and expenses in screening and identifying the mutants. However, the
structural biology study on the functional mechanisms of inhibitors would provide useful clues for
designing IR mutants and thus shorten the time required for screening (Chaikuad et al., 2014).
74
Perspective and Future Directions
By studying the two highly homologous ERK1/2 isoforms, we revealed the redundant and non-
redundant roles of ERK1/2 isoforms in ESC self-renewal and differentiation. Due to the lack of
ERK1/2 isoform specific inhibitors, it is impossible to determine the isoform-specific roles of
ERK1/2 in wild-type ESCs using conventional chemical inhibitors. With our established ERK IR
KI ESC lin, we can selectively inhibit one of the two ERK1/2 isoforms in ESCs with the ERK
inhibitor, Vx11e, and define the isoform-specific roles of ERK1/2 in ESC self-renewal and
differentiation.
The execution of the diverse functions of the MAPK/ERK pathway relies on the phosphorylation
of different substrates of ERK. We can define ERK1/2 isoform-specific substrates in ESCs by a
combination of phosphoproteomics (Carlson et al., 2011) and selective inhibition of ERK1/2
isoforms in our established ERK IR KI ESCs. The identification of the direct substrates of each
ERK isoform in ESCs will enrich our knowledge about how substrates of ERK1/2 mediate the
important roles of MAPK/ERK signaling in ESC self-renewal and differentiation. In addition, we
can narrow down the substrate lists by finding common subsets of ERK1 and ERK2 substrates,
which will help us define key substrates for ESC self-renewal and differentiation. Furthermore,
combining transcriptomic profiling of ERK1/2 isoform specific regulatory mechanisms with ERK
substrates, we can better characterize the signaling networks underlying ERK inhibition in the
maintenance of ESC self-renewal in the presence of CHIR.
ESCs can self-renew indefinitely while retaining the ability to produce any cell type in different
tissues. We can now readily explore ERK1/2-isoform dependent roles in terminally differentiated
cells and in different types of stem cells and somatic cells. Moreover, using engineered ESCs
carrying causative mutations for certain diseases, we can investigate and define ERK1/2 isoform-
75
specific functions in different diseases. Furthermore, we can apply the ERK inhibitor resistant
strategy in human pluripotent stem cells and decipher ERK1/2-isoform specific roles in human
physiology and pathological diseases, which will provide rationale for developing inhibitors
selectively targeting each ERK isoform.
Last but not the least, we can identify isoform-specific downstream target genes via transcriptomic
and proteomic approaches and establish ERK-isoform specific reporter cell lines. With ERK1/2-
isoform specific reporter cell lines, we would be able to monitor the relative contribution of each
ERK isoform to the overall activity of MAPK/ERK signaling in the phenotype of our interest,
which will accelerate studies defining ERK-isoform specific phenotypes. In addition, these
ERK1/2 isoform-specific reporter cell lines will enable us to perform drug screening and develop
ERK1/2 isoform-specific drugs to better target diseases in which ERK1/2 isoforms play distinct
roles.
76
Materials and Methods
Cell Culture
Mouse ESCs were maintained under feeder-free condition and cultured on 0.1% gelatin-coated
plates at 37 °C in a 5% CO
2
incubator. The 0.1% gelatin solution was prepared by diluting 22 mL
of 1% gelatin stock solution into 200 mL of 1x phosphate-buffered saline (PBS) solution. The
gelatin stock solution was made by dissolving 5 grams of gelatin powder (Sigma, St. Louis, MO)
in 500 mL double-distilled water (ddH2O) and then autoclaving. The basal medium (Table 1.1)
was prepared by mixing high glucose DMEM medium (Thermo Fisher Scientific, Waltham, MA)
with Hyclone fetal bovine serum (Thermo Fisher Scientific, Waltham, MA), sodium pyruvate
(Sigma, St. Louis, MO), MEM non-essential amino acids (Thermo Fisher Scientific, Waltham,
MA), and b-mercaptoethanol (Sigma, St. Louis, MO).
Table 1.1 Basal medium composition for mouse ESC culture
Components (stock concentration) Volume Final concentration in medium
High glucose DMEM with L-glutamine 500mL -
Fetal bovine serum 50mL 10%
MEM NEAA (100x) 50mL 1%
Sodium Pyruvate (100mM) 5.5mL 1mM
b-Mercaptoethanol (0.1M) 550µL 0.1mM
To routinely maintain 46c mouse ESCs, which is a cell line derived from the 129 strain of mouse,
10ng/mL leukemia inhibitory factor (LIF; PeproTech, Rocky Hill, NJ) was added to the basal
medium. For the maintenance of 46c mouse ESCs under a serum-free condition, 1µM PD0325901
77
(PD; Selleckchem, Houston, TX) and 3µM CHIR99021 (CHIR; Selleckchem, Houston, TX) were
added to the N2B27 medium (Table 1.2). Serum-free N2B27 medium was prepared by mixing
Neurobasal medium (Thermo Fisher Scientific, Waltham, MA) with DMEM/F12 medium
(Thermo Fisher Scientific, Waltham, MA) at a ratio of 1:1 and then adding N-2 supplement
(Thermo Fisher Scientific, Waltham, MA), B-27
TM
supplement (Thermo Fisher Scientific,
Waltham, MA), L-glutamine (Thermo Fisher Scientific, Waltham, MA), and b-mercaptoethanol
(Sigma, St. Louis, MO).
Table 1.2 Composition of N2B27 medium
Components (stock concentration) Volume Final concentration in medium
DMEM/F12 medium 500mL -
Neurobasal medium 500mL -
N-2 supplement (100x) 5mL 0.5% (0.5x)
B-27
TM
supplement (50x) 10mL 1% (0.5x)
L-glutamine (200mM) 10mL 2mM
2-Mercaptoethanol (0.1M) 1000µL 0.1mM
Cells were passaged when confluency reached 60-70%, which took 2-4 days in most cases. To
detach cells from culture plates, a 0.025% trypsin solution (Thermo Fisher Scientific, Waltham,
MA) containing 1% chicken serum (Thermo Fisher Scientific, Waltham, MA) was used. After
neutralization of trypsin with serum-containing medium and collection of cells by centrifugation
(300g, 3min), cells were sub-cultured at a split ratio of 1: (6-10).
78
Genome Editing in Mouse ESCs
To generate knock-out ESC lines, we employed CRISPR/Cas9 technology with the PX330 vector
(plasmid # 42230; Addgene, Cambridge, MA) co-expressing Cas9 protein and guide RNA (gRNA).
To carry out screening in PX330-transfected cell pools, a P2A-pac cassette expressing puromycin
was inserted in-frame with Cas9 into the PX330 vector using the FseI and EcoRI enzymes.
Two different gRNAs specifically targeting exon 2 of Erk1 and exon 3 of Erk2, respectively, were
designed by an online gRNA-designing tool (crispr.mit.edu). The sequences of these two gRNAs
(NGG not included) are listed in Table 1.3. Sense and antisense DNA oligonucleotides containing
two different gRNA sequences for knocking out Erk were synthesized by Integrated DNA
Technologies (IDT) Incorporation and cloned into the PX330-P2A-pac vector using two BbsI
restriction enzyme sites.
To generate 46c ESC lines with the deletion of both ERK1 and ERK2, approximately 5x10
5
46c
ESCs were plated into each well in a 12-well plate, transfected with either PX330 control plasmids
(2.5µg/well) or the Erk1/2 gRNA PX330-P2A-pac constructs (2.5µg/well) using Lipofectamine®
LTX & Plus Reagent (Life Technologies, Grand Island, NY) according to the manufacturer’s
instructions. Next morning, the Opti-MEM medium (Life Technologies, Grand Island, NY) used
for transfection was replaced by fresh ESC culture medium (LIF plus serum) containing 1 µg/ml
puromycin for selection. After selection with puromycin for 72-96 hours, surviving cells were
trypsinized and seeded at a clonal density (~800 cells/well in a six-well plate) for expansion under
2i condition. After one week of culture, dome-shaped colonies were individually picked and
expanded for further analysis. Candidate ESC clones with both Erk1/2 knocked out (Erk DKO
ESCs) were first screened by detecting the expression of ERK1/2 with Western blot and then
confirmed by DNA sequencing on the genomic regions of Erk1/2 targeted by the gRNAs. The
79
primers for PCR gRNA targeted genomic regions of Erk1 and Erk2 in Erk DKO ESCs are listed
in Table 1.3.
Table 1.3 DNA oligonucleotides for genome editing with CRISPR/Cas9 technique
Name Sequences (5’-3’)
Erk1 gRNA for knock-out TATATACTTGAGGCCCCGG
Erk2 gRNA for knock-out TTTGCTCAATGGTTGGTGCC
Erk1 knock-out genotyping-F TCCTTTTGAGCACCAGACCT
Erk1 knock-out genotyping-R AGCGCAAACATTATAATCTCCTCT
Erk2 knock-out genotyping-F TCAAGTCTCAGTGTAGGCCC
Erk2 knock-out genotyping-R CACTGTCACTGTGAGCCTTGT
Erk1 G55A KI gRNA TACATCGGCGAGGGCGCGTA
Erk2 G35S KI gRNA TACTCACCAAACCATGCCGT
Erk1 G55A KI ssODN ctccctgagccccccatgtcccgagcacacacctgctgtccctgctccgaagccccctcccgg
gacgccccctcacctgaccat ggcgtacgcgccctcgccgatgtactgcagctgcgtgtagcgt
Erk2 G35S KI ssODN gcggcggcgggcccggagatggtccgcgggcaggtgttcgacgtagggccgcgctacaccaa
cctctcgtacatcggagaaggcgcatacagtatggtttggtgagtatccgcgctggatttcaggc
Erk1 G55A KI genotyping-F GGCCTAAAGGCAGGAGGATG
Erk1 G55A KI genotyping-R GCAGCGAGAACTCACAAACC
Erk2 G35S KI genotyping-F TGTGGGGTCCTTATGCCTAAAT
Erk2 G35S KI genotyping-R CAGTGTACTTCCGTTCCCGT
80
To generate Erk1 knock-out ESCs (Erk1
-/-
ESCs) and Erk2 knock-out ESCs (Erk2
-/-
ESCs),
approximately 5x10
5
46c ESCs were plated into each well of a 12-well plate and transfected with
Erk1 gRNA PX330-P2A-pac plasmids and Erk2 gRNA PX330-P2A-pac plasmids, respectively,
using Lipofectamine® LTX & Plus Reagent (Life Technologies, Grand Island, NY). Similar to
the generation of Erk DKO ESCs, selection by puromycin was performed, followed by clonal
expansion and colony picking. Candidate clones of Erk1
-/-
ESCs and Erk2
-/-
ESCs were also
confirmed by detecting the expression of ERK1/2 with Western blot and by DNA sequencing on
the genomic region of Erk1 or Erk2 targeted by the gRNAs. The primers used for genotyping PCR
in Erk1
-/-
ESCs and Erk2
-/-
ESCs were the same as these primers for genotyping PCR in Erk DKO
ESCs.
To generate Erk1 G55A knock-in (KI) ESC lines and Erk2 G35S KI cell lines, we used
CRIPSR/Cas9-mediated homologous recombination. The gRNAs were designed based on the
results from the online software CRISPOR (crispor.tefor.net). To increase the knock-in efficiency,
we selected gRNAs targeting DNA sequences in proximity to our intended point mutation sites.
The gRNA sequences used to generate Erk1 G55A KI and Erk2 G35S KI cell lines are listed in
Table 1.3, and the sense and antisense oligonucleotides containing the two gRNA sequences were
synthesized by IDT Incorporation. After annealing, the two gRNAs were cloned into the PX330-
P2A-pac vector. Additionally, donors used as the templates for introducing the ERK1 G55A
mutation and the ERK2 G35S mutation were synthesized as single strand DNA (ssODN) by IDT
Incorporation. We also introduced a new restrictive enzyme digestion site, which could not be
found in the genotyping products of wild-type alleles, into the donors for KI to facilitate our
identification of the clones with the mutated allele(s). The sequences of two donors are listed in
Table 1.3.
81
A DNA mixture of ssODN donor (200pmol) and PX330-P2A-puro containing the gRNA for Erk
KI (5µg) was made on ice and electroporated into (1-2) x10
6
46c mouse ESCs using
Nucleofector™ Kits for Mouse Embryonic Stem Cells (Lonza, Allendale, NJ) according to the
instructions of the manufacturer. Electroporated ESCs were plated in a 3.5mm dish and cultured
with LIF plus serum. After overnight culture, the medium was changed with fresh ESC culture
medium (PD + CHIR condition) supplemented with puromycin (1µg/ml) for selection. The
medium was changed every other day and drug selection lasted for 5-7 days. The surviving,
expanded colonies were picked, trypsinized, and split into two 0.1% gelatin-coated 96-well plates
where cells at the same position of the two plates were from the same colony. After one week of
expansion, one of the two 96-well plates was chosen, and genomic DNA was isolated from these
cells with the Quick-DNA 96 Kit (Zymo Research, Irvine, CA) according to the manufacturer’s
protocol.
Genotyping PCR was carried out with the designed primers (Table 1.3) to amplify the region
surrounding the gRNA targeting site. The genotyping PCR products were first screened with a
restriction fragment length polymorphism (RFLP) assay to identify the candidate clones with an
insertion of the intended point mutation. In the RFLP assay, the genotyping PCR products from
the mutated allele (Erk1 G55A or Erk2 G35S) were cut into two smaller pieces since a new
restriction enzyme digestion site was inserted. In contrast, genotyping PCR products from wild-
type alleles would remain the same size. Running PCR products on the agarose gel after the RFLP
assay enabled us to identify the potentially positive clones with the correct point mutation insertion.
For candidate clones with the correct point mutation insertion, we recorded their positions and
passaged clones of the same location in the second 96-well plate for expansion. The expanded
clones were again confirmed by a RFLP assay. Genotyping PCR products of homozygous clones
82
that had point mutation insertions in both alleles were further confirmed by DNA sequencing.
Clones with the correct sequencing were further expanded and established as KI ESC lines.
Establishment of Erk Transgenic Embryonic Stem Cell Lines
Total RNA was isolated from 46c ESCs using Quick-RNA MiniPrep Kit (Zymo Research, Irvine,
CA) and then 0.5-1µg RNA was transcribed into cDNA using TaqMan™ Reverse Transcription
Reagents (Invitrogen, Carlsbad, CA) following the manufacturers’ instructions. Coding sequences
(CDS) of mouse Erk1 and Erk2 were amplified from 46c ESC cDNA by PCR with PrimeSTAR
GXL DNA Polymerase (Clontech, Mountain View, CA), and PCR products were then cloned into
T vectors using TOPO™ TA Cloning™ Kit (Thermo Fisher Scientific, Waltham, MA) following
the manufacturer’s instructions. Sequences of Erk1 and Erk2 were confirmed by DNA sequencing
of cloned T vectors.
For the establishment of Erk transgenic ESC lines, SBI’s PiggyBac Transposon System (System
Bioscience, Palo Alto, CA) was used. Two PiggyBac (PB) transposon-based plasmids (PB
Transposon vector #PB511B-1, PB Transposase vector #PB200PA-1) were purchased. PB
transposase recognizes the transposon-specific inverted terminal repeats (ITRs) located on both
ends of the PB transposon vectors and efficiently integrates DNA sequences between the two ITRs
(including themselves) into the genome (Yusa et al., 2011). The CMV-MCS-EF1a-pac dual
promoter cassette between the two ITRs in the PB Transposon vector was replaced by CAG-MCS-
IRES-bsd or CAG-MCS-IRES-BleoR using NheI and ClaI. These modifications ensured the high
transgene expression levels, which were driven by the CAG promoter, and facilitated the screening
of positive cells with PB Transposon vectors via different drugs (Blasticidin for the bsd vector and
Zeocin for the BleoR vector). Additionally, a turboRFP-P2A sequence was inserted into multiple
cloning sites (MCS) of the PB-CAG-MCS-IRES-bsd/BleoR backbone, and dual restrictive
83
enzyme sites (SbfI and AscI) were inserted after the P2A sequence for the cloning of Erk mutants
into the modified PB Transposon vectors. The turboRFP-P2A-Erk cassette allowed for the tracing
of ERK protein expression under a fluorescent microscope based on the expression of red
fluorescent protein (RFP) in Erk transgenic ESC lines.
A one step mutagenesis protocol was used to introduce individual Erk mutation into wild-type
(WT) Erk1/2 sequences that had been cloned into T vectors. In this mutagenesis protocol, a pair
of primers carrying the intended mutation were ordered from IDT Incorporation. PCR was
performed using T vectors carrying the wild-type Erk1/2 sequences as templates, with the primers
introducing mutation at the same time. After PCR, DNA templates with WT Erk sequences were
digested and removed by a DpnI enzyme. The remaining PCR products formed T vectors carrying
Erk mutant sequences after ligation with a T4 ligase. The Erk sequences in the T vectors were
confirmed by DNA sequencing.
The following mutations were introduced into Erk sequences and used in this study: Erk1 Y54H,
Erk1 G55A, Erk1 P76L, Erk1 K72R, Erk2 Y34N, Erk2 G35S, Erk2 P56L, and Erk2 K52R. Among
the generated ERK mutants, the ERK1 K72R mutant and the ERK2 K52R mutant were also called
ERK1/2 kinase-dead (KD) mutants. DNA sequences of wild-type Erk and mutated Erk were cut
out from confirmed T vectors using SbfI and AscI and then inserted into modified PB Transposon
vectors. Erk1 and Erk2 were cloned into PB-CAG-turboRFP-P2A-MCS-IRES-BleoR and PB-
CAG-turboRFP-P2A-MCS-IRES-bsd, respectively.
For the generation of Erk transgenic ESC lines, approximately 5 x 10
5
ESCs were plated into each
well of a 12-well plate, transfected with both PB Transposase vectors (1.25µg/well) and the
aforementioned modified PB Transposon vectors containing the sequences of Erk mutants
(2.5µg/well) using Lipofectamine® LTX & Plus Reagent (Life Technologies, Grand Island, NY)
84
according to the manufacturer’s instructions. After 16-24 hours of culture, Zeocin (75µg/mL;
Thermo Fisher Scientific, Waltham, MA) and Blasticidin S (10µg/mL; ThermoFisher Scientific,
Waltham, MA) were used for the screening of Erk1 transgenic ESCs and Erk2 transgenic ESCs,
respectively. Medium was changed every 2-3 days. After drug selection for 5-7 days, different
transgenic ESC pools were split and plated in different wells of 6-well plates at a density of 1000
cells/well. After expansion of cells under the 2i condition, red fluorescent protein positive (RFP+)
colonies were picked under a fluorescence microscope and further expanded to establish different
clones of Erk transgenic ESC lines. The expression of ERK mutants in transgenic ESCs was
confirmed with Western blot.
Blastocyst Injection and Chimera Formation
Blastocyst injection was performed by the staff members of the Transgenic/Knockout Core
Facility at USC Norris Comprehensive Cancer Center. Blastocysts from E3.5 timed-pregnant
C57BL/6 mice (Charles River, Wilmington, MA) were placed into a droplet of M2 medium and
incubated in M16 medium (Sigma, St. Louis, MO) for 2–3 hours, preparing the well-expanded
blastocysts for microinjection. Seven to fifteen mouse ESCs of Erk2 WT-transgenic ESC line (Erk
DKO + Erk2 WT) were injected into each blastocyst and incubated at 37°C for 1 hour in M16
medium to allow for the recovery of embryos. Eight to ten embryos were then transferred into the
uterine horn of each E3.5 pseudopregnant female mouse. In total, we used two pseudopregnant
recipient mice for the transplantation of embryos.
Embryo-transplanted female mice were sacrificed when the embryos reached E12.5. Embryos
were flushed from the uterine with 1x PBS solution and washed once in 10 cm petri-dishes. The
contribution of ESCs to chimera formation was analyzed by checking the expression of RFP in all
85
embryos under a Fluorescence Stereo Zoom Microscope (Carl Zeiss Microscopy, Thornwood,
NY). The RFP+ embryos were then counted and imaged.
Self-Renewal Assay and Alkaline Phosphatase Staining
ESCs were plated at a density of (500-1000) cells/well in 12-well plates coated with 0.1% gelatin
in ESC maintenance medium (LIF + serum). Next morning, the serum medium was changed to
serum free N2B27 medium supplemented with inhibitors. The inhibitors used in this study
included PD, CHIR, and Vx11e (APExBIO, Houston, TX), and the conditions I tested were PD +
CHIR, Vx11e + CHIR, and CHIR. For each tested condition, cells of different ESC lines were
plated in triplicate. Culture medium was changed every two days. After six days of culture, cells
were washed with 1x PBS solution and stained for alkaline phosphatase using an Alkaline
Phosphatase Detection Kit (Sigma, St. Louis, MO) according to manufacturer’s instructions. At
least two independent self-renewal assays were performed for ESCs of different genotypes.
In the AP staining assay, undifferentiated ESC colonies are stained in red (termed AP+) because
of the expression of alkaline phosphatase on the cell membrane; in contrast, differentiated colonies
cannot be stained (termed AP-) due to the loss of alkaline phosphatase over the course of
differentiation. Dome-shaped AP+ colonies with clear boundaries were classified as self-renewing
colonies, and the colonies with AP- cells and rough boundaries were designated as differentiated
colonies. After the AP staining assay, the colonies in a whole well were counted and classified into
self-renewing colonies and differentiated colonies. The percentage of self-renewing colonies for
each well was calculated for further statistical analysis.
86
Embryoid Body-Based ESC Differentiation
A 4-/4+ retinoic acid-based embryoid body (EB) differentiation protocol was used to evaluate the
differentiation potential of ESCs. For EB formation, ESCs were plated in 35mm petri-dishes
(Genesee Scientific, El Cajon, CA) at a density of (5-10) ×10
5
cells/dish and cultured in basal
medium without LIF or inhibitors. EB aggregates were cultured in suspension for 8 days, and
0.1mM retinoic acid were supplemented to EB culture from day 4 to day 8. To change medium,
EB aggregates were collected by the centrifugation (100g, 3min), and cell pellets were gently
suspended with fresh medium for culture. For each ESC lines, experiments were performed in
triplicate. EB samples (50~100 EBs for each condition) were collected on day 4 and day 8 for
quantitative real-time PCR analysis.
For further three-germ layer differentiation assays, EB aggregates (4-5 EBs/well) were transferred
to laminin-coated Nunc
TM
4-well dishes (Thermo Fisher Scientific, Waltham, MA) for
neuroectoderm differentiation in N2B27 medium and to gelatin coated Nunc
TM
4-well dishes for
mesendoderm differentiation in basal medium with serum. For differentiation into each germ layer,
experiments were quadruplicated. Differentiated cells were further analyzed by immunostaining.
In Vitro Monolayer Differentiation of ESCs
For neural differentiation, ESCs were plated at 2 x10
3
/cm
2
in LIF + serum medium on the plates
coated with laminin (1µg/ml; Life Technologies, Grand Island, NY). After 24 hours, cells were
cultured with serum free N2B27 medium. Medium was changed every other day. The cells under
neural differentiation were kept in N2B27 medium for 4 days to perform qPCR analysis and for 6
days to check green fluorescent protein positive (GFP+) cells under a fluorescence microscope.
87
For mesoderm differentiation, cells were plated at 2 x10
3
/cm
2
in basal medium with LIF on 0.1%
gelatin-coated plates. After 24 hours, cells were cultured with N2B27 medium containing 20 ng/ml
Activin A (PeproTech, Rocky Hill, NJ), 5 ng/ml bFGF (PeproTech, Rocky Hill, NJ) and 1 µg/ml
Heparin (Sigma, St. Louis, MO). After four days of culture in N2B27 medium, 3 µM CHIR was
added into N2B27 medium together with Activin A and bFGF. Cells undergoing mesoderm
differentiation were kept in N2B27-based medium for 5 days to perform qPCR analysis and for 6
days to perform immunostaining.
For definitive endoderm differentiation, cells were plated at 5 x 10
3
/cm
2
in basal medium with LIF
on 0.1% gelatin-coated plates. For the first two days, N2B27 medium containing 20 ng/ml Activin
A, 3 µM CHIR, 10 ng/ml FGF4 (PeproTech, Rocky Hill, NJ), 1 µg/ml Heparin, and 100 nM PI103
(Cayman Chemical, Ann Arbor, Michigan) was used for inducing differentiation. After three days
of differentiation, cells were switched to SF5 basal medium containing 20 ng/ml Activin A, 3 µM
CHIR, 10 ng/ml FGF4, 1 µg/ml Heparin, 100 nM PI103, and 20 ng/ml EGF (PeproTech, Rocky
Hill, NJ). 100 ml SF5 basal medium was made by mixing 98.5ml DMEM/F12 medium (Thermo
Fisher Scientific, Waltham, MA) with 500 µl N-2 supplement (100X) (Thermo Fisher Scientific,
Waltham, MA), 1 ml B-27
TM
without Vitamin A supplement (50X) (Thermo Fisher Scientific,
Waltham, MA), 1g bovine serum albumin (BSA; 1%, final concentration) (CST, Danvers, MA),
and 100 µl β-mercaptoethanol. For cells undergoing endoderm differentiation, qPCR analysis was
performed on day 5 and immunostaining on day 6.
Western Blot
Before cell lysate collection, the cells were washed twice with an ice-cold 1x PBS solution to
remove the dead cells and culture medium. The cell lysis buffer used for collecting protein lysates
from cells was prepared by mixing the RIPA lysis buffer (Teknova, Hollister, CA) with
88
cOmplete™ Protease Inhibitor Cocktail (Roche, Pleasanton, CA) and PhosSTOP -Phosphatase
Inhibitor Cocktail (Roche, Pleasanton, CA) following the manufacturer’s instructions. To lyse the
cells, an ice-cold lysis buffer (100µl per 1x10
6
cells) was added. Cells were scraped off the plate
using a cold plastic cell scraper, and the cell suspension was gently transferred into a pre-cooled
microcentrifuge tube. The cell suspension was constantly agitated at 4°C every 5 min and in total
for 30min followed by centrifugation (16000g, 45min) at 4°C. The supernatant was then
transferred to a new ice-cold tube for protein quantification.
Protein concentrations were determined using the Piece BCA Protein Assay Kit (Thermo Fisher
Scientific, Waltham, MA) in a 96 well plate according to the manufacturer’s instructions. Plates
were read with SoftMax Pro software using a SpectraMax i3x microplate reader (Molecular
Devices, San Jose, CA). For each sample, approximately 30µg of protein was mixed with the
appropriate volume of 5x Laemmli Sample buffer and b-Mercaptoethanol (5% final concentration),
heated (100°C, 8min), and separated on a 10% SDS-polyacrylamide electrophoresis gel. After
electrophoresis, proteins were transferred onto a PVDF membrane (Biorad, Hercules, CA), which
was then blocked with 5% nonfat milk (blocking-grade; Biorad, Hercules, CA) in 1x Tris-buffered
saline (TBS) buffer (150mM NaCl, 50mM Tris-HCl, pH 7.9) for 1h at room temperature.
Following blocking, the membrane was incubated overnight at 4°C with primary antibodies that
were diluted with 5% BSA in 1x TBS buffer. After washing the membrane with 1x TBS buffer
containing Triton X-100 (0.1%), species-specific and HRP-conjugated secondary antibodies
(Santa Cruz, Santa Cruz, CA) were diluted (1:1000) in blocking solution and incubated with the
membrane for 1h at room temperature. Primary antibodies and their dilutions used in the study
were as follows: ERK1/2 (1:1000, CST, #9102S), Phospho-p44/42 MAPK (ERK1/2)
89
(Thr202/Tyr204) (1:1000, CST, #4370), GAPDH (1:2000, CST, #5174), RSK1/2/3 (1:1000, CST,
#9355T), and Phospho-p90RSK (Ser380) (1:300, CST, #9341).
For the detection of proteins, the Pierce™ ECL Western Blotting Substrate (Thermo Fisher
Scientific, Waltham, MA) was applied on the antibody-incubated PVDF membrane according to
the manufacturer’s recommendation. Chemiluminescent signals were captured by a CCD camera-
based imager FluorChem E (ProteinSimple, San Jose, CA). In the case of weak signals, the
SuperSignal™ West Femto Maximum Sensitivity Substrate (Thermo Fisher Scientific, Waltham,
MA) was used instead. ImageJ (https://imagej.nih.gov/ij/) was used to compare the densities of the
Western blot bands.
Immunostaining
Cells were first fixed in 4% paraformaldehyde for 15 min at room temperature and then incubated
with the blocking buffer (1x PBS solution with 5% BSA and 0.2% TritonX-100) at 37°C for 1
hour. After fixation and perforation, cells were incubated at 4°C overnight with primary antibodies
diluted in blocking buffer and subsequently washed three times with a 1x PBS solution for 5 min
per wash. Cells were then incubated with secondary antibodies at 37°C for 1 h (or at 4°C overnight).
Nuclei were counter-stained with Hoechst 33342 (1:5000; Thermo Fisher Scientific, Waltham,
MA) at room temperature for 10 min. Before imaging, cells were washed three times with a 1x
PBS solution for 5 min per wash. Immunofluorescence and phase contrast images were acquired
on a BZ-X700 All-in-one Fluorescence Microscope (KEYENCE, Itasca, IL) and processed using
BZX Analyzer software (KEYENCE, Itasca, IL).
The primary antibodies and their dilutions used in this study were as follows: Rex1 (1:100, Santa
Cruz, sc-50669), Oct4 (1:200, Santa Cruz, sc-5279), Nanog (1:1000, R&D, AF2729-SP), β3-
tubulin (Tuj1) (1:200, CST, #4466S), Myosin (1:50, DSHB, MF-20), Brachyury (1:1000, Abcam,
90
ab209665), and FoxA2 (1:1000, CST, #8186T). Ig-subtype- and species-specific Alexa Fluor 488,
546 and 647 conjugated secondary antibodies (1:1000; Thermo Fisher Scientific, Waltham, MA)
were used.
To quantify the immunostaining results, 7 fields were taken at random for each condition and
analyzed by CellProfiler (cellprofiler.org). The DAPI channel (Hoechst staining) was used to
identify nuclei of live cells before staining, and the total number of cells was automatically counted
by CellProfiler. The threshold fluorescence for positive/negative staining was determined by
staining samples with secondary antibodies only, allowing for the number of cells with positive
staining to be quantified by the software. The percentage of positive staining cells was calculated
via dividing the total number of live cells (Hoechst positive) by the number of positive staining
cells (e.g. Nanog positive). At least two independent experiments were performed. Data were
shown as mean ± SEM.
Quantitative Real-Time PCR (qRT-PCR)
Total RNA was extracted from cells using a Quick-RNA MiniPrep Kit (Zymo Research, Irvine,
CA), and cDNA was transcribed from 1µg RNA in a 20µL reaction system using an iScript cDNA
Synthesis Kit (Biorad, Hercules, CA) according to the manufacturers’ instructions. All samples
within an experiment were reverse transcribed at the same time, and the resulting cDNA was
diluted 1:3 in nuclease-free water and stored in aliquots at -80°C until needed for use. For each
sample, wells were used in triplicate. All primers of the targets used in this study are listed in Table
1.4. Relative expression levels were determined using an iTaq Universal SYBR® Green Supermix
(Biorad, Hercules, CA) in 10µl final volumes and run on a ViiA7 Real-Time PCR machine
(Thermo Fisher Scientific, Waltham, MA). The following qPCR program was used based on the
91
manufacturer’s recommendation: 95°C for 25s and then 40 cycles of 95°C for 2s and 60°C for 25s.
The specificity of the reaction was verified by melting curve analyses.
The expression levels of gene targets were normalized using GAPDH as an internal control. The
Ct was automatically determined by the instrument software and the relative quantification of each
target was performed using the comparative Ct method and was expressed in 2
-∆∆Ct
.
Table 1.4 The sequences of primers used in this study.
Target Forward Sequence (5’-3’) Reverse Sequence (5’-3’)
GAPDH TGAAGCAGGCA TCTGAGGG CGAAGGTGGAAGAGTGGGAG
Rex1 TCACTGTGCTGCCTCCAAGT GGGCACTGA TCCGCAAAC
Oct4 GAAGCAGAAGAGGA TCACCTTG TTCTTAAGGCTGAGCTGCAAG
Nanog CGGCTCACTTCCTTCTGACT GGCGAGGAGAGGCAGC
Egr1 CCACAACAACAGGGAGACCT ACTGAGTGGCGAAGGCTTTA
Sox1 CTCCTCGGCTGAATTCTTTG TGTAATCCGGGTGTTCCTTC
Brachyury CCGGTGCTGAAGGTAAATGT CCTCCATTGAGCTTGTTGGT
Mixl1 TTGAATTGAACCCTGTTGTCCC GAAACCCGTTCTCCCATCCACC
FoxA2 CCTCAAGGGAGCAGTCTCAC TTTCTCCTGGTCCGGTACAC
Sox17 AGCCATTTCCTCCGTGGTGT AACACTGCTTCTGGCCCTCAG
Statistical Analysis
Images were taken as results of a representative experiment. Data are based on at least two
independent experiments. Statistical differences were analyzed using two-tailed Student’s t-tests
for comparisons of the two groups (e.g. CHIR vs. PD + CHIR, Vx11e + CHIR vs. CHIR,
92
serum/bFGF vs. non-treatment, (serum + PD) / (bFGF + PD) vs. non-treatment, or (serum + Vx11e)
/ (bFGF + Vx11e) vs. non-treatment) in each experiment. All data are shown as mean ± SEM. Data
with p<0.05 was considered statistically significant. *: p<0.05, ***: p<0.001.
93
Bibliography
Aceves-Luquero, C.I., Agarwal, A., Callejas-Valera, J.L., Arias-Gonzalez, L., Esparis-Ogando,
A., del Peso Ovalle, L., Bellon-Echeverria, I., de la Cruz-Morcillo, M.A., Galan Moya, E.M.,
Moreno Gimeno, I., et al. (2009). ERK2, but not ERK1, mediates acquired and "de novo"
resistance to imatinib mesylate: implication for CML therapy. PLoS One 4, e6124.
Alaimo, P.J., Shogren-Knaak, M.A., and Shokat, K.M. (2001). Chemical genetic approaches for
the elucidation of signaling pathways. Curr Opin Chem Biol 5, 360-367.
Bain, J., Plater, L., Elliott, M., Shpiro, N., Hastie, C.J., McLauchlan, H., Klevernic, I., Arthur, J.S.,
Alessi, D.R., and Cohen, P. (2007). The selectivity of protein kinase inhibitors: a further update.
Biochem J 408, 297-315.
Baker, C.L., and Pera, M.F. (2018). Capturing Totipotent Stem Cells. Cell Stem Cell 22, 25-34.
Barsyte-Lovejoy, D., Galanis, A., and Sharrocks, A.D. (2002). Specificity determinants in MAPK
signaling to transcription factors. J Biol Chem 277, 9896-9903.
Bessonnard, S., De Mot, L., Gonze, D., Barriol, M., Dennis, C., Goldbeter, A., Dupont, G., and
Chazaud, C. (2014). Gata6, Nanog and Erk signaling control cell fate in the inner cell mass through
a tristable regulatory network. Development 141, 3637-3648.
Bi, L., Okabe, I., Bernard, D.J., and Nussbaum, R.L. (2002). Early embryonic lethality in mice
deficient in the p110beta catalytic subunit of PI 3-kinase. Mamm Genome 13, 169-172.
Bishop, A.C., Ubersax, J.A., Petsch, D.T., Matheos, D.P., Gray, N.S., Blethrow, J., Shimizu, E.,
Tsien, J.Z., Schultz, P.G., Rose, M.D., et al. (2000). A chemical switch for inhibitor-sensitive
alleles of any protein kinase. Nature 407, 395-401.
Bost, F., Aouadi, M., Caron, L., Even, P., Belmonte, N., Prot, M., Dani, C., Hofman, P., Pages,
G., Pouyssegur, J., et al. (2005). The extracellular signal-regulated kinase isoform ERK1 is
specifically required for in vitro and in vivo adipogenesis. Diabetes 54, 402-411.
Brickman, J.M., and Serup, P. (2017). Properties of embryoid bodies. Wiley Interdiscip Rev Dev
Biol 6.
Brook, F.A., and Gardner, R.L. (1997). The origin and efficient derivation of embryonic stem cells
in the mouse. Proc Natl Acad Sci U S A 94, 5709-5712.
Buehr, M., Meek, S., Blair, K., Yang, J., Ure, J., Silva, J., McLay, R., Hall, J., Ying, Q.L., and
Smith, A. (2008). Capture of authentic embryonic stem cells from rat blastocysts. Cell 135, 1287-
1298.
Buehr, M., and Smith, A. (2003). Genesis of embryonic stem cells. Philos Trans R Soc Lond B
Biol Sci 358, 1397-1402; discussion 1402.
94
Burdon, T., Stracey, C., Chambers, I., Nichols, J., and Smith, A. (1999). Suppression of SHP-2
and ERK signalling promotes self-renewal of mouse embryonic stem cells. Dev Biol 210, 30-43.
Busca, R., Christen, R., Lovern, M., Clifford, A.M., Yue, J.X., Goss, G.G., Pouyssegur, J., and
Lenormand, P. (2015). ERK1 and ERK2 present functional redundancy in tetrapods despite higher
evolution rate of ERK1. BMC Evol Biol 15, 179.
Busca, R., Pouyssegur, J., and Lenormand, P. (2016). ERK1 and ERK2 Map Kinases: Specific
Roles or Functional Redundancy? Front Cell Dev Biol 4, 53.
Carlson, S.M., Chouinard, C.R., Labadorf, A., Lam, C.J., Schmelzle, K., Fraenkel, E., and White,
F.M. (2011). Large-scale discovery of ERK2 substrates identifies ERK-mediated transcriptional
regulation by ETV3. Sci Signal 4, rs11.
Catalanotti, F., Reyes, G., Jesenberger, V., Galabova-Kovacs, G., de Matos Simoes, R., Carugo,
O., and Baccarini, M. (2009). A Mek1-Mek2 heterodimer determines the strength and duration of
the Erk signal. Nat Struct Mol Biol 16, 294-303.
Chaikuad, A., Tacconi, E.M., Zimmer, J., Liang, Y., Gray, N.S., Tarsounas, M., and Knapp, S.
(2014). A unique inhibitor binding site in ERK1/2 is associated with slow binding kinetics. Nat
Chem Biol 10, 853-860.
Chan, Y.S., Goke, J., Ng, J.H., Lu, X., Gonzales, K.A., Tan, C.P., Tng, W.Q., Hong, Z.Z., Lim,
Y.S., and Ng, H.H. (2013). Induction of a human pluripotent state with distinct regulatory circuitry
that resembles preimplantation epiblast. Cell Stem Cell 13, 663-675.
Chazaud, C., Yamanaka, Y., Pawson, T., and Rossant, J. (2006). Early lineage segregation between
epiblast and primitive endoderm in mouse blastocysts through the Grb2-MAPK pathway. Dev Cell
10, 615-624.
Chen, H., Guo, R., Zhang, Q., Guo, H., Yang, M., Wu, Z., Gao, S., Liu, L., and Chen, L. (2015).
Erk signaling is indispensable for genomic stability and self-renewal of mouse embryonic stem
cells. Proc Natl Acad Sci U S A 112, E5936-5943.
Chen, X., Wang, R., Liu, X., Wu, Y., Zhou, T., Yang, Y., Perez, A., Chen, Y.C., Hu, L.,
Chadarevian, J.P., et al. (2017). A Chemical-Genetic Approach Reveals the Distinct Roles of
GSK3alpha and GSK3beta in Regulating Embryonic Stem Cell Fate. Dev Cell 43, 563-576 e564.
Chen, X., Ye, H., Kuruvilla, R., Ramanan, N., Scangos, K.W., Zhang, C., Johnson, N.M., England,
P.M., Shokat, K.M., and Ginty, D.D. (2005). A chemical-genetic approach to studying
neurotrophin signaling. Neuron 46, 13-21.
Cho, H.N., Lee, Y.J., Cho, C.K., Lee, S.J., and Lee, Y.S. (2002). Downregulation of ERK2 is
essential for the inhibition of radiation-induced cell death in HSP25 overexpressed L929 cells. Cell
Death Differ 9, 448-456.
Corey, D.R., and Abrams, J.M. (2001). Morpholino antisense oligonucleotides: tools for
investigating vertebrate development. Genome Biol 2, REVIEWS1015.
95
Dani, C., Chambers, I., Johnstone, S., Robertson, M., Ebrahimi, B., Saito, M., Taga, T., Li, M.,
Burdon, T., Nichols, J., et al. (1998). Paracrine induction of stem cell renewal by LIF-deficient
cells: a new ES cell regulatory pathway. Dev Biol 203, 149-162.
Desideri, E., Cavallo, A.L., and Baccarini, M. (2015). Alike but Different: RAF Paralogs and Their
Signaling Outputs. Cell 161, 967-970.
Dhillon, A.S., Hagan, S., Rath, O., and Kolch, W. (2007). MAP kinase signalling pathways in
cancer. Oncogene 26, 3279-3290.
Doble, B.W., and Woodgett, J.R. (2003). GSK-3: tricks of the trade for a multi-tasking kinase. J
Cell Sci 116, 1175-1186.
Doetschman, T.C., Eistetter, H., Katz, M., Schmidt, W., and Kemler, R. (1985). The in vitro
development of blastocyst-derived embryonic stem cell lines: formation of visceral yolk sac, blood
islands and myocardium. J Embryol Exp Morphol 87, 27-45.
Duggal, G., Warrier, S., Ghimire, S., Broekaert, D., Van der Jeught, M., Lierman, S., Deroo, T.,
Peelman, L., Van Soom, A., Cornelissen, R., et al. (2015). Alternative Routes to Induce Naive
Pluripotency in Human Embryonic Stem Cells. Stem Cells 33, 2686-2698.
Emrick, M.A., Lee, T., Starkey, P.J., Mumby, M.C., Resing, K.A., and Ahn, N.G. (2006). The
gatekeeper residue controls autoactivation of ERK2 via a pathway of intramolecular connectivity.
Proc Natl Acad Sci U S A 103, 18101-18106.
Endo, S., Satoh, Y., Shah, K., and Takishima, K. (2006). A single amino-acid change in ERK1/2
makes the enzyme susceptible to PP1 derivatives. Biochem Biophys Res Commun 341, 261-265.
Evans, M.J., and Kaufman, M.H. (1981). Establishment in culture of pluripotential cells from
mouse embryos. Nature 292, 154-156.
Fischer, A.M., Katayama, C.D., Pages, G., Pouyssegur, J., and Hedrick, S.M. (2005). The role of
erk1 and erk2 in multiple stages of T cell development. Immunity 23, 431-443.
Fox, I.J., Daley, G.Q., Goldman, S.A., Huard, J., Kamp, T.J., and Trucco, M. (2014). Stem cell
therapy. Use of differentiated pluripotent stem cells as replacement therapy for treating disease.
Science 345, 1247391.
Fremin, C., Bessard, A., Ezan, F., Gailhouste, L., Regeard, M., Le Seyec, J., Gilot, D., Pages, G.,
Pouyssegur, J., Langouet, S., et al. (2009). Multiple division cycles and long-term survival of
hepatocytes are distinctly regulated by extracellular signal-regulated kinases ERK1 and ERK2.
Hepatology 49, 930-939.
Fremin, C., Ezan, F., Guegan, J.P., Gailhouste, L., Trotard, M., Le Seyec, J., Rageul, J., Theret,
N., Langouet, S., and Baffet, G. (2012). The complexity of ERK1 and ERK2 MAPKs in multiple
hepatocyte fate responses. J Cell Physiol 227, 59-69.
96
Fremin, C., Saba-El-Leil, M.K., Levesque, K., Ang, S.L., and Meloche, S. (2015). Functional
Redundancy of ERK1 and ERK2 MAP Kinases during Development. Cell Rep 12, 913-921.
Gafni, O., Weinberger, L., Mansour, A.A., Manor, Y.S., Chomsky, E., Ben-Yosef, D., Kalma, Y.,
Viukov, S., Maza, I., Zviran, A., et al. (2013). Derivation of novel human ground state naive
pluripotent stem cells. Nature 504, 282-286.
Garske, A.L., Peters, U., Cortesi, A.T., Perez, J.L., and Shokat, K.M. (2011). Chemical genetic
strategy for targeting protein kinases based on covalent complementarity. Proc Natl Acad Sci U S
A 108, 15046-15052.
Goetz, E.M., Ghandi, M., Treacy, D.J., Wagle, N., and Garraway, L.A. (2014). ERK mutations
confer resistance to mitogen-activated protein kinase pathway inhibitors. Cancer Res 74, 7079-
7089.
Goplen, N., Karim, Z., Guo, L., Zhuang, Y., Huang, H., Gorska, M.M., Gelfand, E., Pages, G.,
Pouyssegur, J., and Alam, R. (2012). ERK1 is important for Th2 differentiation and development
of experimental asthma. FASEB J 26, 1934-1945.
Greber, B., Wu, G., Bernemann, C., Joo, J.Y., Han, D.W., Ko, K., Tapia, N., Sabour, D.,
Sterneckert, J., Tesar, P., et al. (2010). Conserved and divergent roles of FGF signaling in mouse
epiblast stem cells and human embryonic stem cells. Cell Stem Cell 6, 215-226.
Guegan, J.P., Ezan, F., Theret, N., Langouet, S., and Baffet, G. (2013). MAPK signaling in
cisplatin-induced death: predominant role of ERK1 over ERK2 in human hepatocellular carcinoma
cells. Carcinogenesis 34, 38-47.
Guihard, S., Clay, D., Cocault, L., Saulnier, N., Opolon, P., Souyri, M., Pages, G., Pouyssegur, J.,
Porteu, F., and Gaudry, M. (2010). The MAPK ERK1 is a negative regulator of the adult steady-
state splenic erythropoiesis. Blood 115, 3686-3694.
Guo, W., Hao, B., Wang, Q., Lu, Y., and Yue, J. (2013). Requirement of B-Raf, C-Raf, and A-Raf
for the growth and survival of mouse embryonic stem cells. Exp Cell Res 319, 2801-2811.
Gusenbauer, S., Zanucco, E., Knyazev, P., and Ullrich, A. (2015). Erk2 but not Erk1 regulates
crosstalk between Met and EGFR in squamous cell carcinoma cell lines. Mol Cancer 14, 54.
Hamazaki, T., Kehoe, S.M., Nakano, T., and Terada, N. (2006). The Grb2/Mek pathway represses
Nanog in murine embryonic stem cells. Mol Cell Biol 26, 7539-7549.
Hamilton, W.B., and Brickman, J.M. (2014). Erk signaling suppresses embryonic stem cell self-
renewal to specify endoderm. Cell Rep 9, 2056-2070.
Hamilton, W.B., Kaji, K., and Kunath, T. (2013). ERK2 suppresses self-renewal capacity of
embryonic stem cells, but is not required for multi-lineage commitment. PLoS One 8, e60907.
97
Hatano, N., Mori, Y., Oh-hora, M., Kosugi, A., Fujikawa, T., Nakai, N., Niwa, H., Miyazaki, J.,
Hamaoka, T., and Ogata, M. (2003). Essential role for ERK2 mitogen-activated protein kinase in
placental development. Genes Cells 8, 847-856.
Hollands, P. (1988). Differentiation of embryonic haemopoietic stem cells from mouse blastocysts
grown in vitro. Development 102, 135-141.
Hong, S.K., Yoon, S., Moelling, C., Arthan, D., and Park, J.I. (2009). Noncatalytic function of
ERK1/2 can promote Raf/MEK/ERK-mediated growth arrest signaling. J Biol Chem 284, 33006-
33018.
Huh, S., Song, H.R., Jeong, G.R., Jang, H., Seo, N.H., Lee, J.H., Yi, J.Y., Lee, B., Choi, H.W.,
Do, J.T., et al. (2018). Suppression of the ERK-SRF axis facilitates somatic cell reprogramming.
Exp Mol Med 50, e448.
Kang, M., Piliszek, A., Artus, J., and Hadjantonakis, A.K. (2013). FGF4 is required for lineage
restriction and salt-and-pepper distribution of primitive endoderm factors but not their initial
expression in the mouse. Development 140, 267-279.
Kauppila, J.H.K., Baines, H.L., Bratic, A., Simard, M.L., Freyer, C., Mourier, A., Stamp, C.,
Filograna, R., Larsson, N.G., Greaves, L.C., et al. (2016). A Phenotype-Driven Approach to
Generate Mouse Models with Pathogenic mtDNA Mutations Causing Mitochondrial Disease. Cell
Rep 16, 2980-2990.
Kawasumi, M., and Nghiem, P. (2007). Chemical genetics: elucidating biological systems with
small-molecule compounds. J Invest Dermatol 127, 1577-1584.
Kelly, K.F., Ng, D.Y., Jayakumaran, G., Wood, G.A., Koide, H., and Doble, B.W. (2011). beta-
catenin enhances Oct-4 activity and reinforces pluripotency through a TCF-independent
mechanism. Cell Stem Cell 8, 214-227.
Kim, E.K., and Choi, E.J. (2010). Pathological roles of MAPK signaling pathways in human
diseases. Biochim Biophys Acta 1802, 396-405.
Kim, M.O., Kim, S.H., Cho, Y.Y., Nadas, J., Jeong, C.H., Yao, K., Kim, D.J., Yu, D.H., Keum,
Y.S., Lee, K.Y., et al. (2012). ERK1 and ERK2 regulate embryonic stem cell self-renewal through
phosphorylation of Klf4. Nat Struct Mol Biol 19, 283-290.
Kim, S.H., Kim, M.O., Cho, Y.Y., Yao, K., Kim, D.J., Jeong, C.H., Yu, D.H., Bae, K.B., Cho,
E.J., Jung, S.K., et al. (2014). ERK1 phosphorylates Nanog to regulate protein stability and stem
cell self-renewal. Stem Cell Res 13, 1-11.
Krens, S.F., Corredor-Adamez, M., He, S., Snaar-Jagalska, B.E., and Spaink, H.P. (2008a). ERK1
and ERK2 MAPK are key regulators of distinct gene sets in zebrafish embryogenesis. BMC
Genomics 9, 196.
98
Krens, S.F., He, S., Lamers, G.E., Meijer, A.H., Bakkers, J., Schmidt, T., Spaink, H.P., and Snaar-
Jagalska, B.E. (2008b). Distinct functions for ERK1 and ERK2 in cell migration processes during
zebrafish gastrulation. Dev Biol 319, 370-383.
Kunath, T., Saba-El-Leil, M.K., Almousailleakh, M., Wray, J., Meloche, S., and Smith, A. (2007).
FGF stimulation of the Erk1/2 signalling cascade triggers transition of pluripotent embryonic stem
cells from self-renewal to lineage commitment. Development 134, 2895-2902.
Kung, A., Schimpl, M., Ekanayake, A., Chen, Y.C., Overman, R., and Zhang, C. (2017). A
Chemical-Genetic Approach to Generate Selective Covalent Inhibitors of Protein Kinases. ACS
Chem Biol 12, 1499-1503.
Lancaster, M.A., and Knoblich, J.A. (2014). Organogenesis in a dish: modeling development and
disease using organoid technologies. Science 345, 1247125.
Lee, S.Y., Coffey, F., Fahl, S.P., Peri, S., Rhodes, M., Cai, K.Q., Carleton, M., Hedrick, S.M.,
Fehling, H.J., Zuniga-Pflucker, J.C., et al. (2014). Noncanonical mode of ERK action controls
alternative alphabeta and gammadelta T cell lineage fates. Immunity 41, 934-946.
Lefloch, R., Pouyssegur, J., and Lenormand, P. (2008). Single and combined silencing of ERK1
and ERK2 reveals their positive contribution to growth signaling depending on their expression
levels. Mol Cell Biol 28, 511-527.
Li, J., and Johnson, S.E. (2006). ERK2 is required for efficient terminal differentiation of skeletal
myoblasts. Biochem Biophys Res Commun 345, 1425-1433.
Li, P., Tong, C., Mehrian-Shai, R., Jia, L., Wu, N., Yan, Y., Maxson, R.E., Schulze, E.N., Song,
H., Hsieh, C.L., et al. (2008). Germline competent embryonic stem cells derived from rat
blastocysts. Cell 135, 1299-1310.
Li, Z., Fei, T., Zhang, J., Zhu, G., Wang, L., Lu, D., Chi, X., Teng, Y., Hou, N., Yang, X., et al.
(2012). BMP4 Signaling Acts via dual-specificity phosphatase 9 to control ERK activity in mouse
embryonic stem cells. Cell Stem Cell 10, 171-182.
Lips, D.J., Bueno, O.F., Wilkins, B.J., Purcell, N.H., Kaiser, R.A., Lorenz, J.N., Voisin, L., Saba-
El-Leil, M.K., Meloche, S., Pouyssegur, J., et al. (2004). MEK1-ERK2 signaling pathway protects
myocardium from ischemic injury in vivo. Circulation 109, 1938-1941.
Marchi, M., D'Antoni, A., Formentini, I., Parra, R., Brambilla, R., Ratto, G.M., and Costa, M.
(2008). The N-terminal domain of ERK1 accounts for the functional differences with ERK2. PLoS
One 3, e3873.
Marchi, M., Pancrazi, L., Maffei, M., Ratto, G.M., and Costa, M. (2010). ERK1 nucleocytoplasmic
shuttling rate depends on specific N-terminal aminoacids. Biochem Biophys Res Commun 398,
166-172.
99
Martello, G., Sugimoto, T., Diamanti, E., Joshi, A., Hannah, R., Ohtsuka, S., Gottgens, B., Niwa,
H., and Smith, A. (2012). Esrrb is a pivotal target of the Gsk3/Tcf3 axis regulating embryonic stem
cell self-renewal. Cell Stem Cell 11, 491-504.
Martin, G.R. (1981). Isolation of a pluripotent cell line from early mouse embryos cultured in
medium conditioned by teratocarcinoma stem cells. Proc Natl Acad Sci U S A 78, 7634-7638.
Mazzucchelli, C., Vantaggiato, C., Ciamei, A., Fasano, S., Pakhotin, P., Krezel, W., Welzl, H.,
Wolfer, D.P., Pages, G., Valverde, O., et al. (2002). Knockout of ERK1 MAP kinase enhances
synaptic plasticity in the striatum and facilitates striatal-mediated learning and memory. Neuron
34, 807-820.
McManus, E.J., Sakamoto, K., Armit, L.J., Ronaldson, L., Shpiro, N., Marquez, R., and Alessi,
D.R. (2005). Role that phosphorylation of GSK3 plays in insulin and Wnt signalling defined by
knockin analysis. EMBO J 24, 1571-1583.
Mitsui, K., Tokuzawa, Y., Itoh, H., Segawa, K., Murakami, M., Takahashi, K., Maruyama, M.,
Maeda, M., and Yamanaka, S. (2003). The homeoprotein Nanog is required for maintenance of
pluripotency in mouse epiblast and ES cells. Cell 113, 631-642.
Morikawa, M., Koinuma, D., Mizutani, A., Kawasaki, N., Holmborn, K., Sundqvist, A., Tsutsumi,
S., Watabe, T., Aburatani, H., Heldin, C.H., et al. (2016). BMP Sustains Embryonic Stem Cell
Self-Renewal through Distinct Functions of Different Kruppel-like Factors. Stem Cell Reports 6,
64-73.
Morrison, D.K. (2012). MAP kinase pathways. Cold Spring Harb Perspect Biol 4.
Murphy, L.O., and Blenis, J. (2006). MAPK signal specificity: the right place at the right time.
Trends Biochem Sci 31, 268-275.
Murry, C.E., and Keller, G. (2008). Differentiation of embryonic stem cells to clinically relevant
populations: lessons from embryonic development. Cell 132, 661-680.
Niakan, K.K., Schrode, N., Cho, L.T., and Hadjantonakis, A.K. (2013). Derivation of
extraembryonic endoderm stem (XEN) cells from mouse embryos and embryonic stem cells. Nat
Protoc 8, 1028-1041.
Nichols, J., Jones, K., Phillips, J.M., Newland, S.A., Roode, M., Mansfield, W., Smith, A., and
Cooke, A. (2009). Validated germline-competent embryonic stem cell lines from nonobese
diabetic mice. Nat Med 15, 814-818.
Nichols, J., and Smith, A. (2009). Naive and primed pluripotent states. Cell Stem Cell 4, 487-492.
Niwa, H., Burdon, T., Chambers, I., and Smith, A. (1998). Self-renewal of pluripotent embryonic
stem cells is mediated via activation of STAT3. Genes Dev 12, 2048-2060.
Otsubo, Y., Satoh, Y., Kodama, M., Araki, Y., Satomoto, M., Sakamoto, E., Pages, G.,
Pouyssegur, J., Endo, S., and Kazama, T. (2012). Mechanical allodynia but not thermal
100
hyperalgesia is impaired in mice deficient for ERK2 in the central nervous system. Pain 153, 2241-
2252.
Pages, G., Guerin, S., Grall, D., Bonino, F., Smith, A., Anjuere, F., Auberger, P., and Pouyssegur,
J. (1999). Defective thymocyte maturation in p44 MAP kinase (Erk 1) knockout mice. Science
286, 1374-1377.
Palmiter, R.D., and Brinster, R.L. (1986). Germ-line transformation of mice. Annu Rev Genet 20,
465-499.
Papa, F.R., Zhang, C., Shokat, K., and Walter, P. (2003). Bypassing a kinase activity with an ATP-
competitive drug. Science 302, 1533-1537.
Papkoff, J., Chen, R.H., Blenis, J., and Forsman, J. (1994). p42 mitogen-activated protein kinase
and p90 ribosomal S6 kinase are selectively phosphorylated and activated during thrombin-
induced platelet activation and aggregation. Mol Cell Biol 14, 463-472.
Pereira, L., Yi, F., and Merrill, B.J. (2006). Repression of Nanog gene transcription by Tcf3 limits
embryonic stem cell self-renewal. Mol Cell Biol 26, 7479-7491.
Pritchard, A.L., and Hayward, N.K. (2013). Molecular pathways: mitogen-activated protein kinase
pathway mutations and drug resistance. Clin Cancer Res 19, 2301-2309.
Radtke, S., Milanovic, M., Rosse, C., De Rycker, M., Lachmann, S., Hibbert, A., Kermorgant, S.,
and Parker, P.J. (2013). ERK2 but not ERK1 mediates HGF-induced motility in non-small cell
lung carcinoma cell lines. J Cell Sci 126, 2381-2391.
Rauch, J., Volinsky, N., Romano, D., and Kolch, W. (2011). The secret life of kinases: functions
beyond catalysis. Cell Commun Signal 9, 23.
Reinholdt, L.G., Howell, G.R., Czechanski, A.M., Macalinao, D.G., Macnicoll, K.H., Lin, C.S.,
Donahue, L.R., and John, S.W. (2012). Generating embryonic stem cells from the inbred mouse
strain DBA/2J, a model of glaucoma and other complex diseases. PLoS One 7, e50081.
Ring, D.B., Johnson, K.W., Henriksen, E.J., Nuss, J.M., Goff, D., Kinnick, T.R., Ma, S.T., Reeder,
J.W., Samuels, I., Slabiak, T., et al. (2003). Selective glycogen synthase kinase 3 inhibitors
potentiate insulin activation of glucose transport and utilization in vitro and in vivo. Diabetes 52,
588-595.
Robbins, D.J., Zhen, E., Owaki, H., Vanderbilt, C.A., Ebert, D., Geppert, T.D., and Cobb, M.H.
(1993). Regulation and properties of extracellular signal-regulated protein kinases 1 and 2 in vitro.
J Biol Chem 268, 5097-5106.
Rogers, M.B., Hosler, B.A., and Gudas, L.J. (1991). Specific expression of a retinoic acid-
regulated, zinc-finger gene, Rex-1, in preimplantation embryos, trophoblast and spermatocytes.
Development 113, 815-824.
101
Roskoski, R., Jr. (2012). ERK1/2 MAP kinases: structure, function, and regulation. Pharmacol Res
66, 105-143.
Runcie, A.C., Chan, K.H., Zengerle, M., and Ciulli, A. (2016). Chemical genetics approaches for
selective intervention in epigenetics. Curr Opin Chem Biol 33, 186-194.
Saba-El-Leil, M.K., Fremin, C., and Meloche, S. (2016). Redundancy in the World of MAP
Kinases: All for One. Front Cell Dev Biol 4, 67.
Saba-El-Leil, M.K., Vella, F.D., Vernay, B., Voisin, L., Chen, L., Labrecque, N., Ang, S.L., and
Meloche, S. (2003). An essential function of the mitogen-activated protein kinase Erk2 in mouse
trophoblast development. EMBO Rep 4, 964-968.
Saijoh, Y., Fujii, H., Meno, C., Sato, M., Hirota, Y., Nagamatsu, S., Ikeda, M., and Hamada, H.
(1996). Identification of putative downstream genes of Oct-3, a pluripotent cell-specific
transcription factor. Genes Cells 1, 239-252.
Samatar, A.A., and Poulikakos, P.I. (2014). Targeting RAS-ERK signalling in cancer: promises
and challenges. Nat Rev Drug Discov 13, 928-942.
Sarbassov, D.D., Jones, L.G., and Peterson, C.A. (1997). Extracellular signal-regulated kinase-1
and -2 respond differently to mitogenic and differentiative signaling pathways in myoblasts. Mol
Endocrinol 11, 2038-2047.
Shimizu, T., Ueda, J., Ho, J.C., Iwasaki, K., Poellinger, L., Harada, I., and Sawada, Y. (2012).
Dual inhibition of Src and GSK3 maintains mouse embryonic stem cells, whose differentiation is
mechanically regulated by Src signaling. Stem Cells 30, 1394-1404.
Shin, J., Yang, J., Lee, J.C., and Baek, K.H. (2013). Depletion of ERK2 but not ERK1 abrogates
oncogenic Ras-induced senescence. Cell Signal 25, 2540-2547.
Shin, M., Franks, C.E., and Hsu, K.L. (2018). Isoform-selective activity-based profiling of ERK
signaling. Chem Sci 9, 2419-2431.
Shin, S., Dimitri, C.A., Yoon, S.O., Dowdle, W., and Blenis, J. (2010). ERK2 but not ERK1
induces epithelial-to-mesenchymal transformation via DEF motif-dependent signaling events. Mol
Cell 38, 114-127.
Smith, A.G., Heath, J.K., Donaldson, D.D., Wong, G.G., Moreau, J., Stahl, M., and Rogers, D.
(1988). Inhibition of pluripotential embryonic stem cell differentiation by purified polypeptides.
Nature 336, 688-690.
Stewart, C.L., Kaspar, P., Brunet, L.J., Bhatt, H., Gadi, I., Kontgen, F., and Abbondanzo, S.J.
(1992). Blastocyst implantation depends on maternal expression of leukaemia inhibitory factor.
Nature 359, 76-79.
Takahashi, K., and Yamanaka, S. (2006). Induction of pluripotent stem cells from mouse
embryonic and adult fibroblast cultures by defined factors. Cell 126, 663-676.
102
Tang, Z., Dai, S., He, Y., Doty, R.A., Shultz, L.D., Sampson, S.B., and Dai, C. (2015). MEK
guards proteome stability and inhibits tumor-suppressive amyloidogenesis via HSF1. Cell 160,
729-744.
Tee, W.W., Shen, S.S., Oksuz, O., Narendra, V., and Reinberg, D. (2014). Erk1/2 activity
promotes chromatin features and RNAPII phosphorylation at developmental promoters in mouse
ESCs. Cell 156, 678-690.
Theunissen, T.W., Powell, B.E., Wang, H., Mitalipova, M., Faddah, D.A., Reddy, J., Fan, Z.P.,
Maetzel, D., Ganz, K., Shi, L., et al. (2014). Systematic identification of culture conditions for
induction and maintenance of naive human pluripotency. Cell Stem Cell 15, 471-487.
Thomson, J.A., Itskovitz-Eldor, J., Shapiro, S.S., Waknitz, M.A., Swiergiel, J.J., Marshall, V.S.,
and Jones, J.M. (1998). Embryonic stem cell lines derived from human blastocysts. Science 282,
1145-1147.
Valamehr, B., Robinson, M., Abujarour, R., Rezner, B., Vranceanu, F., Le, T., Medcalf, A., Lee,
T.T., Fitch, M., Robbins, D., et al. (2014). Platform for induction and maintenance of transgene-
free hiPSCs resembling ground state pluripotent stem cells. Stem Cell Reports 2, 366-381.
Vantaggiato, C., Formentini, I., Bondanza, A., Bonini, C., Naldini, L., and Brambilla, R. (2006).
ERK1 and ERK2 mitogen-activated protein kinases affect Ras-dependent cell signaling
differentially. J Biol 5, 14.
von Kriegsheim, A., Baiocchi, D., Birtwistle, M., Sumpton, D., Bienvenut, W., Morrice, N.,
Yamada, K., Lamond, A., Kalna, G., Orton, R., et al. (2009). Cell fate decisions are specified by
the dynamic ERK interactome. Nat Cell Biol 11, 1458-1464.
von Thun, A., Birtwistle, M., Kalna, G., Grindlay, J., Strachan, D., Kolch, W., von Kriegsheim,
A., and Norman, J.C. (2012). ERK2 drives tumour cell migration in three-dimensional
microenvironments by suppressing expression of Rab17 and liprin-beta2. J Cell Sci 125, 1465-
1477.
Vrbsky, J., Tereh, T., Kyrylenko, S., Dvorak, P., and Krejci, L. (2015). MEK and TGF-beta
Inhibition Promotes Reprogramming without the Use of Transcription Factor. PLoS One 10,
e0127739.
Ware, C.B., Horowitz, M.C., Renshaw, B.R., Hunt, J.S., Liggitt, D., Koblar, S.A., Gliniak, B.C.,
McKenna, H.J., Papayannopoulou, T., Thoma, B., et al. (1995). Targeted disruption of the low-
affinity leukemia inhibitory factor receptor gene causes placental, skeletal, neural and metabolic
defects and results in perinatal death. Development 121, 1283-1299.
Ware, C.B., Nelson, A.M., Mecham, B., Hesson, J., Zhou, W., Jonlin, E.C., Jimenez-Caliani, A.J.,
Deng, X., Cavanaugh, C., Cook, S., et al. (2014). Derivation of naive human embryonic stem cells.
Proc Natl Acad Sci U S A 111, 4484-4489.
Weinberger, L., Ayyash, M., Novershtern, N., and Hanna, J.H. (2016). Dynamic stem cell states:
naive to primed pluripotency in rodents and humans. Nat Rev Mol Cell Biol 17, 155-169.
103
Williams, C.A., Fernandez-Alonso, R., Wang, J., Toth, R., Gray, N.S., and Findlay, G.M. (2016).
Erk5 Is a Key Regulator of Naive-Primed Transition and Embryonic Stem Cell Identity. Cell Rep
16, 1820-1828.
Williams, R.L., Hilton, D.J., Pease, S., Willson, T.A., Stewart, C.L., Gearing, D.P., Wagner, E.F.,
Metcalf, D., Nicola, N.A., and Gough, N.M. (1988). Myeloid leukaemia inhibitory factor
maintains the developmental potential of embryonic stem cells. Nature 336, 684-687.
Wood, H.B., and Episkopou, V. (1999). Comparative expression of the mouse Sox1, Sox2 and
Sox3 genes from pre-gastrulation to early somite stages. Mech Dev 86, 197-201.
Wray, J., Kalkan, T., Gomez-Lopez, S., Eckardt, D., Cook, A., Kemler, R., and Smith, A. (2011).
Inhibition of glycogen synthase kinase-3 alleviates Tcf3 repression of the pluripotency network
and increases embryonic stem cell resistance to differentiation. Nat Cell Biol 13, 838-845.
Wray, J., Kalkan, T., and Smith, A.G. (2010). The ground state of pluripotency. Biochem Soc
Trans 38, 1027-1032.
Wu, J., Greely, H.T., Jaenisch, R., Nakauchi, H., Rossant, J., and Belmonte, J.C. (2016). Stem
cells and interspecies chimaeras. Nature 540, 51-59.
Wu, J., Platero-Luengo, A., Sakurai, M., Sugawara, A., Gil, M.A., Yamauchi, T., Suzuki, K.,
Bogliotti, Y.S., Cuello, C., Morales Valencia, M., et al. (2017). Interspecies Chimerism with
Mammalian Pluripotent Stem Cells. Cell 168, 473-486 e415.
Yao, Y., Li, W., Wu, J., Germann, U.A., Su, M.S., Kuida, K., and Boucher, D.M. (2003).
Extracellular signal-regulated kinase 2 is necessary for mesoderm differentiation. Proc Natl Acad
Sci U S A 100, 12759-12764.
Ye, S., Li, P., Tong, C., and Ying, Q.L. (2013). Embryonic stem cell self-renewal pathways
converge on the transcription factor Tfcp2l1. EMBO J 32, 2548-2560.
Yeo, J.C., Jiang, J., Tan, Z.Y., Yim, G.R., Ng, J.H., Goke, J., Kraus, P., Liang, H., Gonzales, K.A.,
Chong, H.C., et al. (2014). Klf2 is an essential factor that sustains ground state pluripotency. Cell
Stem Cell 14, 864-872.
Ying, Q.L., Nichols, J., Chambers, I., and Smith, A. (2003a). BMP induction of Id proteins
suppresses differentiation and sustains embryonic stem cell self-renewal in collaboration with
STAT3. Cell 115, 281-292.
Ying, Q.L., and Smith, A. (2017). The Art of Capturing Pluripotency: Creating the Right Culture.
Stem Cell Reports 8, 1457-1464.
Ying, Q.L., Stavridis, M., Griffiths, D., Li, M., and Smith, A. (2003b). Conversion of embryonic
stem cells into neuroectodermal precursors in adherent monoculture. Nat Biotechnol 21, 183-186.
Ying, Q.L., Wray, J., Nichols, J., Batlle-Morera, L., Doble, B., Woodgett, J., Cohen, P., and Smith,
A. (2008). The ground state of embryonic stem cell self-renewal. Nature 453, 519-523.
104
Yoshida, K., Taga, T., Saito, M., Suematsu, S., Kumanogoh, A., Tanaka, T., Fujiwara, H., Hirata,
M., Yamagami, T., Nakahata, T., et al. (1996). Targeted disruption of gp130, a common signal
transducer for the interleukin 6 family of cytokines, leads to myocardial and hematological
disorders. Proc Natl Acad Sci U S A 93, 407-411.
Yusa, K., Zhou, L., Li, M.A., Bradley, A., and Craig, N.L. (2011). A hyperactive piggyBac
transposase for mammalian applications. Proc Natl Acad Sci U S A 108, 1531-1536.
Zhang, C., Kenski, D.M., Paulson, J.L., Bonshtien, A., Sessa, G., Cross, J.V., Templeton, D.J., and
Shokat, K.M. (2005). A second-site suppressor strategy for chemical genetic analysis of diverse
protein kinases. Nat Methods 2, 435-441.
Zhang, C., Lopez, M.S., Dar, A.C., Ladow, E., Finkbeiner, S., Yun, C.H., Eck, M.J., and Shokat,
K.M. (2013). Structure-guided inhibitor design expands the scope of analog-sensitive kinase
technology. ACS Chem Biol 8, 1931-1938.
Zuberi, A., and Lutz, C. (2016). Mouse Models for Drug Discovery. Can New Tools and
Technology Improve Translational Power? ILAR J 57, 178-185.
105
Part II The Effect of Medium Osmolality on Neural Differentiation
of Mouse Embryonic Stem Cell
Abstract
It is well-known that osmotic forces play a role in diverse cellular processes. The effect of change
in osmolality on neural differentiation, however, remains unknown. Here, we investigated the
effect of change in medium osmolality on neural differentiation of mouse embryonic stem cells
(ESCs). We found that mouse ESCs cultured in serum-free medium with low osmolality (~200
mOsm/kg) rarely differentiated into Sox1 positive neural stem cells (NSCs). Increasing the
osmolality of the serum-free medium by adding NaCl significantly increased the efficiency of ESC
differentiation into Sox1 positive NSCs. Increasing osmolality by adding other osmolytes such as
KCl, MgCl
2
, Na
2
SO
4
and glucose to the serum-free medium had a similar effect in promoting
neural differentiation of ESCs. Increasing osmolality activated the MAPK/ERK pathway, and
inhibition of this pathway partially abolish the effect of increasing osmolality in promoting neural
differentiation of ESCs. Activation of the MAPK/ERK pathway by FGF4 promoted neural
differentiation of ESCs culture in serum-free medium with low osmolality. To further confirm the
role of the MAPK/ERK pathway in this process, we generated Erk1/2 double knock-out ESCs with
CRISPR/Cas9 technology. Erk1/2 double knock-out ESCs differentiated into NSCs with a much
lower efficiency compared with wild-type ESCs, indicating the importance of ERK signaling in
the neural induction. Interestingly, increased osmolality could still promote neural differentiation
of Erk1/2 double knock-out ESCs suggesting that osmolality affects neural differentiation of ESCs
through both MAPK/ERK pathway-dependent and -independent mechanisms.
106
Introduction
Patients suffering from neurological diseases and injury can potentially benefit from cell
replacement therapy using neural cells derived from embryonic stem cells (ESCs) (Snyder, 2018).
If the full potential of ESCs in cell-based therapies is to be realized, a better understanding of the
regulation of their differentiation is critical.
1. Neural Induction in vivo
Neural induction is the first step of neural development in the embryo (Ozair et al., 2013; Stern,
2006). Neural progenitor cells first appear during development with the formation of the neural
plate from a portion of the ectoderm during gastrulation (Ozair et al., 2013). The ends of the neural
plate, known as neural folds, then push themselves up and then fuse together, leading to the
formation of the neural tube. The neural tube consists of neuroepithelial cells and extends along
the anterior-posterior (AP) axis. During the same time window, neural crest and epidermal cells
are also formed from the ectoderm. Neural crest cells are able to migrate across the whole embryo
(Nieto, 2001) and differentiate into an astonishing number of cell types in different organs.
Epidermal cells contribute to the skin.
In exploration of the cellular basis of neural induction, Spemann and Mangold found that the
ventral ectoderm of a host frog embryo could be induced to form a complete nervous system when
exposed to a piece of tissue from another frog’s dorsal mesoderm (Lenhoff, 1991). This
mesodermal region capable of inducing neural tissue is termed as the Spemann organizer. A
homolog of the Spemann organizer, called the node, has also been found in chicks, rabbits, mice,
and recently in humans (Beddington, 1994; Martinez Arias and Steventon, 2018; Martyn et al.,
107
2018; Xu, 2006), suggesting that the conservation of mechanisms for neural induction among
different vertebrates.
After decades of work investigating the molecular basis of the Spemann organizer’s ability to
induce formation of neural tissue, a default model of neural induction was proposed (Figure 2.1)
(Hemmati-Brivanlou and Melton, 1997; Stern, 2006). In this model, the secretion of bone
morphogenetic protein (BMP) inhibitors by the organizer releases the brake on ectodermal cells
and allow them differentiate into neural progenitors following their default fate (Stern, 2006). The
default model of neural induction was proposed based on two linked findings in amphibian
embryos. Firstly, cultured naïve ectoderm explanted from blastula-stage embryos gave rise to the
epidermis, whereas the dissociation of uncommitted ectodermal explants into single cells, wherein
cell-cell communication was disrupted, led to the formation of neural tissue (Kuroda et al., 2005).
Figure 2.1 Default model of neural induction.
A zygote, the fertilized egg, divides and forms a blastula. A region of the early embryo generates the epiblast,
which responds to patterning signals as development progresses from left to right. During the blastula stages, these
signals include ones that induce mesodermal derivatives in the posterior regions of the embryo using growth
factors such as FGF, Wnt, and nodal-related families. During the gastrula stages, the ectoderm on the ventral side
is induced to form the epidermis by BMP signaling. However, a region of ectoderm can evade BMP signaling
through inhibitors produced by the organizer and thus form neural tissue. BMP: bone morphogenetic protein; FGF:
fibroblast growth factor. Adapted from (Kintner & Koyano-Nakagawa, 2013).
108
In addition, the overexpression of dominant negative transforming growth factor beta (TGF-b)
type II receptor in the ectodermal cells could induce the formation of neural tissue (Hemmati-
Brivanlou and Melton, 1994). Together, these two results suggest that a TGF-b related signaling
pathway plays a negative role in inducing neural tissues from ectoderm. Subsequent work revealed
that the inhibitory signaling is BMP signaling, and a very small amount of BMP can stop the
specification of neural cells in dispersed ectodermal cells (Kuroda et al., 2005). Furthermore, BMP
inhibitors such as Noggin, Chordin, and Follistatin have been identified in the organizer tissues
(Hemmati-Brivanlou et al., 1994; Lamb et al., 1993; Sasai et al., 1994), and are potent neural
inducers.
However, a major problem associated with the default model is that BMP-signaling independent
pathways also play some roles in the neural induction (Stern, 2005). For instance, neural tissues
can be formed in mouse embryos lacking functional Follistatin, Chordin, Noggin, or both Noggin
and Chordin (Bachiller et al., 2000; Matzuk et al., 1995; McMahon et al., 1998). Furthermore,
mouse embryos carrying a malfunctioning FoxA2 (HNF3b), failed to generate the node and its
derivatives, but still generated neural plates (Klingensmith et al., 1999). This suggests that
generation of neural cells in mouse embryos does not require a functional node or node derivatives.
In addition, a neural plate was formed when the organizer was surgically removed in the chick or
zebrafish (Psychoyos and Stern, 1996; Shih and Fraser, 1996). Therefore, the organizer-dependent
BMP-antagonizing signaling may not be required for neural induction and alternative pathways
operate in embryos to induce neural tissues.
Fibroblast growth factor (FGF) signaling and Wnt signaling have been implicated in neural
induction in chick and frog embryos (Kuriyama and Mayor, 2009; Kuroda et al., 2005; Streit et al.,
2000; Wilson et al., 2000). Active FGF signaling seems to be required for neural induction in the
109
chick, as the expression of neural markers and the induction of an ectopic neural plate by a grafted
organizer are blocked by FGF signaling inhibitors (Streit et al., 2000). In addition, FGF signaling
and Wnt signaling in frog embryos are both necessary for neural specification in addition to BMP
signaling inhibition (Delaune et al., 2005; Marchal et al., 2009). Moreover, inhibition of Wnt
signaling promotes the formation of neural tissue in chick and frog embryos (Heeg-Truesdell and
LaBonne, 2006; Wilson et al., 2001).
2. Neural induction in vitro: ESC neural differentiation
Neural induction can be recapitulated in vitro through ESC differentiation towards neural stem
cells (NSCs), neural progenitors, and mature neural subtypes (neurons and glia cells). Diverse
approaches have been developed to achieve in vitro neural differentiation of ESCs, and all these
methods try to recapitulate, in different ways, the multistep process of neural development that
occurs in the embryo. Based on the strategies used, these approaches of ESC neural differentiation
can be classified into at least three categories: embryoid body (EB)-based differentiation protocols,
stromal cell co-culture protocols, and direct differentiation protocols based on the default model
of neural induction (Figure 2.2).
EB-Based Differentiation Protocols
When cultured in suspension without differentiation inhibitory factors, ESCs form three-
dimensional multicellular aggregates called EBs and differentiate spontaneously (Brickman and
Serup, 2017). EBs recapitulate many aspects of early embryo development and in particular
gastrulation. Accordingly, the derivatives of all three germ layers can be found in EBs, and a
heterogeneous population including neural progenitors can be harvested from EB differentiation.
110
To improve the derivation efficiency of neural cells in EB protocols, several modified protocols
have been developed to enhance neural differentiation and expansion of neural progenitors.
The first EB-mediated neural differentiation protocol described was a 4-/4+ protocol (Bain et al.,
1995), in which ESCs were cultured in the serum containing medium without leukemic inhibitory
factor (LIF) for 4 days followed by the culture in the serum containing medium with the addition
of retinoic acid (RA) for another 4 days. RA is an active derivative of vitamin A and one of the
most important extrinsic inductive signals for neural differentiation (Cai and Grabel, 2007; Ross
Figure 2.2 Protocols for embryonic stem cell (ESC) neural differentiation.
These approaches of ESC neural differentiation can be classified into at least three categories: embryoid body
(EB)-based differentiation, stromal cell co-culture differentiation, and direct differentiation based on the default
model of neural induction. The time frame for completing the neural differentiation from ESCs is roughly
estimated and shown in the leftmost part of figure. FGF2: fibroblast growth factor 2. LIF: leukemia inhibitory
factor.
111
et al., 2000). RA has a concentration-dependent effect on neural differentiation, and also plays a
role in dorso-ventral and rostro-caudal identity acquisition during neural patterning (Okada et al.,
2004). Consequently, many modified versions based on the basic 4-/4+ protocol have been
developed to optimize the generation of neural cells (Fraichard et al., 1995; Kothapalli and Kamm,
2013; Okada et al., 2004; Renoncourt et al., 1998). One drawback of the RA-based EB protocols
is that derived neural progenitors have differentiation defects, as evidenced by impaired axon
elongation out of the dorsal root ganglia when transplanted into chick neural tube, a phenomenon
not seen in neural cells derived from RA-free ESC differentiation protocols (Plachta et al., 2004).
An alternative EB-based strategy is based on the conditioned medium derived from a HepG2
hepatocarcinoma cell line that can promote the homogeneous differentiation of primitive
ectoderm-like cells from mouse ESCs. EBs formed using the conditioned medium do not have an
extraembryonic endoderm layer. It seems that the conditioned medium contains special molecules
that are capable of inducing neuroectoderm differentiation at the expense of mesendoderm
differentiation. Accordingly, cells in these EBs express neural progenitor markers, such as Sox1
and nestin, but do not express any mesodermal or endodermal markers. More importantly, neural
progenitors generated using this protocol do not appear to have a restricted regional identity and
can be further differentiated into neurons or glia, as well as neural crest cells (Rathjen et al., 2002).
A third EB-based neural differentiation protocol includes formation of an EB-intermediate and
selection of neural cells (Guan et al., 2001; Okabe et al., 1996). ESCs are first cultured as
suspension to form EBs, followed by the culture of cells on adhesive substrates in a serum-free
medium supplemented with insulin, transferrin, selenium, and fibronectin. This serum-free
medium is termed as the ITSFn medium. After several days of culture in the ITSFn medium, a
large proportion of cells die. Most of the surviving cells are neural progenitors that express nestin.
112
The addition of fibroblast growth factor 2 (FGF2) can help expand nestin+ neural progenitor cells
(Okabe et al., 1996). Neural progenitors generated with this approach have been used to produce
different types of neurons and glial cells, including dopaminergic neurons, glutamatergic neurons,
and oligodendrocytes.
A common drawback shared by the above EB-based protocols is the high variability of EB size.
EB size is crucial for the differentiation of ESCs into specific cell types, and the mixed sizes of
EBs can contribute to the heterogeneity during EB-based differentiation (Moon et al., 2014; Ng et
al., 2005). To address this issue, low-adherence 96-well plates and microscale technologies (e.g.
the spinner flask and bioreactor) have been developed to induce formation of EBs with uniform
and controlled sizes (Kurosawa, 2007), which may further optimize the available EB
differentiation protocols to induce neural differentiation more efficient in the future.
Stromal Cell Co-Culture Protocols
Stromal cells have been shown to support the survival, proliferation, self-renewal and
differentiation of various stem cells (Muller-Sieburg and Deryugina, 1995). In addition, stromal
feeder cell lines derived from bone marrow have been used for the expansion of undifferentiated
hematopoietic stem cells (HSCs) (Collins and Dorshkind, 1987; Itoh et al., 1989; Okada et al.,
1992). These results highlight the usefulness of stromal cells in cell culture practice.
Most recently, it has been reported that the stromal cell lines originally used for supporting the
growth of HSCs can induce neural differentiation when co-cultured with primate and mouse ESCs
(Barberi et al., 2003; Kawasaki et al., 2000). The pro-neural differentiation ability of various
stromal cells is called stromal cell-derived inducing activity (SDIA). SDIA is shared by multiple
stromal cell lines (including MS5, S17, and PA6) and primary cells obtained from the aorta-gonad-
mesonephros (AGM) region. These cell lines and primary cells always show SDIA when co-
113
cultured with a variety of ESC lines. ESCs co-cultured with stromal-feeders can be induced to
differentiate towards a wide variety of specific neural fates, including multiple subtypes of neurons
and glial cells. In addition, early neural precursors that emerge in co-culture can be expanded with
a combination of N2 medium and FGF2. The co-culture protocol is more efficient in inducing
neural differentiation of ESCs. In terms of the basis of SDIA, the survival of the stromal cells is
not necessary since paraformaldehyde fixation does not abolish, and can even enhance, the SDIA
of stromal cells (Kawasaki et al., 2000). Physical contacts between stromal cells and ESCs are not
required as well, suggesting that the SDIA comes from some soluble inducing factor(s). At present,
the molecular nature of those factors is unclear.
Direct Differentiation Protocols
Two simpler methods to reconstitute neural differentiation in vitro and achieve efficient neural
progenitor production have been developed based on the default model of neural induction
(Smukler et al., 2006; Tropepe et al., 2001; Ying et al., 2003). In these two protocols, ESCs are
cultured, either in suspension or adherently, under chemically defined serum- and feeder-free
conditions. Without BMP signaling from serum, ESCs directly undergo neural specification
through an autocrine induction mechanism, in which FGF signaling is crucial. Blocking FGF
signaling by genetic deletion, neutralizing antibodies, or chemical inhibitors, significantly reduced
the percentage of NSCs derived from ESCs, confirming the early role of FGF signaling in the
neural development in vivo. In the suspension neurosphere-forming protocol, most ESCs die and
a low percent of cells (~0.2%) contribute to the formation of neurospheres (Smukler et al., 2006).
In contrast, the adherent monolayer differentiation protocol has a much higher cell viability, and
around 60% of surviving cells becomes Sox1 positive NSCs on day 5-6 (Ying and Smith, 2003;
Ying et al., 2003). The advantages of this monolayer differentiation protocol include the relatively
114
short time-frame of differentiation, and the ease of isolating a pure NSC population using
Fluorescence-activated cell sorting if a Sox1 reporter ESC line (e.g. 46c ESCs) is used. In addition,
the moderate plating density and the robust proliferation of differentiating ESCs make the
monolayer neural differentiation protocol feasible for the large-scale NSC production. In addition,
NSC lines derived from humans and mice are able to differentiate into neurons, astrocytes, and
oligodendrocytes, and are used for disease treatments (Banda and Grabel, 2016; Daadi et al., 2008).
3. Signaling Pathways Implicated in ESC Neural Differentiation
Neural differentiation of ESCs provides an ideal in vitro model for studying the roles of different
signaling pathways in neurogenesis. In addition, an improving understanding of the overall
signaling pathways underlying neural differentiation of ESCs will help refine the protocols of
neural differentiation from ESCs and advance the application of ESCs in cell therapy.
BMP Signaling
The role of BMP signaling in inhibiting neural differentiation has been well established since
BMP4-addition inhibits the emergence of neural progenitors from ESCs (Finley et al., 1999).
Regarding the molecular circuits underlying the neural differentiation inhibition, BMPs as part of
the transforming growth factor-b (TGFb) superfamily of proteins, can bind to type II receptors in
dimeric form and lead to the propagation of signaling by the canonical Smad-dependent pathway
(Wang et al., 2014). In the canonical pathway, BMPs induce the phosphorylation of R-Smads
(Smad 1/5/8), and their association with co-Smad (Smad4). The complex of R-Smads and co-
Smad is translocated into nucleus and participates in the regulation of gene expression. BMPs’
inhibitory role in neural specification of ESCs is mediated by various Smads. Knocking down
Smad1, a response element of BMP signaling, promotes neural differentiation of ESCs, as
115
evidenced by the increased expression of Sox1 (Fei et al., 2010). This ESC-based study is
consistent with the study of Smad1 in frog embryos in which inhibition of Smad1/2 pathways
induces neural induction more efficiently (Chang and Harland, 2007). The Smad4 knock-down
has a stronger effect than that of Smad1 in the promotion of neural differentiation. This may be
because Smad4 is also the target of Activin/Nodal activated TGFb signaling and the ablation of
Activin/Nodal signaling in ESCs facilitates neural differentiation (Sonntag et al., 2005). In addition,
Smad1/5 mediates the crosstalk between BMP signaling and FGF signaling in the regulation of
neural commitment (Cho et al., 2014; Pera et al., 2003). The phosphorylation dependent retention
of Smad1/5 in the cytoplasm can promote neural differentiation.
FGF Signaling
FGF signaling is crucial for neural differentiation of ESCs since Fgf receptor 1 (Fgfr1) null ESCs
produce much fewer neural spheres and monolayer ESCs expressing dominant negative FGF4
receptors generate a lower percentage of Sox1-GFP positive NSCs in comparison to their wild-
type counterparts (Smukler et al., 2006; Ying et al., 2003). In addition, the addition of FGF
signaling inhibitors impairs neural differentiation. An alternative explanation is that the activation
of FGF signaling specifically promotes the survival and proliferation of NSCs, similar to the role
of FGF2 in the expansion of NSCs (Okabe et al., 1996). Using the monolayer neural differentiation
protocol, Chen et al. revealed that FGF1, FGF2, and FGF4 are able to increase neural
differentiation, which was not mediated by decreased apoptosis or enhanced proliferation in NSCs
(Chen et al., 2010). These results demonstrate that FGF proteins do promote ESC neural
differentiation. In fact, FGF signaling is necessary for the transition of ESCs from self-renewal to
differentiation (Kunath et al., 2007). Moreover, FGF signaling opposes BMP signaling by
hampering the nuclear translocation of pSmad1/5. On one hand, FGF/ERK signaling triggers an
116
influx of Ca
2+
, which activates the CaN phosphatase complex and then dephosphorylates
pSmad1/5 at the C-terminal (Cho et al., 2014). On the other hand, FGF/ERK signaling may
phosphorylate the linker region in pSmad1 and retain Smad1 in the cytoplasm (Pera et al., 2003).
Therefore, FGF signaling directs the differentiation of ESCs into NSCs rather than epidermal or
mesendodermal cells.
Wnt Signaling
The role of Wnt signaling in neural differentiation of ESCs is controversial. Wnt signaling
activation in ESCs can inhibit neural differentiation. Overexpression of Wnt1, the adenomatous
polyposis coli (APC) mutant, or the dominant active b-catenin mutant in ESCs leading to the
accumulation of cytoplasmic b-catenin inhibits the formation of NSCs and contributes to the
maintenance of pluripotency (Atlasi et al., 2013; Aubert et al., 2002; Bakre et al., 2007; Haegele
et al., 2003). Knock-down of Dkk1, a negative regulator of WNT signaling, or the depletion of
GSK3a/b in ESCs impairs neural differentiation (Chen et al., 2017; Verani et al., 2007). Moreover,
the addition of GSK3 inhibitors that reduce b-catenin degradation suppresses the production of
neural progenitors (Aubert et al., 2002; Sato et al., 2004). In contrast, inhibition of Wnt signaling
by either Dkk1 proteins or overexpression of SFRP2 facilitates the differentiation of ESCs into
neural progenitors (Haegele et al., 2003; Kong and Zhang, 2009; Watanabe et al., 2005). Contrary
to the above findings, Otero et al. found that activation of Wnt pathway by the expression of Wnt3a
or a dominant negative E-cadherin increases cytoplasmic b-catenin and promotes neural
commitment (Otero et al., 2004). The discrepancy regarding the role of Wnt signaling in neural
differentiation may have resulted from different differentiation conditions, variation in ESC status
(e.g. seeding density), and b-catenin dose-dependent effects. In addition, Wnt signaling can
activate both b-catenin -dependent and -independent downstream cascades (Wray et al., 2011),
117
which may have opposite roles in neural induction. Consequently, genetic deletion of b-catenin in
ESCs, compared with chemical inhibition of Wnt signaling in ESCs, can yield different phenotypes
in terms of neural differentiation.
The c-Jun N-terminal Kinase Pathway
The c-Jun N-terminal kinase (JNK) is one of the classic MAPK family members and is expressed
in mammals in three isoforms. The role of the JNK pathway in stress responses and the
development has been established via gene disruption in mice. In particular, JNK1/JNK2 double
knockout mice have defects in neural tubes and hindbrains, suggesting the role of JNK1 and JNK2
in the development of nervous systems (Kuan et al., 1999; Sabapathy et al., 1999). The deletion of
JNK1 but not JNK2 or JNK3 in ESCs significantly reduces the outgrowth of neurites using the
RA-based EB differentiation protocol (Amura et al., 2005). Neurite outgrowth is also reduced
when JNK/stress-activated protein kinase associated protein 1 (JSAP1), a JNK3 interacting protein,
is knocked out (Xu et al., 2003). Shan et al. reported that JNK participated in RA-based neural
differentiation by phosphorylating cyclic AMP response element-binding protein (CREB) (Shan
et al., 2008). Following the activation of JNK by RA, levels of pCREB were increased and neural
differentiation was enhanced based on the expression of nestin. Furthermore, activation of JNK
signaling facilitates neural differentiation in the monolayer neural differentiation (Chen et al.,
2010). The addition of a JNK inhibitor (SP600125) in the medium or the expression of dominant
negative JNK mutants in ESCs impairs the production of NSCs. Taken together, activation of the
JNK pathway contributes to neural differentiation of ESCs.
On top of the aforementioned signaling pathways, Notch signaling, Hedgehog signaling, and
calcium signaling also play some roles in neural differentiation (Chuang et al., 2015). Of note, the
RA, a strong neural lineage inducing agent in ESCs, promotes the neural induction via multiple
118
pathways, including Smad1-, Wnt-, ERK-, and JNK- dependent signal transduction (Chuang et al.,
2015; Shan et al., 2008; Sheng et al., 2010). This indicates us that manipulation of multiple
pathways may maximize the potential of neural differentiation from ESCs.
4. Implication of Physical Factors in Neural Differentiation
In addition to growth factors, matrix and physical parameters are also crucial for stem cell fate
control (Discher et al., 2009). The roles of physical forces in embryogenesis, such as blastocyst
compaction and gastrulation, have been well-established (Brunet et al., 2013; Ingber, 2006).
Moreover, we use different matrices with distinct physical properties to coat tissue culture plates
in different protocols for the expansion and differentiation of ESCs. This fact underscores the
significance of the physical environment in cell fate control (Discher et al., 2009). Accordingly,
efforts have been invested to uncover the effects of physical factors in ESC neural differentiation.
The stiffness, or the rigidity, of coating matrices can determine cell fates. Works from multiple
labs have revealed that neurogenic matrices with a stiffness range similar to that of tissues found
in the brain dramatically promote the production of neural progenitor cells from mouse and human
ESCs (Ali et al., 2015; Keung et al., 2012; Macri-Pellizzeri et al., 2015; Sun et al., 2014; Zoldan
et al., 2011). The soft extracellular matrix (ECM) can also promote neurogenic differentiation from
mesenchymal stem cells (MSCs), multipotent stem cells that are able to undergo neurogenesis (Du
et al., 2011; Engler et al., 2006). The underlying mechanism is attributed to the regulation of YAP
protein distribution and the phosphorylation of Smad by ECM rigidity. Soft ECM promotes the
retention of YAP/TAZ and Smads proteins in the cytoplasm, thereby enhancing neural
differentiation in an anti-BMP signaling dependent manner. In addition, a soft environment may
suppress the BMP/Smad pathway through the integrin-regulated BMP receptor endocytosis.
Reducing oxygen concentration in the incubator can also greatly enhances the neural
119
differentiation (Mondragon-Teran et al., 2009). The temperature, pH, and the osmolality of culture
medium can likewise influence neural induction by affecting the cytoplasmic-nuclear distribution
of Smad5, a mediator of BMP signaling (Fang et al., 2017; Kim et al., 2017).
5. Rationale of This Study
In 2003, Dr. Ying and colleagues developed the monolayer neural differentiation protocol for the
derivation of NSCs from ESCs. In this neural differentiation protocol, ESCs are cultured in N2B27
medium, a 1:1 mixture of DMEM/F12 medium and Neurobasal medium with the addition of N-2
and B-27
TM
supplements. N2B27 medium could induce mouse Sox1-GFP knock-in (46c) ESCs to
differentiate and more than 50% of cells were Sox1-GFP positive NSCs on day 6 (Ying et al.,
2003). If DMEM/F12 medium supplemented with N-2 and B-27
TM
supplements (termed DFNB
medium) was used in the monolayer neural differentiation protocol, the percentage of Sox1-GFP
positive (Sox1-GFP
+
) NSCs increased to 70-80% on day 6 (Ying and Smith, 2003); in contrast, if
Neurobasal medium with N-2 and B-27
TM
supplements (termed NB medium) was used, there were
much fewer Sox1-GFP
+
NSCs on day 6 (unpublished data from Dr. Ying). The great variation in
the percentages of Sox1-GFP
+
NSCs among the ESCs cultured in the three different media (NB,
N2B27, DFNB) were likely due to differences among the three different neural differentiation
media. The working concentration of N-2 and B-27
TM
supplements, the critical components for
neural differentiation of ESCs, are the same in the above three neural differentiation media. As
such, the differences in the basal media for preparing NB medium, N2B27 medium, and DFNB
medium lead to the variation in the percentage of Sox1-GFP
+
NSCs. These two basal media are
Neurobasal medium and DMEM/F12 medium. Neurobasal medium is a modified version of
DMEM/F12 medium, with reduction in the concentrations of several amino acids and removal of
ferrous sulfate (Brewer et al., 1993). The major difference between these two basal media is the
120
total amount of solutes in the solution, which defines the osmolality of the medium. The role of
osmolality in neural lineage differentiation was established in as early as 1975 (Fisher et al., 1975),
and its role in the differentiation of other cell types has also been reported (Caron et al., 2013).
Based on the above clues, we hypothesized that increased medium osmolality promotes the
differentiation of ESCs into the neural lineage under the monolayer differentiation protocol.
Here, we investigated the effects of increased osmolality on neural differentiation of mouse 46c
ESCs using the monolayer neural differentiation protocol. By adding NaCl and other neutral
(pH=7.0) osmolytes into NB medium, we elevated the medium osmolality and found that the
elevation in medium osmolality enhanced the production of Sox1-GFP positive NSCs. MAPK
pathways play a key role in response to a variety of environmental stresses, including osmolality.
Through chemical screening with inhibitors of MAPK pathways, we revealed that MAPK/ERK
signaling was involved in promoting neural differentiation of ESCs mediated by increased
osmolality. Additionally, we cultured Erk1/2 DKO ESCs in Neurobasal medium supplemented
with NaCl or glucose for neural differentiation. In the media with elevated osmolality, we still
could see increased numbers of Sox1-GFP
+
NSCs and enhanced neuronal lineage differentiation.
Based on the findings, we conclude that increased medium osmolality promotes neural
differentiation of mouse ESCs through MAPK/ERK signaling -dependent and -independent
mechanisms.
121
Experimental Results
1. Addition of NaCl Promotes Neural Differentiation of ESCs Cultured in
Neurobasal Medium with N-2 and B-27
TM
Supplements
To investigate the effect of medium osmolality on neural differentiation of ESCs, we needed a
medium with low osmolality as a basal medium, which was easily changeable by adding various
osmolytes (e.g. salts or organic molecules). To this end, we first measured the osmolality of
Neurobasal medium and DMEM/F12 medium using an osmometer: the osmolalities of Neurobasal
and DMEM/F12 media are 207 ± 0.6 mOsm/kg and 288 ± 1.1 mOsm/kg, respectively. DMEM/F12
medium is optimal for mammalian cell culture since its osmolality is similar to that of blood in
mice (Waymouth, 1970). In consideration that the osmolality of Neurobasal medium was much
lower, we chose Neurobasal medium as basal medium and tried to increase its osmolality to close
to the osmolality of blood by adding salts. To adjust the medium osmolality, we added different
amounts of NaCl into Neurobasal medium and the osmolalities of the prepared media were
measured. The osmolalities of Neurobasal +10 mM NaCl, Neurobasal + 30mM NaCl, and
Neurobasal + 50mM NaCl were 227 ± 1.5 mOsm/kg, 263 ± 1.2 mOsm/kg, and 300 ± 1.5 mOsm/kg,
respectively. These data suggested that NaCl could linearly increase the osmolality of Neurobasal
medium. Neurobasal media supplemented with N-2 and B-27
TM
supplements were termed as NB
medium for neural differentiation. The addition of these two supplements into the basal medium
increased the medium osmolality by another 2-5 mOsm/kg. I then used NaCl to adjust the NB
medium with different osmolalities.
I first investigated the effects of these NaCl-adjusted NB media on ESC neural differentiation. The
mouse ESC line I used was the 46c ESC line, which carried a Sox1-GFP reporter. Sox1 is the
122
earliest neural lineage marker in mouse and Sox1 positive cells derived from mouse ESCs are
NSCs (Ying et al., 2003). As such, I could assess the neural differentiation of the 46c ESCs by
checking the GFP expression under a fluorescence microscope. After culturing the ESCs in the
NB medium for 4 days, I could not see any Sox1-GFP positive (Sox1-GFP
+
) cells under a
fluorescence microscope. In contrast, I could see Sox1-GFP
+
cells among the ESCs cultured with
NB media supplemented with different concentrations of NaCl. In particular, there were a large
number of Sox1-GFP
+
cells among the ESCs subjected to neural differentiation in the NB medium
supplemented with 30mM NaCl or 50mM NaCl (Figure 2.3). These data suggest that NaCl-
addition promotes the emergence of Sox1-GFP
+
NSCs in the ESCs cultured in the NB medium.
To further assess the effect of NaCl-addition on the neural differentiation, I quantified the
percentage of Sox1-GFP
+
cells in the ESCs subjected to neural differentiation in the NB medium
with NaCl-addition on day 6 by flow cytometry (FC) analysis. It was revealed that NaCl-addition
Figure 2.3 Addition of NaCl accelerates the emergence of Sox1-GFP+ NSCs in the differentiating ESCs
cultured in NB medium.
Representative images for 46c ESCs carrying a Sox1-GFP knock-in reporter subjected to neural differentiation
under the indicated conditions on day 4. NB: NeuroBasal medium with N-2 and B-27
TM
supplements. Scale bar:
100µm. BF: bright-field.
123
groups had higher percentages of Sox1-GFP
+
cells (Figure 2.4A). For cells differentiated in the
NB medium without additional NaCl, the percentage of Sox1-GFP
+
cells was 40.7 ± 1.5 % (Figure
2.4B); however, the percentages of Sox1-GFP
+
cells in the three NaCl supplemented groups were
Figure 2.4 Addition of NaCl increases the percentages of Sox1-GFP
+
NSCs among the differentiating ESCs
cultured in NB medium.
(A) Representative histograms of flow cytometry (FC) data of Sox1-GFP
+
cells among ESCs subjected to neural
differentiation under the indicated conditions on day 6. The undifferentiated ESCs was used as the negative control
for gating in FC analysis. (B) The percentages of Sox1-GFP
+
cells in FC analysis under indicated conditions. The
data represent two independent experiments and are shown as mean ± SEM. *: p<0.05; **: p<0.01; ***: p<0.001.
Figure 2.5 NaCl dose-dependently promotes the production of Sox1-GFP
+
NSCs from ESCs cultured in NB
medium.
Linear regression analysis of the average percentages of Sox1-GFP
+
NSCs (from Figure 2.4B) in ESCs subjected
to neural differentiation under the indicated conditions on day 6. NB: NeuroBasal medium with N-2 and B-27
TM
supplements.
124
52.5 ± 0.7 % in NB +10mM NaCl group, 59.8 ± 1.6 % in NB + 30 mM NaCl group, and 87.3 ±
1.3 % in NB + 50mM NaCl group, respectively. Interestingly, there was a linear relationship
between the concentrations of NaCl added and the percentages of Sox1-GFP
+
cells formed in the
medium (Figure 2.5). Taken together, I conclude that the addition of NaCl in NB medium is able
to promote neural differentiation of ESCs
2. Addition of Inorganic or Organic Osmolytes Promotes Neural Differentiation of
ESCs in NeuroBasal Medium with N-2 and B-27
TM
Supplements
To examine whether the neural differentiation-promoting effect of NaCl is specifically due to the
Na
+
or Cl
-
ions, I next tried adding different salts into NB medium and evaluated the production
of GFP+ cells. Salts with neutral pH were selected to avoid changing the pH of the medium, which
may affect neural differentiation as well (Fang et al., 2017). Additionally, salts with bivalent
cations (e.g. Mg
2+
) or bivalent anions (e.g. SO
4
2−
) were tested to examine the Na
+
or Cl
-
specific
effect. Based on the two above criteria, I chose KCl, MgCl
2
, and Na
2
SO
4
for the further
experiments.
To make the final osmolalities of NB media with salts addition similar to each other, I adjusted the
concentrations of the supplemented salts in the NB medium. For instance, 1mM KCl in solution
generates two milliosmoles of osmolality per liter. In contrast, 1mM MgCl
2
and 1mM Na
2
SO
4
both produce three milliosmoles per liter. As such, NB medium with 30mM KCl is identical to
that with 20mM MgCl
2
or 20mM Na
2
SO
4
with regards to osmolality. The osmolalities of NB +
30mM KCl, NB + 20 mM MgCl2, and NB + 20mM Na2SO4 were measured, which were 265 ±
0.7 mOsm/kg, 262 ± 0.8 mOsm/kg, and 266 ± 1.5 mOsm/kg, respectively. The osmolalities of
125
these media were similar to those of N2B27 medium (265 ± 0.6 mOsm/kg) and the NB medium
with 30mM NaCl.
I first checked if the addition of KCl, MgCl
2
, and Na
2
SO
4
could facilitate the differentiation of
ESCs into Sox1-GFP
+
NSCs. All the three salts accelerated the emergence of Sox1-GFP
+
NSCs in
the ESCs undergoing neural differentiation (Figure 2.6A). I then quantified the differentiating
ESCs with the addition of different salts on day 6, and found that there were significantly higher
percentages of Sox1-GFP
+
NSCs among the differentiating ESCs with the addition of different
salts (KCl, MgCl
2
, or Na
2
SO
4
) than that in the NB medium alone (Figure 2.6B and 2.7). These
results suggest the addition of different salts to the NB medium all promotes the efficiency of
neural differentiation of ESCs.
To further determine whether the neural differentiation-promoting effect is mediated by ions or a
shared feature of increased osmolality by both inorganic or organic solutes, I added glucose into
NB medium and checked the neural differentiation. Glucose does not dissociate into cations and
Figure 2.6 Addition of different salts promotes the differentiation of ESCs into Sox1-GFP
+
NSCs among the
differentiating ESCs cultured in NB medium.
(A) Representative images for ESCs subjected to neural differentiation in NB medium and the NB medium
supplemented with KCl, MgCl2, or Na
2
SO
4
on day 4. BF: bright-field. Scale bar: 100µm. (B) Representative
histograms of FC data of Sox1-GFP
+
cells among ESCs subjected to neural differentiation under the indicated
conditions on day 6. The undifferentiated ESCs was the negative control for gating in FC analysis.
126
anions, and therefore eliminates the specific ion mediated effects. 60mM glucose and 100mM
glucose were individually supplemented to NB medium, and the osmolalities of prepared media
were increased to 260 ± 1.2 mOsm/kg and 295 ± 0.5 mOsm/kg, respectively. I used the NB media
supplemented with glucose for neural differentiation and the Sox1-GFP
+
NSCs were checked
under a fluorescence microscope. Results showed that there were very few, if any, Sox1-GFP
+
NSCs in the NB medium with 60mM glucose on day 4 (Figure 2.8A), which was similar to that in
the NB medium. In contrast, a large population of Sox1-GFP
+
cells appeared in the NB with
100mM glucose group. Consistent with the above observation, FC analysis revealed that the
percentage of Sox1-GFP+ NSCs in the NB plus 100mM glucose was significantly higher than that
in the NB with/without 60mM glucose (Figure 2.7 and 2.8B). These data revealed that the addition
of organic omsolyte was also able to promote the formation of NSCs during neural differenation
of ESCs in NB medium.
Figure 2.7 The addition of different osmolytes increases the percentages of Sox1-GFP
+
NSCs among the
differentiating ESCs cultured in NB medium.
The percentages of Sox1-GFP
+
cells in FC analysis under indicated conditions. The data represent two independent
experiments and are shown as mean ± SEM. *: p<0.05; **: p<0.01; ***: p<0.001.
127
Taken together, our findings suggest that increased medium osmolalities by adding different
osmolytes in NB medium facilitate the differentiation of ESCs into NSCs.
3. Activation of MAPK/ERK Signaling Is Involved in the Neural Differentiation of
ESCs Promoted by Increased Medium Osmolality
The differences between medium osmolality and intracellular osmolality define the osmotic
pressure on the cell membrane and cytoskeleton (Zhou et al., 2016). Osmotic pressure, as a stress
signal, activates multiple signaling pathways in cells. The members of the MAPK superfamily are
especially the crucial mediators in response to extracellular stimuli and evolutionarily conserved
from invertebrates to vertebrates (Plotnikov et al., 2011). Indeed, not only stress kinases (JNK and
p38) but also ERK1/2 and ERK5 are involved in the cellular responses to osmotic pressure. The
roles of highly conserved MAPK pathways, including ERK1/2-, JNK-, and ERK5-dependent
Figure 2.8 Addition of glucose promotes the differentiation of ESCs into Sox1-GFP
+
NSCs among the
differentiating ESCs cultured in NB medium.
(A) Representative images for ESCs subjected to neural differentiation under the indicated conditions on day 4.
BF: bright-field. Scale bar: 100µm. (B) Representative histograms of FC data of Sox1-GFP
+
cells among ESCs
subjected to neural differentiation under the indicated conditions on day 6. The undifferentiated ESCs was the
negative control for gating in FC analysis.
128
signaling, in neural differentiation have been well established (Chuang et al., 2015; Nishimoto et
al., 2005; Williams et al., 2016).
Figure 2.9 The MAPK/ERK pathway inhibitors suppress the formation of Sox1-GFP
+
NSCs from
differentiating ESCs cultured in NB medium supplemented with NaCl.
Representative images of ESCs subjected to neural differentiation under the indicated conditions on day 4 with
the addition of MAPK pathway inhibitors. The NB medium group was used as the negative control. SP600125:
JNK inhibitor; SB203580: p38 inhibitor; BIX02189: MEK 5 inhibitor; PD: MEK inhibitor; Vx11e: ERK inhibitor.
BF: bright-field. Scale bar: 100µm.
129
To investigate the molecular pathways involved in the promotion of Sox1-GFP
+
cells formation
due to increased osmolality, I tested the effects of various MAPK pathway inhibitors on the neural
differentiation of ESCs cultured in NB medium with the addition of 30mM NaCl. The results
indicate none of the p38 inhibitor (SB203580), the MEK5 inhibitor (BIX02189), or the JNK
inhibitor (SP600125) could blocked the emergence of Sox1-GFP
+
NSCs in this medium on day 4
(Figure 2.9). In contrast, the well-known MEK inhibitor (PD0325901, PD) significantly reduced
the number of Sox1-GFP
+
cells in the 30mM NaCl supplemented medium. Additionally, the ERK
inhibitor (Vx11e) had a similar inhibitory effect as that of the MEK inhibitor on Sox1-GFP
+
cell
production mediated by NaCl-addition. Furthermore, I quantified the Sox1-GFP
+
cells in the ESCs
subjected to neural differentiation cultured in NB medium with 30mM NaCl and MAPK pathway
inhibitors by flow cytometry. Consistent with the results seen under a fluorescence microscope,
neither the p38 inhibitor nor the MEK5 inhibitor could not decrease the percentage of Sox1-GFP
+
Figure 2.10 The MAPK/ERK pathway inhibitors suppress the promoting effect of NaCl-addition on neural
differentiation of ESCs cultured in NB medium.
(A) Representative histograms of FC data of Sox1-GFP
+
cells among ESCs subjected to neural differentiation under
the indicated conditions on day 6. The undifferentiated ESCs was used as the negative control for gating in FC
analysis. (B) The quantification of Sox1-GFP
+
NSCs in (A). SP600125: JNK inhibitor; SB203580: p38 inhibitor;
BIX02189: MEK 5 inhibitor; PD03: MEK inhibitor; Vx11e: ERK inhibitor. The data represent two independent
experiments and are shown as mean ± SEM. *: p<0.05; **: p<0.01.
130
NSCs among those ESCs treated with 30mM NaCl (Figure 2.10). Though JNK inhibitor could
significantly decrease the percentage of Sox1-GFP
+
cells, there was still a very high percentage of
Sox1-GFP
+
cells formed (32.80 ± 4.81 %) (Figure 2.10). In contrast, both the MEK inhibitor and
the ERK inhibitor significantly reduced the percentages of Sox1-GFP
+
NSCs cultured in the NB
medium with 30mM NaCl (Figure 2.10B). These results suggested that the MAPK/ERK pathway
was involved in the promoting effects of increased osmolality on neural differentiation of ESCs.
To explore whether NaCl can independently activate MAPK/ERK pathway to mediate the
osmolality-specific role in neural differentiation, I next examined the changes in the level of
phosphorylated ERK (pERK1/2), an indicator of the activation of MAPK/ERK signaling, in
response to the addition of NaCl. To avoid the stimulatory effects of autocrine FGF4 on the
MAPK/ERK signaling (Ying et al., 2003), I first evaluated the response of MAPK/ERK signaling
to NaCl addition in Fgf4
-/-
ESCs. It was revealed that NaCl could activate MAPK/ERK signaling
based on the level of pERK in Fgf4
-/-
ESCs, and the addition of exogenous FGF4 further enhanced
the activation effect of MAPK/ERK signaling by NaCl (Figure 2.11A). Similarly, the addition of
NaCl was able to activate MAPK/ERK signaling in the 46c ESCs (Figure 2.11B). The combination
of NaCl and FGF4 also had a synergistic effect on the activation of MAPK/ERK signaling. These
results showed that the supplementation of NaCl into NB medium could activate MAPK/ERK
signaling, which may mediate the effect of increased osmolality on neural differentiation of ESCs
in NB medium.
131
To test the above hypothesis about the role of MAPK/ERK signaling in osmolality regulated neural
differentiation, I added FGF4 into the NB medium to see if FGF4 could mimic the effect of
increased osmolality on neural differentiation. I first examined the Sox1-GFP
+
NSCs in NB
Figure 2.11 NaCl activates the MAPK/ERK pathway in ESCs cultured in NB medium.
Western blot analysis of the expression of phosphorylated ERK (pERK1/2) and total ERK (tERK1/2) in Fgf4
-/-
ESCs (A) and WT ESCs (B) under the indicated conditions. Cells were starved in serum-free condition overnight
and cell lysates were then collected after stimulation with LIF (10ng/mL), NaCl (30mM), FGF4 (5ng/mL), or the
combination of NaCl and FGF4 for 24 hours. NT: NB medium. LIF: leukemia inhibitory factor; FGF4: fibroblast
growth factor. ERK1:44 kDa; ERK2:42 kDa.
Figure 2.12 Addition of FGF4 promotes the differentiation of ESCs into Sox1-GFP
+
NSCs among the
differentiating ESCs cultured in NB medium.
(A) Representative images of ESCs subjected to neural differentiation in the NB medium with or without FGF4
(10ng/mL) on day 4. Scale bar: 100µm. (B) Representative histograms of FC data of Sox1-GFP
+
cells among ESCs
subjected to neural differentiation in the NB medium with or without FGF4 (10ng/mL) on day 6. The
undifferentiated ESCs were used for gating in FC analysis. (C) The quantification of Sox1-GFP
+
cells in (B). The
data represent two independent experiments and are shown as mean ± SEM. *: p<0.05.
132
medium after the FGF4-addition under a fluorescence microscope, and there were some Sox1-
GFP
+
NSCs on day 4 (Figure 2.12A). This finding implied the promoting effect of FGF4 on Sox-
GFP
+
NSCs formation. Consistent with the above result, FCS analysis showed that the percentage
of Sox1-GFP
+
cells in the differentiating ESC population was also significantly increased (Figure
2.12). In summary, the increased osmolality of differentiation medium activates the MAPK/ERK
pathway, which facilitates the differentiation of ESCs into NSCs.
4. ERK-Independent Signaling Is Involved in the Neural Differentiation of ESCs
Promoted by Increased Medium Osmolality
To test whether MAPK/ERK independent pathway(s) also plays a role in promoting the neural
differentiation in the NB medium with increased osmolality, I used the Erk1/2 double knock-out
ESCs (Erk1/2 DKO ESCs) (Figure 1.9 in Part I) as a tool to test neural differentiation in NB
medium. While Erk1/2 DKO ESCs preserved their ability to differentiate into Sox1-GFP
+
NSCs,
the low efficiency of neural differentiation in these Erk1/2 DKO ESCs highlights that MAPK/ERK
signaling is important but not essential in promoting neural induction (Figure 1.16). I then asked
whether elevated osmolality could still promote the formation of Sox1-GFP
+
cells in the Erk1/2
DKO ESCs. Interestingly, the supplementation of NaCl into the NB medium accelerated the
emergence of GFP+ cells and the outgrowth of neurites from the Erk1/2 DKO ESCs (Figure 2.13).
Furthermore, the addition of glucose had similar effects on the Erk1/2 DKO ESCs. These results
suggest that there are other signaling pathways responsible for the increased neural differentiation
by NaCl-elevated osmolality.
Considering the critical role of bone morphogenetic protein (BMP) signaling inhibition in the
neural differentiation (Stern, 2005), I reasoned that BMP signaling inhibition by the elevated
133
osmolality may play a role in the acceleration of neural induction. Our data showed that the
addition of BMP4 into the NB medium with 30mM NaCl blocked the emergence of Sox1-GFP
+
cells in wild-type ESCs, as evidenced by the negligible proportion of Sox1-GFP
+
cells in the flow
cytometry data (Figure 2.14). In addition, BMP4 could also suppress neural differentiation of the
Erk1/2 DKO cells in the NB plus 30mM NaCl medium, suggesting that BMP signaling inhibition
Figure 2.13 Increased osmolality by NaCl promotes neural differentiation of ESC cultured in NB medium
via an ERK-independent pathway.
(A) Representative images of Erk1/2 DKO ESCs subjected to neural differentiation under the indicated conditions
on day 6. The outgrowth of neurites is indicated by red arrow and Sox1-GFP
+
NSC clusters are indicated with white
arrowhead. BF: bright-field. DMH1: BMP receptor inhibitor. Scale bars: 100µm. (B) Representative
immunofluorescent staining images for TUJ1 in the Erk1/2 DKO ESCs subjected to neural differentiation under the
indicated conditions on day 7. The white arrow indicates the sparse neurites under the glucose-supplemented
condition. TUJ1: neuronal lineage marker. Scale bars: 100µm.
134
is required for osmolality-dependent enhancement of neural differentiation. These results also
highlight the necessity and importance of BMP signaling inhibition in neural differentiation. In
line with this, a combination of the BMP receptor inhibitor (DMH1) and NaCl further enhances
the outgrowth of neurites and advances the formation of Sox1-GFP
+
cells in wild-type ESCs
(Figure 2.14). Collectively, these data demonstrate that increased osmolality in Neurobasal
medium could promote neural differentiation of ESCs via ERK-independent pathway(s).
Figure 2.14 The enhancement of neural differentiation by increased osmolality by NaCl is dependent on the
inhibition of BMP signaling.
(A) Representative images of WT ESCs and the Erk1/2 DKO ESCs subjected to neural differentiation in the NB
medium supplemented with 30mM NaCl and BMP4 (10ng/ml) on day 4. BMP4: bone morphogenetic protein 4. WT:
wild-type. Scale bar: 100µm. (B) Representative histograms of FC data of Sox1-GFP
+
cells among ESCs subjected to
neural differentiation under the indicated conditions. The undifferentiated ESCs were used for gating in FC analysis.
(C) The quantification of Sox1-GFP
+
cells in NB + 30mM NaCl group and NB + 30mM NaCl + BMP4 group. The
data represent two independent experiments and are shown as mean ± SEM. **: p<0.01.
135
Discussion
In this study, we adjusted the osmolality of the neural differentiation medium by supplementing
different osmolytes and revealed that elevated osmolality promoted ESC differentiation towards
Sox1-GFP
+
NSCs. By screening with different MAPK inhibitors, we found that the JNK pathway,
p38 pathway, and ERK5 pathway are not involved in the induction of Sox1-GFP
+
NSCs mediated
by increased osmolality. Activation of MAPK/ERK signaling by increased osmolality facilitates
neural conversion of ESCs, which can be suppressed by MAPK/ERK pathway inhibitors.
Additionally, we generated Erk1/2 double knock-out ESCs using CRISPR/Cas9 genome editing
technology and provided compelling evidence that the MAPK/ERK pathway is not necessary for
neural differentiation of ESCs. Finally, we found that activation of MAPK/ERK-independent
pathway(s) is implicated in the regulation of neural induction by increased osmolality as well.
The importance of medium osmolality in cell culture has long been established (Waymouth, 1970).
The effects of osmolality on survival, proliferation, metabolism, and differentiation have been
reported in mesenchymal stem cells, neurons, and kidney cells (Brewer et al., 1993; Caron et al.,
2013; Casali et al., 2013; Goransson et al., 2001). However, current knowledge relating to the
effects of osmolality on ESC self-renewal and differentiation is very limited (Fang et al., 2017;
Slater et al., 2014). We here demonstrate for the first time that the osmolality of medium, as a
physical parameter, can regulate neural differentiation. Different osmolytes, including both
inorganic and organic solutes, were tested in our monolayer neural differentiation system. Of note,
media with identical osmolalities prepared with different osmolytes have different efficiencies in
promoting neural differentiation (Figure 2.7). In particular, supplementation of 60mM glucose
subtly increased the percentage of Sox1-GFP
+
NSCs in the whole population, though increasing
the glucose concentration to 100mM significantly promoted neural induction. The effects observed
136
in using glucose as a solute may be due to two reasons. Firstly, there are ion-specific activities in
initiating and accelerating neural induction. In support of this notion, ESCs cultured in PBS buffer
consisting of ionic salts can generate Nestin
+
/Sox1
+
neural progenitors in only 4 hours (Smukler
et al., 2006). In cell aggregates isolated from frog embryos, ionic strength determined ectoderm
cell fates (Barth and Barth, 1974). Secondly, the glucose may have been quickly consumed by the
proliferating cells and thus sharply decrease the osmolality of medium as time goes. As a result, a
higher concentration of glucose is necessary to keep the osmolality beyond an appropriate
threshold for promoting neural differentiation.
In addition, a combination of drug screening and genetic manipulation allowed us to recognize
that both MAPK/ERK -dependent and -independent pathways were implicated in enhancing neural
differentiation due to increased osmolality. With regards to MAPK/ERK signaling, increased
osmolality may have either directly triggered the activation of the MAPK/ERK pathway or
indirectly activated the MAPK/ERK pathway by stimulating the synthesis of cytokines such as
growth factors (Couper et al., 1994; Zhang et al., 2012). Autocrine cytokines may maintain the
long-term activation of Erk signaling and thus enhance the neural induction, similar to the addition
of FGFs (Ying et al., 2003). On the other hand, ERK-independent pathways in the osmolality-
dependent promotion of neural differentiation rely on the inhibition of BMP signaling, supporting
the role of BMP signaling in the default model of neural induction. Furthermore, the requirement
of BMP inhibition for the promotion of high-osmolality dependent neural differentiation indicates
that the ERK-independent pathway may work upstream of the Smad4 or Smad1/5/8 proteins, the
key nodes of BMP signaling. In fact, increased osmolality can induce an influx of Ca
2+
and activate
calcium signaling, which may affect the nuclear translocation of Smad proteins (Caron et al., 2013;
137
Cho et al., 2014; Fang et al., 2017). The cytoplasmic accumulation of Smads helps stop the signal
transduction of BMP signaling, potentiating ESC neuralization.
138
Perspectives and Future Directions
My studies on Erk1/2 DKO cells demonstrate that Erk-independent but BMP-dependent signaling
is responsible for enhanced neural induction due to increased osmolality. As they are important
mediators of BMP signaling, the regulation of R-Smads (1/5/8) and Smad4 by changes in
osmolality is worth further exploration. Different means of regulating Smads include
phosphorylation at different domains, ubiquitination, cytoplasmic retention, and so on (Xu, 2006).
In fact, the soft matrix has been reported to promote neural differentiation by keeping Smads
proteins in the cytoplasm, thereby suppressing BMP dependent signaling (Sun et al., 2014). On
top of the BMP signaling mediators, we may need to further investigate other regulators in BMP
signaling in the future. Pathways with established roles in neural differentiation and in cross-talk
with BMP signaling are worth study as well. The identification of the ERK-independent pathway(s)
in neural induction will help us understand neural lineage specification and provide clues for us to
further optimize the neural differentiation protocol and establish industry-scale procedures for use
in regenerative medicine.
The effect of medium osmolality on neural differentiation was tested using the adherent monolayer
differentiation protocol. It would be interesting to explore the role of medium osmolality on neural
induction in other types of neural differentiation protocols, and in particular, EB-based protocols
that prime ESCs to commit to the three germ layers. This would help address the effects of
osmolality changes on differentiation into the ectoderm, mesoderm, and endoderm. Furthermore,
osmolality can have diverse effects on the terminal differentiation of a specific germ layer. For
instance, the Neurobasal medium with low osmolality favors the expansion of neurons rather than
glial cells (Brewer et al., 1993). With this new knowledge in mind, we may optimize current
protocols for the expansion of specific cell types of interest.
139
Materials and Methods
Cell Culture
Mouse embryonic stem cells (ESCs) were cultured on 0.1% gelatin-coated plates without a feeder
layer at 37 °C in a 5% CO2 incubator. The basal medium (refer to Table 1.1 in the section of
Materials and Methods in Part I) was a high glucose DMEM medium (Thermo Fisher Scientific,
Waltham, MA) containing 10% Hyclone fetal bovine serum (Thermo Fisher Scientific, Waltham,
MA), 1mM sodium pyruvate (Sigma, St. Louis, MO), 1% MEM non-essential amino acids
(Thermo Fisher Scientific, Waltham, MA), and 0.1mM b-mercaptoethanol (Sigma, St. Louis, MO).
To maintain the mouse ESCs, 10ng/mL leukemia inhibitory factor (LIF; PeproTech, Rocky Hill,
NJ) was added to the basal medium. Cells were passaged when confluency reached 60-70%. To
dissociate the cells from culture plates, a 0.025% trypsin solution with 1% chicken serum (Thermo
Fisher Scientific, Waltham, MA) was used. After neutralization of trypsin and collection of cells
by centrifugation (300g, 3min), cells were sub-cultured at a split ratio of 1: (6-10).
Two different mouse embryonic stem (ES) cell lines, 46c ESCs and Fgf4 knock-out (Fgf4
-/-
) ESCs,
were used in this study. 46c ESCs (a Sox1-GFP knock-in ESC line) were a gift from Dr. Austin
Smith’s lab (Wellcome-MRC Cambridge Stem Cell Institute, Cambridge, UK) (Ying et al., 2003).
In 46c ESCs, green fluorescent protein (GFP) is under the control of the promoter for the
neuroectoderm and neural stem cell (NSC) marker gene Sox1. When 46c ESCs differentiate into
Sox1 positive (Sox1+) NSCs, GFP is expressed together with Sox1, providing an easy readout for
NSC transition in live cell culture under a fluorescence microscope. Fgf4
-/-
ESCs were a kind gift
from Angie Rizzino’s lab (Eppley Institute for Cancer Research, Omaha, NE) (Wilder et al., 1997).
140
Fibroblast growth factor 4 (FGF4), the major autocrine cytokine stimulating the MAPK/ERK
pathway, is not expressed in the Fgf4
-/-
ESCs.
Monolayer Neural Differentiation
Serum-free N2B27 medium was a 1:1 mixture of Neurobasal medium (Thermo Fisher Scientific,
Waltham, MA) and DMEM/F12 medium (Thermo Fisher Scientific, Waltham, MA) supplemented
with 0.5% N2 supplements (Thermo Fisher Scientific, Waltham, MA), 1% B27 supplements
(Thermo Fisher Scientific, Waltham, MA), 2mM L-glutamine (Thermo Fisher Scientific, Waltham,
MA) and 0.1mM b-mercaptoethanol (Sigma, St. Louis, MO) as described previously in the
Materials and Methods section in Part I. The Neurobasal medium with 0.5% N2 supplements
(Thermo Fisher Scientific, Waltham, MA), 1% B27 supplements (Thermo Fisher Scientific,
Waltham, MA) is termed NB medium.
46c ESCs were plated in the basal medium with LIF at a density of (7.5-10) x10
3
cells/well in a
12-well plate. If a 6-well plate was used, the cell number for each well was doubled. After
overnight culture, the basal medium was aspirated and cells were washed with pre-warmed 1x PBS
solution to remove serum, which suppresses neural induction. Following the wash, pre-warmed
NB or N2B27 medium was used to induce ESC neural differentiation. The stock solution of
glucose and different salts (NaCl, KCl, MgCl
2
, and Na
2
SO
4
) were supplemented into the medium
at the desired concentrations, and the medium for neural differentiation was changed every other
day.
Chemical Inhibitors and Cytokines
Chemical inhibitors and cytokines were prepared following the manufacturers’ instructions. The
stock solutions of chemical inhibitors were dissolved in dimethyl sulfoxide (DMSO) and cytokines
141
in water with 0.1% bovine serum albumin (BSA). The aliquots of stock solution were stored at -
80°C until use. Inhibitors and their working concentrations used in this study were as follows:
SB203580 (p38 inhibitor; Sigma, S8307), 10µM; SP600125 (JNK inhibitor; Sigma, S5567), 5µM;
BIX02189 (MEK5 inhibitor; APExBio, A5801), 10µM; PD0325901 (MEK1/2 inhibitor; Tocris,
4192), 1µM; Vx11e (ERK1/2 inhibitor; APExBio, A9301), 1µM. FGF4 (PeproTech, Rocky Hill,
NJ) was supplemented into medium at a working concentration of 5ng/ml. Among the chemical
inhibitor-treated groups, the control group was treated with DMSO. Among the cytokine-treated
groups, the control group was treated with 0.1% BSA in water.
Osmolality Control and Measurement
Medium osmolality was measured with a Vapro 5520 Vapor Pressure Osmometer (Wescor, Logan,
UT) following the manufacturers' protocols. Each type of medium (e.g. Neurobasal medium,
N2B27 medium, or DMEM/F12 medium) was measured three times. Three independent batches
of different media were measured and the osmolality of medium was presented in mean ± SEM.
Western Blotting
Before cell lysate collection, cells were pre-washed with an ice-cold 1x PBS solution twice to
remove the dead cells and medium. The cell lysis buffer for collecting protein lysates from cells
was a mixture of RIPA lysis buffer (Teknova, Hollister, CA) with cOmplete™ Protease Inhibitor
Cocktail (Roche, Pleasanton, CA) and PhosSTOP-Phosphatase Inhibitor Cocktail (Roche,
Pleasanton, CA) following the manufacturer’s instructions. To lyse the cells, ice-cold lysis buffer
(100µl per 1x10
6
cells) was added. Cells were scraped off the plate using a cold plastic cell scraper,
and the cell suspension was gently transferred into a pre-cooled Eppendorf tube. The cell
suspension was constantly agitated at 4°C every 5min and in total for 30min followed by
142
centrifugation (16000g, 45min) at 4°C. The supernatant was then transferred to a new ice-cold
tube for protein quantification.
Protein concentrations were determined using the Piece BCA Protein Assay Kit (Thermo Fisher
Scientific, Waltham, MA) following the manufacturer’s instructions. Plates were read with
SoftMax Pro software using a SpectraMax i3x microplate reader (Molecular Devices, San Jose,
CA). For each sample, approximately 25µg total protein was mixed with the appropriate volume
of 5x Laemmli Sample buffer and b-Mercaptoethanol (5% final concentration), heated (100°C,
8min), and separated on a 10% SDS-polyacrylamide electrophoresis gel. After electrophoresis,
proteins were transferred onto a PVDF membrane (Biorad, Hercules, CA), which was then blocked
with 5% nonfat milk (blocking-grade; Biorad, Hercules, CA) in 1x Tris-buffered saline (TBS)
buffer (150mM NaCl, 50mM Tris-HCl, pH 7.9) for 1h at room temperature. Following blocking,
the membrane was incubated overnight at 4°C with primary antibodies that were diluted with 5%
bovine serum albumin in 1x TBS buffer. After washing the membrane with 1x TBS buffer
containing Triton X-100 (0.1%), species-specific and HRP-conjugated secondary antibodies
(Santa Cruz, Santa Cruz, CA) were diluted (1:1000) in blocking solution and incubated with the
membrane for 1h at room temperature. Primary antibodies and their dilutions used in the study
were as follows: ERK1/2 (1:1000, CST, #9102S), Phospho-p44/42 MAPK (pErk1/2)
(Thr202/Tyr204) (1:1000, CST, #4370). Anti-Rabbit HRP-conjugated secondary antibodies
(Santa Cruz, Santa Cruz, CA) were diluted with 5% non-fat milk in 1x TBS buffer at a ratio of
1:2000.
For the detection of proteins, the Pierce™ ECL Western Blotting Substrate (Thermo Fisher
Scientific, Waltham, MA) was applied on the antibody-conjugated PVDF membrane according to
143
the manufacturer’s recommendation. Chemiluminescent signals were captured by a CCD camera-
based imager FluorChem E (ProteinSimple, San Jose, CA).
Flow Cytometry Analysis
Flow cytometry (FCS) analysis was performed to quantify the percentage of Sox1-GFP+ neural
stem cells in the differentiating 46c ESC population using an adherent monolayer neural
differentiation protocol.
Cells were detached from the plate with a 0.025% trypsin/1% serum solution (Thermo Fisher
Scientific, Waltham, MA) and collected with centrifugation (300g, 3min). The supernatant was
discarded and cell pellets were washed with a 1x PBS solution twice. After the washing, cells were
suspended in the ice-cold 1x PBS solution with 0.1% Hyclone fetal bovine serum (Thermo Fisher
Scientific, Waltham, MA) and kept on ice before loading on a flow cytometer for analysis.
Before quantification of the percentage of Sox1-GFP+ cells in a sample, cells were re-suspended
by gentle tapping before loading, and the data were collected on a BD
TM
LSR II microflow
cytometer (Becton Dickinson, Franklin Lakes, NJ). Dead cells and debris were eliminated from
the analysis using forward and side scatter parameters. The cell number of alive cells collected for
analysis was set as 1x 10
4
. GFP was excited by an argon laser and fluorescence was detected using
a 530/30 nm bandpass filter in the FL1 channel.
To analyze the data, FlowJo X (the 10.0.7r2 version) was used. GFP negative undifferentiated 46c
ESCs were used as the negative control for gating. Each tested condition was assessed in triplicate.
Two independent experiments were performed.
144
Immunofluorescence Staining
Cells were first fixed in 4% paraformaldehyde for 15 min at room temperature and then incubated
with the blocking buffer (1x PBS solution with 5% BSA and 0.2% TritonX-100) at 37°C for 1
hour. After fixation and perforation, cells were incubated at 4°C overnight with primary antibodies
diluted in blocking buffer and subsequently washed three times with 1x PBS solution for 5 min
per wash. Cells were then incubated with secondary antibodies at 37°C for 1 h (or at 4°C overnight).
Nuclei were counter-stained with Hoechst 33342 (1:5000, Thermo Fisher Scientific, Waltham,
MA) at room temperature for 10 min. Before imaging, cells were washed three times with 1x PBS
solution for 5 min per wash. The primary antibody and its dilution used in this study was β3-
tubulin (Tuj1) (1:200, CST, #4466S), and the Alexa Fluor® 546 fluorescent secondary antibody
(Thermo Fisher Scientific, Waltham, MA) was used at a 1:1000 dilution. Images from random
fields (at least five) were taken with a Zeiss Fluorescence Microscope (Carl Zeiss Microscopy,
Thornwood, NY).
Statistical Analysis
Images were from results of a representative experiment. Data are based on at least two
independent experiments. Statistical differences were analyzed in GraphPad Prism 6 (GraphPad,
La Jolla, CA) using two-tailed Student’s t-tests for comparisons of the two groups (e.g. NB +
10mM/30mM/50mM NaCl vs. NB, NB + 30mM KCl vs. NB, NB + 20mM MgCl
2
vs. NB, NB +
20mM Na
2
SO
4
vs. NB, or NB + 30mM NaCl + inhibitors vs. NB + 30mM NaCl) in each
experiment. All data were presented as mean ± SEM. Data with p<0.05 was considered statistically
significant. *: p<0.05; **: p<0.01; ***: p<0.001.
145
Bibliography
Ali, S., Wall, I.B., Mason, C., Pelling, A.E., and Veraitch, F.S. (2015). The effect of Young's
modulus on the neuronal differentiation of mouse embryonic stem cells. Acta Biomater 25, 253-
267.
Amura, C.R., Marek, L., Winn, R.A., and Heasley, L.E. (2005). Inhibited neurogenesis in JNK1-
deficient embryonic stem cells. Mol Cell Biol 25, 10791-10802.
Atlasi, Y., Noori, R., Gaspar, C., Franken, P., Sacchetti, A., Rafati, H., Mahmoudi, T., Decraene,
C., Calin, G.A., Merrill, B.J., et al. (2013). Wnt signaling regulates the lineage differentiation
potential of mouse embryonic stem cells through Tcf3 down-regulation. PLoS Genet 9, e1003424.
Aubert, J., Dunstan, H., Chambers, I., and Smith, A. (2002). Functional gene screening in
embryonic stem cells implicates Wnt antagonism in neural differentiation. Nat Biotechnol 20,
1240-1245.
Bachiller, D., Klingensmith, J., Kemp, C., Belo, J.A., Anderson, R.M., May, S.R., McMahon, J.A.,
McMahon, A.P., Harland, R.M., Rossant, J., et al. (2000). The organizer factors Chordin and
Noggin are required for mouse forebrain development. Nature 403, 658-661.
Bain, G., Kitchens, D., Yao, M., Huettner, J.E., and Gottlieb, D.I. (1995). Embryonic stem cells
express neuronal properties in vitro. Dev Biol 168, 342-357.
Bakre, M.M., Hoi, A., Mong, J.C., Koh, Y.Y., Wong, K.Y., and Stanton, L.W. (2007). Generation
of multipotential mesendodermal progenitors from mouse embryonic stem cells via sustained Wnt
pathway activation. J Biol Chem 282, 31703-31712.
Banda, E., and Grabel, L. (2016). Directed Differentiation of Human Embryonic Stem Cells into
Neural Progenitors. Methods Mol Biol 1307, 289-298.
Barberi, T., Klivenyi, P., Calingasan, N.Y., Lee, H., Kawamata, H., Loonam, K., Perrier, A.L.,
Bruses, J., Rubio, M.E., Topf, N., et al. (2003). Neural subtype specification of fertilization and
nuclear transfer embryonic stem cells and application in parkinsonian mice. Nat Biotechnol 21,
1200-1207.
Barth, L.G., and Barth, L.J. (1974). Ionic regulation of embryonic induction and cell differentiation
in Rana pipiens. Dev Biol 39, 1-22.
Beddington, R.S. (1994). Induction of a second neural axis by the mouse node. Development 120,
613-620.
Brewer, G.J., Torricelli, J.R., Evege, E.K., and Price, P.J. (1993). Optimized survival of
hippocampal neurons in B27-supplemented Neurobasal, a new serum-free medium combination.
J Neurosci Res 35, 567-576.
146
Brickman, J.M., and Serup, P. (2017). Properties of embryoid bodies. Wiley Interdiscip Rev Dev
Biol 6.
Brunet, T., Bouclet, A., Ahmadi, P., Mitrossilis, D., Driquez, B., Brunet, A.C., Henry, L., Serman,
F., Bealle, G., Menager, C., et al. (2013). Evolutionary conservation of early mesoderm
specification by mechanotransduction in Bilateria. Nat Commun 4, 2821.
Cai, C., and Grabel, L. (2007). Directing the differentiation of embryonic stem cells to neural stem
cells. Dev Dyn 236, 3255-3266.
Caron, M.M., van der Windt, A.E., Emans, P.J., van Rhijn, L.W., Jahr, H., and Welting, T.J.
(2013). Osmolarity determines the in vitro chondrogenic differentiation capacity of progenitor
cells via nuclear factor of activated T-cells 5. Bone 53, 94-102.
Casali, C.I., Weber, K., Favale, N.O., and Tome, M.C. (2013). Environmental hyperosmolality
regulates phospholipid biosynthesis in the renal epithelial cell line MDCK. J Lipid Res 54, 677-
691.
Chang, C., and Harland, R.M. (2007). Neural induction requires continued suppression of both
Smad1 and Smad2 signals during gastrulation. Development 134, 3861-3872.
Chen, C.W., Liu, C.S., Chiu, I.M., Shen, S.C., Pan, H.C., Lee, K.H., Lin, S.Z., and Su, H.L. (2010).
The signals of FGFs on the neurogenesis of embryonic stem cells. J Biomed Sci 17, 33.
Chen, X., Wang, R., Liu, X., Wu, Y., Zhou, T., Yang, Y., Perez, A., Chen, Y.C., Hu, L.,
Chadarevian, J.P., et al. (2017). A Chemical-Genetic Approach Reveals the Distinct Roles of
GSK3alpha and GSK3beta in Regulating Embryonic Stem Cell Fate. Dev Cell 43, 563-576 e564.
Cho, A., Tang, Y., Davila, J., Deng, S., Chen, L., Miller, E., Wernig, M., and Graef, I.A. (2014).
Calcineurin signaling regulates neural induction through antagonizing the BMP pathway. Neuron
82, 109-124.
Chuang, J.H., Tung, L.C., and Lin, Y. (2015). Neural differentiation from embryonic stem cells in
vitro: An overview of the signaling pathways. World J Stem Cells 7, 437-447.
Collins, L.S., and Dorshkind, K. (1987). A stromal cell line from myeloid long-term bone marrow
cultures can support myelopoiesis and B lymphopoiesis. J Immunol 138, 1082-1087.
Couper, J.J., Littleford, K.D., Couper, R.T., Nakamura, T., and Ferrante, A. (1994). High glucose
and hyperosmolality stimulate hepatocyte growth factor secretion from cultured human mesangial
cells. Diabetologia 37, 533-535.
Daadi, M.M., Maag, A.L., and Steinberg, G.K. (2008). Adherent self-renewable human embryonic
stem cell-derived neural stem cell line: functional engraftment in experimental stroke model. PLoS
One 3, e1644.
Delaune, E., Lemaire, P., and Kodjabachian, L. (2005). Neural induction in Xenopus requires early
FGF signalling in addition to BMP inhibition. Development 132, 299-310.
147
Discher, D.E., Mooney, D.J., and Zandstra, P.W. (2009). Growth factors, matrices, and forces
combine and control stem cells. Science 324, 1673-1677.
Du, J., Chen, X., Liang, X., Zhang, G., Xu, J., He, L., Zhan, Q., Feng, X.Q., Chien, S., and Yang,
C. (2011). Integrin activation and internalization on soft ECM as a mechanism of induction of stem
cell differentiation by ECM elasticity. Proc Natl Acad Sci U S A 108, 9466-9471.
Engler, A.J., Sen, S., Sweeney, H.L., and Discher, D.E. (2006). Matrix elasticity directs stem cell
lineage specification. Cell 126, 677-689.
Fang, Y., Liu, Z., Chen, Z., Xu, X., Xiao, M., Yu, Y., Zhang, Y., Zhang, X., Du, Y., Jiang, C., et
al. (2017). Smad5 acts as an intracellular pH messenger and maintains bioenergetic homeostasis.
Cell Res 27, 1083-1099.
Fei, T., Xia, K., Li, Z., Zhou, B., Zhu, S., Chen, H., Zhang, J., Chen, Z., Xiao, H., Han, J.D., et al.
(2010). Genome-wide mapping of SMAD target genes reveals the role of BMP signaling in
embryonic stem cell fate determination. Genome Res 20, 36-44.
Finley, M.F., Devata, S., and Huettner, J.E. (1999). BMP-4 inhibits neural differentiation of
murine embryonic stem cells. J Neurobiol 40, 271-287.
Fisher, K.R., Fedoroff, R.S., and Wenger, E.L. (1975). Effect of osmotic pressure on neurogenesis
in cultures of chich embryo spinal cords. In Vitro 11, 329-337.
Fraichard, A., Chassande, O., Bilbaut, G., Dehay, C., Savatier, P., and Samarut, J. (1995). In vitro
differentiation of embryonic stem cells into glial cells and functional neurons. J Cell Sci 108 ( Pt
10), 3181-3188.
Goransson, V., Hansell, P., Moss, S., Alcorn, D., Johnsson, C., Hallgren, R., and Maric, C. (2001).
Renomedullary interstitial cells in culture; the osmolality and oxygen tension influence the
extracellular amounts of hyaluronan and cellular expression of CD44. Matrix Biol 20, 129-136.
Guan, K., Chang, H., Rolletschek, A., and Wobus, A.M. (2001). Embryonic stem cell-derived
neurogenesis. Retinoic acid induction and lineage selection of neuronal cells. Cell Tissue Res 305,
171-176.
Haegele, L., Ingold, B., Naumann, H., Tabatabai, G., Ledermann, B., and Brandner, S. (2003).
Wnt signalling inhibits neural differentiation of embryonic stem cells by controlling bone
morphogenetic protein expression. Mol Cell Neurosci 24, 696-708.
Heeg-Truesdell, E., and LaBonne, C. (2006). Neural induction in Xenopus requires inhibition of
Wnt-beta-catenin signaling. Dev Biol 298, 71-86.
Hemmati-Brivanlou, A., Kelly, O.G., and Melton, D.A. (1994). Follistatin, an antagonist of
activin, is expressed in the Spemann organizer and displays direct neuralizing activity. Cell 77,
283-295.
148
Hemmati-Brivanlou, A., and Melton, D. (1997). Vertebrate embryonic cells will become nerve
cells unless told otherwise. Cell 88, 13-17.
Hemmati-Brivanlou, A., and Melton, D.A. (1994). Inhibition of activin receptor signaling
promotes neuralization in Xenopus. Cell 77, 273-281.
Ingber, D.E. (2006). Mechanical control of tissue morphogenesis during embryological
development. Int J Dev Biol 50, 255-266.
Itoh, K., Tezuka, H., Sakoda, H., Konno, M., Nagata, K., Uchiyama, T., Uchino, H., and Mori,
K.J. (1989). Reproducible establishment of hemopoietic supportive stromal cell lines from murine
bone marrow. Exp Hematol 17, 145-153.
Kawasaki, H., Mizuseki, K., Nishikawa, S., Kaneko, S., Kuwana, Y., Nakanishi, S., Nishikawa,
S.I., and Sasai, Y. (2000). Induction of midbrain dopaminergic neurons from ES cells by stromal
cell-derived inducing activity. Neuron 28, 31-40.
Keung, A.J., Asuri, P., Kumar, S., and Schaffer, D.V. (2012). Soft microenvironments promote
the early neurogenic differentiation but not self-renewal of human pluripotent stem cells. Integr
Biol (Camb) 4, 1049-1058.
Kim, N., Minami, N., Yamada, M., and Imai, H. (2017). Immobilized pH in culture reveals an
optimal condition for somatic cell reprogramming and differentiation of pluripotent stem cells.
Reprod Med Biol 16, 58-66.
Kintner, C., & Koyano-Nakagawa, N. (2013). Neurogenesis in the Vertebrate Embryo. In R. Lanza
& A. Atala (2nd Ed.), Handbook of Stem Cells (pp. 665-680). Waltham, MA: Academic Press.
Klingensmith, J., Ang, S.L., Bachiller, D., and Rossant, J. (1999). Neural induction and patterning
in the mouse in the absence of the node and its derivatives. Dev Biol 216, 535-549.
Kong, X.B., and Zhang, C. (2009). Dickkopf (Dkk) 1 promotes the differentiation of mouse
embryonic stem cells toward neuroectoderm. In Vitro Cell Dev Biol Anim 45, 185-193.
Kothapalli, C.R., and Kamm, R.D. (2013). 3D matrix microenvironment for targeted
differentiation of embryonic stem cells into neural and glial lineages. Biomaterials 34, 5995-6007.
Kuan, C.Y., Yang, D.D., Samanta Roy, D.R., Davis, R.J., Rakic, P., and Flavell, R.A. (1999). The
Jnk1 and Jnk2 protein kinases are required for regional specific apoptosis during early brain
development. Neuron 22, 667-676.
Kunath, T., Saba-El-Leil, M.K., Almousailleakh, M., Wray, J., Meloche, S., and Smith, A. (2007).
FGF stimulation of the Erk1/2 signalling cascade triggers transition of pluripotent embryonic stem
cells from self-renewal to lineage commitment. Development 134, 2895-2902.
Kuriyama, S., and Mayor, R. (2009). A role for Syndecan-4 in neural induction involving ERK-
and PKC-dependent pathways. Development 136, 575-584.
149
Kuroda, H., Fuentealba, L., Ikeda, A., Reversade, B., and De Robertis, E.M. (2005). Default neural
induction: neuralization of dissociated Xenopus cells is mediated by Ras/MAPK activation. Genes
Dev 19, 1022-1027.
Kurosawa, H. (2007). Methods for inducing embryoid body formation: in vitro differentiation
system of embryonic stem cells. J Biosci Bioeng 103, 389-398.
Lamb, T.M., Knecht, A.K., Smith, W.C., Stachel, S.E., Economides, A.N., Stahl, N., Yancopolous,
G.D., and Harland, R.M. (1993). Neural induction by the secreted polypeptide noggin. Science
262, 713-718.
Lenhoff, H.M. (1991). Ethel Browne, Hans Spemann, and the Discovery of the Organizer
Phenomenon. Biol Bull 181, 72-80.
Macri-Pellizzeri, L., Pelacho, B., Sancho, A., Iglesias-Garcia, O., Simon-Yarza, A.M., Soriano-
Navarro, M., Gonzalez-Granero, S., Garcia-Verdugo, J.M., De-Juan-Pardo, E.M., and Prosper, F.
(2015). Substrate stiffness and composition specifically direct differentiation of induced
pluripotent stem cells. Tissue Eng Part A 21, 1633-1641.
Marchal, L., Luxardi, G., Thome, V., and Kodjabachian, L. (2009). BMP inhibition initiates neural
induction via FGF signaling and Zic genes. Proc Natl Acad Sci U S A 106, 17437-17442.
Martinez Arias, A., and Steventon, B. (2018). On the nature and function of organizers.
Development 145.
Martyn, I., Kanno, T.Y., Ruzo, A., Siggia, E.D., and Brivanlou, A.H. (2018). Self-organization of
a human organizer by combined Wnt and Nodal signalling. Nature 558, 132-135.
Matzuk, M.M., Lu, N., Vogel, H., Sellheyer, K., Roop, D.R., and Bradley, A. (1995). Multiple
defects and perinatal death in mice deficient in follistatin. Nature 374, 360-363.
McMahon, J.A., Takada, S., Zimmerman, L.B., Fan, C.M., Harland, R.M., and McMahon, A.P.
(1998). Noggin-mediated antagonism of BMP signaling is required for growth and patterning of
the neural tube and somite. Genes Dev 12, 1438-1452.
Mondragon-Teran, P., Lye, G.J., and Veraitch, F.S. (2009). Lowering oxygen tension enhances
the differentiation of mouse embryonic stem cells into neuronal cells. Biotechnol Prog 25, 1480-
1488.
Moon, S.H., Ju, J., Park, S.J., Bae, D., Chung, H.M., and Lee, S.H. (2014). Optimizing human
embryonic stem cells differentiation efficiency by screening size-tunable homogenous embryoid
bodies. Biomaterials 35, 5987-5997.
Muller-Sieburg, C.E., and Deryugina, E. (1995). The stromal cells' guide to the stem cell universe.
Stem Cells 13, 477-486.
150
Ng, E.S., Davis, R.P., Azzola, L., Stanley, E.G., and Elefanty, A.G. (2005). Forced aggregation of
defined numbers of human embryonic stem cells into embryoid bodies fosters robust, reproducible
hematopoietic differentiation. Blood 106, 1601-1603.
Nieto, M.A. (2001). The early steps of neural crest development. Mech Dev 105, 27-35.
Nishimoto, S., Kusakabe, M., and Nishida, E. (2005). Requirement of the MEK5-ERK5 pathway
for neural differentiation in Xenopus embryonic development. EMBO Rep 6, 1064-1069.
Okabe, S., Forsberg-Nilsson, K., Spiro, A.C., Segal, M., and McKay, R.D. (1996). Development
of neuronal precursor cells and functional postmitotic neurons from embryonic stem cells in vitro.
Mech Dev 59, 89-102.
Okada, S., Nakauchi, H., Nagayoshi, K., Nishikawa, S., Miura, Y., and Suda, T. (1992). In vivo
and in vitro stem cell function of c-kit- and Sca-1-positive murine hematopoietic cells. Blood 80,
3044-3050.
Okada, Y., Shimazaki, T., Sobue, G., and Okano, H. (2004). Retinoic-acid-concentration-
dependent acquisition of neural cell identity during in vitro differentiation of mouse embryonic
stem cells. Dev Biol 275, 124-142.
Otero, J.J., Fu, W., Kan, L., Cuadra, A.E., and Kessler, J.A. (2004). Beta-catenin signaling is
required for neural differentiation of embryonic stem cells. Development 131, 3545-3557.
Ozair, M.Z., Kintner, C., and Brivanlou, A.H. (2013). Neural induction and early patterning in
vertebrates. Wiley Interdiscip Rev Dev Biol 2, 479-498.
Pera, E.M., Ikeda, A., Eivers, E., and De Robertis, E.M. (2003). Integration of IGF, FGF, and anti-
BMP signals via Smad1 phosphorylation in neural induction. Genes Dev 17, 3023-3028.
Plachta, N., Bibel, M., Tucker, K.L., and Barde, Y.A. (2004). Developmental potential of defined
neural progenitors derived from mouse embryonic stem cells. Development 131, 5449-5456.
Plotnikov, A., Zehorai, E., Procaccia, S., and Seger, R. (2011). The MAPK cascades: signaling
components, nuclear roles and mechanisms of nuclear translocation. Biochim Biophys Acta 1813,
1619-1633.
Psychoyos, D., and Stern, C.D. (1996). Restoration of the organizer after radical ablation of
Hensen's node and the anterior primitive streak in the chick embryo. Development 122, 3263-
3273.
Rathjen, J., Haines, B.P., Hudson, K.M., Nesci, A., Dunn, S., and Rathjen, P.D. (2002). Directed
differentiation of pluripotent cells to neural lineages: homogeneous formation and differentiation
of a neurectoderm population. Development 129, 2649-2661.
Renoncourt, Y., Carroll, P., Filippi, P., Arce, V., and Alonso, S. (1998). Neurons derived in vitro
from ES cells express homeoproteins characteristic of motoneurons and interneurons. Mech Dev
79, 185-197.
151
Ross, S.A., McCaffery, P.J., Drager, U.C., and De Luca, L.M. (2000). Retinoids in embryonal
development. Physiol Rev 80, 1021-1054.
Sabapathy, K., Jochum, W., Hochedlinger, K., Chang, L., Karin, M., and Wagner, E.F. (1999).
Defective neural tube morphogenesis and altered apoptosis in the absence of both JNK1 and JNK2.
Mech Dev 89, 115-124.
Sasai, Y., Lu, B., Steinbeisser, H., Geissert, D., Gont, L.K., and De Robertis, E.M. (1994).
Xenopus chordin: a novel dorsalizing factor activated by organizer-specific homeobox genes. Cell
79, 779-790.
Sato, N., Meijer, L., Skaltsounis, L., Greengard, P., and Brivanlou, A.H. (2004). Maintenance of
pluripotency in human and mouse embryonic stem cells through activation of Wnt signaling by a
pharmacological GSK-3-specific inhibitor. Nat Med 10, 55-63.
Shan, Z.Y., Shen, J.L., Li, Q.M., Wang, Y., Huang, X.Y., Guo, T.Y., Liu, H.W., Lei, L., and Jin,
L.H. (2008). pCREB is involved in neural induction of mouse embryonic stem cells by RA. Anat
Rec (Hoboken) 291, 519-526.
Sheng, N., Xie, Z., Wang, C., Bai, G., Zhang, K., Zhu, Q., Song, J., Guillemot, F., Chen, Y.G.,
Lin, A., et al. (2010). Retinoic acid regulates bone morphogenic protein signal duration by
promoting the degradation of phosphorylated Smad1. Proc Natl Acad Sci U S A 107, 18886-
18891.
Shih, J., and Fraser, S.E. (1996). Characterizing the zebrafish organizer: microsurgical analysis at
the early-shield stage. Development 122, 1313-1322.
Slater, J.A., Zhou, S., Puscheck, E.E., and Rappolee, D.A. (2014). Stress-induced enzyme
activation primes murine embryonic stem cells to differentiate toward the first extraembryonic
lineage. Stem Cells Dev 23, 3049-3064.
Smukler, S.R., Runciman, S.B., Xu, S., and van der Kooy, D. (2006). Embryonic stem cells assume
a primitive neural stem cell fate in the absence of extrinsic influences. J Cell Biol 172, 79-90.
Snyder, E.Y. (2018). The state of the art in stem cell biology and regenerative medicine: the end
of the beginning. Pediatr Res 83, 191-204.
Sonntag, K.C., Simantov, R., Bjorklund, L., Cooper, O., Pruszak, J., Kowalke, F., Gilmartin, J.,
Ding, J., Hu, Y.P., Shen, M.M., et al. (2005). Context-dependent neuronal differentiation and germ
layer induction of Smad4-/- and Cripto-/- embryonic stem cells. Mol Cell Neurosci 28, 417-429.
Stern, C.D. (2005). Neural induction: old problem, new findings, yet more questions. Development
132, 2007-2021.
Stern, C.D. (2006). Neural induction: 10 years on since the 'default model'. Curr Opin Cell Biol
18, 692-697.
152
Streit, A., Berliner, A.J., Papanayotou, C., Sirulnik, A., and Stern, C.D. (2000). Initiation of neural
induction by FGF signalling before gastrulation. Nature 406, 74-78.
Sun, Y., Yong, K.M., Villa-Diaz, L.G., Zhang, X., Chen, W., Philson, R., Weng, S., Xu, H.,
Krebsbach, P.H., and Fu, J. (2014). Hippo/YAP-mediated rigidity-dependent motor neuron
differentiation of human pluripotent stem cells. Nat Mater 13, 599-604.
Tropepe, V., Hitoshi, S., Sirard, C., Mak, T.W., Rossant, J., and van der Kooy, D. (2001). Direct
neural fate specification from embryonic stem cells: a primitive mammalian neural stem cell stage
acquired through a default mechanism. Neuron 30, 65-78.
Verani, R., Cappuccio, I., Spinsanti, P., Gradini, R., Caruso, A., Magnotti, M.C., Motolese, M.,
Nicoletti, F., and Melchiorri, D. (2007). Expression of the Wnt inhibitor Dickkopf-1 is required
for the induction of neural markers in mouse embryonic stem cells differentiating in response to
retinoic acid. J Neurochem 100, 242-250.
Wang, R.N., Green, J., Wang, Z., Deng, Y., Qiao, M., Peabody, M., Zhang, Q., Ye, J., Yan, Z.,
Denduluri, S., et al. (2014). Bone Morphogenetic Protein (BMP) signaling in development and
human diseases. Genes Dis 1, 87-105.
Watanabe, K., Kamiya, D., Nishiyama, A., Katayama, T., Nozaki, S., Kawasaki, H., Watanabe,
Y., Mizuseki, K., and Sasai, Y. (2005). Directed differentiation of telencephalic precursors from
embryonic stem cells. Nat Neurosci 8, 288-296.
Waymouth, C. (1970). Osmolality of mammalian blood and of media for culture of mammalian
cells. In Vitro 6, 109-127.
Wilder, P.J., Kelly, D., Brigman, K., Peterson, C.L., Nowling, T., Gao, Q.S., McComb, R.D.,
Capecchi, M.R., and Rizzino, A. (1997). Inactivation of the FGF-4 gene in embryonic stem cells
alters the growth and/or the survival of their early differentiated progeny. Dev Biol 192, 614-629.
Williams, C.A., Fernandez-Alonso, R., Wang, J., Toth, R., Gray, N.S., and Findlay, G.M. (2016).
Erk5 Is a Key Regulator of Naive-Primed Transition and Embryonic Stem Cell Identity. Cell Rep
16, 1820-1828.
Wilson, S.I., Graziano, E., Harland, R., Jessell, T.M., and Edlund, T. (2000). An early requirement
for FGF signalling in the acquisition of neural cell fate in the chick embryo. Curr Biol 10, 421-
429.
Wilson, S.I., Rydstrom, A., Trimborn, T., Willert, K., Nusse, R., Jessell, T.M., and Edlund, T.
(2001). The status of Wnt signalling regulates neural and epidermal fates in the chick embryo.
Nature 411, 325-330.
Wray, J., Kalkan, T., Gomez-Lopez, S., Eckardt, D., Cook, A., Kemler, R., and Smith, A. (2011).
Inhibition of glycogen synthase kinase-3 alleviates Tcf3 repression of the pluripotency network
and increases embryonic stem cell resistance to differentiation. Nat Cell Biol 13, 838-845.
Xu, L. (2006). Regulation of Smad activities. Biochim Biophys Acta 1759, 503-513.
153
Xu, P., Yoshioka, K., Yoshimura, D., Tominaga, Y., Nishioka, T., Ito, M., and Nakabeppu, Y.
(2003). In vitro development of mouse embryonic stem cells lacking JNK/stress-activated protein
kinase-associated protein 1 (JSAP1) scaffold protein revealed its requirement during early
embryonic neurogenesis. J Biol Chem 278, 48422-48433.
Ying, Q.L., and Smith, A.G. (2003). Defined conditions for neural commitment and
differentiation. Methods Enzymol 365, 327-341.
Ying, Q.L., Stavridis, M., Griffiths, D., Li, M., and Smith, A. (2003). Conversion of embryonic
stem cells into neuroectodermal precursors in adherent monoculture. Nat Biotechnol 21, 183-186.
Zhang, P., Xing, K., Randazzo, J., Blessing, K., Lou, M.F., and Kador, P.F. (2012). Osmotic stress,
not aldose reductase activity, directly induces growth factors and MAPK signaling changes during
sugar cataract formation. Exp Eye Res 101, 36-43.
Zhou, X., Naguro, I., Ichijo, H., and Watanabe, K. (2016). Mitogen-activated protein kinases as
key players in osmotic stress signaling. Biochim Biophys Acta 1860, 2037-2052.
Zoldan, J., Karagiannis, E.D., Lee, C.Y., Anderson, D.G., Langer, R., and Levenberg, S. (2011).
The influence of scaffold elasticity on germ layer specification of human embryonic stem cells.
Biomaterials 32, 9612-9621.
Abstract (if available)
Abstract
Mouse embryonic stem cells (ESCs) are pluripotent stem cells derived from the inner cell mass of mouse blastocyst. Mouse ESCs can self-renew indefinitely while retaining the ability to differentiate into different cell types in the body. Mouse ESCs can be maintained and propagated in vitro by dual inhibition of GSK3 and MEK. MEK has two direct substrates, extracellular signal-regulated kinase 1 and 2 (ERK1/2), which are effector kinases of the MAPK/ERK pathway. However, the roles of ERK1/2 in the self-renewal and differentiation of mouse ESCs remain ill-defined. Here, we show that ERK1/2 are dispensable for ESC survival and self-renewal, and that chemical inhibition or genetic deletion of ERK1/2 is sufficient to maintain ESC self-renewal when GSK3 is simultaneously inhibited. Genetic deletion of either ERK1 or ERK2, however, is not sufficient to maintain ESC self-renewal in the presence of GSK3 inhibitor. Similarly, selective inhibition of individual ERK isoforms with a novel chemical genetic approach cannot maintain ESC self-renewal in the presence of GSK3 inhibitor. During ESC differentiation, inhibition of ERK1/2 blocks mesendoderm differentiation but not neuroectoderm differentiation. Selective inhibition of either ERK1 or ERK2, however, cannot block the lineage commitment of ESCs into different germ layers. In addition, selective inhibition of ERK1 but not ERK2 is able to promote ESC differentiation towards mesendoderm. These results suggest that the two ERK isoforms have both overlapping and non-overlapping roles in regulating ESC self-renewal and differentiation. This study opens new research avenues for dissecting the roles of individual ERK isoforms in different cell types and during various cellular processes.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Investigating the role of STAT3 in mouse and rat embryonic stem cell self-renewal and differentiation
PDF
Role of beta-catenin in mouse epiblast stem cell, embryonic stem cell self-renewal and differentiation
PDF
Role of STAT3 phosphorylation in mouse embryonic stem cell self-renewal and differentiation
PDF
Molecular basis of mouse epiblast stem cell and human embryonic stem cell self‐renewal
PDF
Pleotropic potential of Stat3 in determining self-renewal, apoptosis, and differentiation in mouse embryonic stem cells
PDF
Hepatic differentiation in human naïve stem cells compared to human embryonic stem cells
PDF
Roles of Klf4 in embryonic stem cells
PDF
Novel roles for Maf1 in embryonic stem cell differentiation and adipogenesis
PDF
Derivation and characterization of human embryonic stem (hES) cells and human induced pluripotent stem (hiPS) cells in clinical grade conditions
PDF
Dissecting the heterogeneity of mouse hematopoietic stem cells in vivo
PDF
Functional compensation between hematopoietic stem cell clones in vivo
PDF
Understanding human nephrogenesis and scaling synthesis of organoids facilitate modeling of kidney development and disease
PDF
Interaction of epigenetics and SMAD signaling in stem cells and diseases
PDF
Epigenetic plasticity of cultured female human embryonic stem cells and regulation of gene expression and chromatin by PR-SET7 mediated H4K20me1
PDF
Exploring stem cell pluripotency through long range chromosome interactions
PDF
Leucine-rich amelogenin peptide induces osteogenesis in mouse embryonic stem cells
PDF
Exploration of the roles of cancer stem cells and survivin in the pathogenesis and progression of prostate cancer
PDF
HLA upregulation via inhibition of the MEK1/2 pathway
PDF
The function of BS69 in mouse embryogenesis and embryonic stem cell differentiation
PDF
Harnessing the power of stem cell self-renewal pathways in cancer: dissecting the role of BMI-1 in Ewing’s sarcoma initiation and maintenance
Asset Metadata
Creator
Hu, Liang
(author)
Core Title
The role of ERK1/2 in mouse embryonic stem cell fate control
School
Keck School of Medicine
Degree
Doctor of Philosophy
Degree Program
Development, Stem Cells and Regenerative Medicine
Publication Date
04/19/2020
Defense Date
08/10/2018
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
differentiation,ERK1,ERK2,isoform-specific,mouse embryonic stem cell,OAI-PMH Harvest,self-renewal
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Ichida, Justin (
committee chair
), Li, Wei (
committee member
), Ying, Qi-Long (
committee member
)
Creator Email
huliang16@163.com,lianghu@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c89-89914
Unique identifier
UC11675841
Identifier
etd-HuLiang-6898.pdf (filename),usctheses-c89-89914 (legacy record id)
Legacy Identifier
etd-HuLiang-6898.pdf
Dmrecord
89914
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Hu, Liang
Type
texts
Source
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 a...
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
Tags
differentiation
ERK1
ERK2
isoform-specific
mouse embryonic stem cell
self-renewal