Content uploaded by Ori Avinoam
Author content
All content in this area was uploaded by Ori Avinoam on Dec 23, 2021
Content may be subject to copyright.
Article
ERK1/2 inhibition promotes robust myotube growth
via CaMKII activation resulting in myoblast-to-
myotube fusion
Graphical abstract
Highlights
dERK inhibition induces robust mouse and chicken myoblast
differentiation and fusion
dMyotubes fuse with mononucleated myoblasts at an
asymmetric fusogenic synapse
dERKi-driven signaling cascade leads to Ca
2+
-CaMKII-
dependent myoblasts to myotube fusion
dCaMKII is required for efficient muscle regeneration following
injury
Authors
Tamar Eigler, Giulia Zarfati,
Emmanuel Amzallag, ...,
Douglas P. Millay, Eldad Tzahor,
Ori Avinoam
Correspondence
eldad.tzahor@weizmann.ac.il (E.T.),
ori.avinoam@weizmann.ac.il (O.A.)
In brief
Eigler et al. show that an evolutionarily
conserved signaling cascade initiated by
ERK inhibition in myoblasts leads to
CaMKII-dependent fusion of
mononucleated myoblasts with early
myotubes at a fusogenic synapse.
Moreover, CaMKII is required for efficient
muscle regeneration following injury.
Mature myotube
ERKi ERK RXR
RYR1/3
CaMKII Ca
2+
Nascent myotube
Myoblasts
Eigler et al., 2021, Developmental Cell 56, 3349–3363
December 20, 2021 ª2021 The Author(s). Published by Elsevier Inc.
https://doi.org/10.1016/j.devcel.2021.11.022 ll
Article
ERK1/2 inhibition promotes robust
myotube growth via CaMKII activation
resulting in myoblast-to-myotube fusion
Tamar Eigler,
1
Giulia Zarfati,
2
Emmanuel Amzallag,
1
Sansrity Sinha,
2
Nadav Segev,
2
Yishaia Zabary,
3
Assaf Zaritsky,
3
Avraham Shakked,
1
Kfir-Baruch Umansky,
1
Eyal D. Schejter,
4
Douglas P. Millay,
5,6
Eldad Tzahor,
1,7,
*and Ori Avinoam
2,
*
1
Department of Molecular Cell Biology, Weizmann Institute of Science, Rehovot, Israel
2
Department of Biomolecular Sciences, Weizmann Institute of Science, Rehovot, Israel
3
Department of Software & Information Systems Engineering, Ben Gurion University, Be’er Sheva, Israel
4
Department of Molecular Genetics, Weizmann Institute of Science, Rehovot, Israel
5
Division of Molecular Cardiovascular Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA
6
Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, USA
7
Lead contact
*Correspondence: eldad.tzahor@weizmann.ac.il (E.T.), ori.avinoam@weizmann.ac.il (O.A.)
https://doi.org/10.1016/j.devcel.2021.11.022
SUMMARY
Myoblast fusion is essential for muscle development and regeneration. Yet, it remains poorly understood
how mononucleated myoblasts fuse with preexisting fibers. We demonstrate that ERK1/2 inhibition (ERKi)
induces robust differentiation and fusion of primary mouse myoblasts through a linear pathway involving
RXR, ryanodine receptors, and calcium-dependent activation of CaMKII in nascent myotubes. CaMKII acti-
vation results in myotube growth via fusion with mononucleated myoblasts at a fusogenic synapse. Mecha-
nistically, CaMKII interacts with and regulates MYMK and Rac1, and CaMKIId/gknockout mice exhibit
smaller regenerated myofibers following injury. In addition, the expression of a dominant negative CaMKII in-
hibits the formation of large multinucleated myotubes. Finally, we demonstrate the evolutionary conservation
of the pathway in chicken myoblasts. We conclude that ERK1/2 represses a signaling cascade leading to
CaMKII-mediated fusion of myoblasts to myotubes, providing an attractive target for the cultivated meat in-
dustry and regenerative medicine.
INTRODUCTION
During embryonic muscle development, myoblasts proliferate
and undergo terminal differentiation, a multistep process which
requires cell-cycle withdrawal, initiation of a muscle-specific
gene transcriptional program, differentiation into fusion-compe-
tent myoblasts, and ultimately cell-to-cell fusion to form nascent
multinucleated myotubes that mature to form contractile muscle
fibers (Chal and Pourquie
´, 2017;Dumont and Rudnicki, 2017;
Herna
´ndez-Herna
´ndez et al., 2017;Schmidt et al., 2019). This
process is recapitulated during muscle regeneration due to the
presence of satellite cells (SCs), the resident muscle stem cell.
(Chal and Pourquie
´, 2017;Dumont and Rudnicki, 2017;Hindi
et al., 2013). However, defining the molecular signaling pathways
that specifically regulate cell-to-cell fusion remain challenging
owing to the difficulty in distinguishing processes that regulate
fusion from those that regulate myogenic differentiation, which
will inevitably, although indirectly, affect fusion.
The study of Drosophila muscle development has highlighted
many facets of myoblast fusion, particularly the critical role of
cytoskeletal rearrangement and the formation of membrane
protrusions that extend from an ‘‘advancing’’ myoblast to a
‘‘receiving’’ myotube (Chen, 2011;Kim and Chen, 2019;Kim
et al., 2015;Lee and Chen, 2019;Lehka and Re
˛dowicz, 2020;
Schejter, 2016;Shilagardi et al., 2013). Drosophila muscle devel-
opment has been described as a two-phase process. The first
phase leads to the formation of founder cells, small nascent my-
otubes consisting of 2–3 nuclei (Beckett and Baylies, 2007;O
¨nel
and Renkawitz-Pohl, 2009;Rau et al., 2001). Founder cells
attract surrounding fusion-competent myoblasts and fuse with
them to form large multinucleated myotubes that mature into
muscle fibers (Abmayr and Pavlath, 2012;Chen and Olson,
2005;Herna
´ndez and Podbilewicz, 2017;Rochlin et al., 2010;
Schejter, 2016;Segal et al., 2016). Despite many conserved sim-
ilarities between Drosophila and vertebrate muscle fusion, differ-
ences do exist. For example, Myomaker (Mymk, a.k.a TMEM8c)
and Myomixer (Mymx, a.k.a GM7325, Myomerger, or Minion),
two muscle-specific proteins which were shown to be essential
and sufficient for myoblast fusion in vertebrates, are absent in in-
vertebrates (Leikina et al., 2018;Millay et al., 2016,2013,2014;
Quinn et al., 2017). Moreover, it is unclear whether the biphasic
phenomenon of myotube growth described in Drosophila is
conserved in vertebrate muscle fusion, and if so, whether the
processes that regulate myoblast-to-myoblast fusion (primary
Developmental Cell 56, 3349–3363, December 20, 2021 ª2021 The Author(s). Published by Elsevier Inc. 3349
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
ll
OPEN ACCESS
fusion) and myoblast-to-myotube fusion (secondary fusion) are
distinct.
The mitogen-activated protein kinases (MAPKs), including
p38, JNK, ERK1/2, and ERK5, mediate diverse signaling path-
ways, and are all implicated in muscle development and
myoblast differentiation (Alter et al., 2008;Knight and Kothary,
2011;Segale
´s et al., 2016;Xie et al., 2018). However, the role
of ERK1/2 in muscle fusion remains unclear and largely contra-
dictory (Bennett and Tonks, 1997;Dinev et al., 2001;Jones
et al., 2001;Sarbassov and Peterson, 1998;Sarbassov et al.,
1997;Shi et al., 2018;Sunadome et al., 2011;Tiffin et al.,
2004;Wu et al., 2000;Yang et al., 2006). ERK1/2 promotes
myoblast proliferation in response to various growth factors
(Campbell et al., 1995;Scata et al., 1999); inhibition of signaling
pathways leading to ERK1/2 activation or sequestering ERK1/2
in the cytoplasm results in cell-cycle exit and differentiation
(Jones et al., 2001;Michailovici et al., 2014;Sarbassov et al.,
1997;Tiffin et al., 2004;Wu et al., 2000). In cancer cell lines,
ERK1/2 phosphorylates the nuclear retinoid-X receptor (RXR),
leading to inhibition of its transactivation potential (Macoritto
et al., 2008;Matsushima-Nishiwaki et al., 2001), and RXR activity
in myoblasts promotes myogenesis through regulation of MyoD
expression and as a MYOG co-factor (Alric et al., 1998;
Froeschle
´et al., 1998;Khilji et al., 2020;Le May et al., 2011;
Zhu et al., 2009).
Calcium (Ca
2+
) has long been implicated as a regulator of
mammalian muscle fusion (Constantin et al., 1996;Shainberg
et al., 1969). Transient Ca
2+
depletion from the endoplasmic re-
ticulum (ER) is associated with myoblast differentiation and
fusion (Nakanishi et al., 2015). Moreover, the Ca
2+
-sensitive tran-
scription factor, NFATc2, was reported to mediate myoblast
recruitment and myotube expansion (Horsley et al., 2003). Yet
the signaling cascades which lead to Ca
2+
-mediated myoblast
fusion remain unclear. Intracellular Ca
2+
levels are regulated
through various Ca
2+
and voltage-gated channels, including
but not limited to ryanodine receptors (RYRs). RYRs are Ca
2+
channels expressed on the ER, which regulate Ca
2+
efflux into
the cytosol. RYRs were previously implicated in the regulation
of muscle terminal differentiation, but not myogenic commitment
in fetal myoblast differentiation (Pisaniello et al., 2003).
CaMKII is a member of the Ca
2+
/calmodulin (CaM)-dependent
serine/threonine kinase family. CaMKII delta (d) and gamma (g),
and to some extent beta (b), are the primary isoforms expressed
in skeletal muscle (Bayer et al., 1996). Upon Ca
2+
/CaM binding to
individual CaMKII subunits, cross-phosphorylation of neigh-
boring subunits at T287 leads to a state of autonomous
activation, by increasing the affinity for Ca
2+
/CaM several thou-
sand-fold. Previously, CaMKII was identified for its role in
Ca
2+
-dependent regulation of gene expression associated with
muscle-oxidative metabolism, as well as components of the
contractile machinery (Eilers, 2014a; Eilers, 2014b; Moradi,
2020; Ojuka, 2012; Richter and Hargreaves, 2013; Rose, 2007).
However, to date, the specific role of CaMKII in the regulation
of myoblast fusion has not been demonstrated.
By using the highly specific ERK1/2 inhibitor SCH772984
(Morris et al., 2013) in primary mouse and chick myoblast cul-
tures, we describe here the pleiotropic role of ERK1/2 during
myogenesis. First, in the inhibition of cell-cycle exit and initiation
of the myogenic transcriptional program, and second in the sup-
pression of a signaling cascade that culminates in CaMKII-
dependent regulation secondary myoblast-to-myotube fusion.
Moreover, we demonstrate a requirement for CaMKII during
muscle regeneration.
RESULTS
ERK1/2 inhibition (ERKi) induces myoblast
differentiation and hyperfusion in proliferation medium
Based on the recent findings by us and others, we hypothesized
that ERK1/2 prevents myogenesis not only through maintenance
of myoblast proliferation but also through the active repression
of pro-myogenic nuclear targets (Michailovici et al., 2014;
Yohe, 2018). In order to examine the role of ERK1/2 in myoblast
differentiation and fusion, early-passage mouse-derived primary
myoblasts were treated with the ERK1/2 inhibitor SCH772984
(ERKi, 1 mM) while in proliferation medium (PM). SCH772984 is
a highly selective, ATP-competitive inhibitor of both ERK1 and
ERK2. It acts by directly effecting ERK kinase activity and simul-
taneously inhibiting MEK-mediated phosphorylation of ERK
through allosteric mechanisms (Morris et al., 2013;Nissan
et al., 2013). ERKi resulted in the robust formation of myotubes
(Figures 1A and 1B; Video S1) as compared with conventional
serum-reduced differentiation medium (DM) (90.5% in ERKi
versus 11.6% in DM after 24 h). The differentiation and fusion
factors MyoD, MyoG,Mymk, and Mymx were upregulated
much earlier in cells treated with ERKi compared with DM alone
(Figure 1C). In addition, the fraction of MYOG
+
nuclei was signif-
icantly higher for ERKi compared with DM alone (Figures 1D and
1E). Moreover, immunofluorescence staining of ERKi cultures
with the proliferation markers KI-67 (Figures 1F and 1G) and
phosphorylated histone 3 (pH3) (Figures 1H and 1I) demon-
strated that myoblasts undergo cell-cycle arrest, consistent
with differentiation. ERKi also induced a similar effect on myo-
blasts cultured in DM (Figures S1A–S1C). Taken together, these
results show that ERKi induces a more robust differentiation and
fusion response in PM and in DM as compared with myoblasts
cultured in DM alone, suggesting that ERK1/2 acts as a
repressor of cell-cycle exit and initiation of the myogenic tran-
scriptional program.
Myotubes grow through recruitment of mononucleated
myoblasts at a fusogenic synapse
As we observed that myoblasts treated with ERKi exhibited a
more robust fusion phenotype compared with cells in conven-
tional DM, we wondered if ERKi was activating processes lead-
ing to increased myotube expansion through secondary fusion
of mononucleated myoblasts and myotubes, as previously
described in Drosophila (O
¨nel and Renkawitz-Pohl, 2009). To
explore this, we performed live-cell imaging of myoblasts ex-
pressing a membrane-targeted GFP and cytoplasmic DsRed
and calculated an hourly fusion index for a period of 8–23 h
post ERKi. We found that after the initial formation of bi- and tri-
nucleated cells, these cells accumulated nuclei and expanded
rapidly through several fusion events with mononucleated cells
(Figures 2A, 2B, and S2A; Videos S1 and S2).
The observed expansion of myotubes at the expense of mono-
nucleated cells is either a regulated phenomenon or a stochastic
process, wherein the larger multinucleated myotubes grow
ll
OPEN ACCESS Article
3350 Developmental Cell 56, 3349–3363, December 20, 2021
A
24h 48h
8h
MyHC Nuclei
MYOG KI-67
F
ERKi Ctrl
DM
PH3
DH
B
C
0
2
4
6**
**
KI-67+ nuclei (%)
0.0
0.5
1.0
1.5
2.0
2.5 *
**
pH3+ nuclei (%)
Ctrl ERKi DM
EGI
Hours post treatment
Fusion index
Ctrl
ERKi
DM
82448
0
25
50
75
100
ns
**** **** **** ****
0 8 12 24 48 72
0
5
10
15
20
Mymk
****
****
****
****
Fold expression
0 8 12 24 48 72
0
5
10
15
20
Mymx
****
**** ****
****
****
0 8 12 24 48 72
0
5
10
15
20
Myog
****
****
**** **** ****
0 8 12 24 48 72
0.0
0.5
1.0
1.5
2.0
Myod
****
**
Hours post treatment
Ctrl ERKi DM
MYOG+ nuclei (%)
0
20
40
60
80
100 **** ***
**
Ctrl ERKi DM
Ctrl
ERKi
DM
ERKi CtrlDM
Figure 1. ERK1/2 inhibition induces myoblast differentiation and hyperfusion in proliferation medium
(A) Representative immunofluorescent (IF) images of myoblasts at 8, 24, and 48 h after treatment with DMSO (Ctrl) or 1 mM SCH772984 (ERKi) in proliferation
medium (PM) or in differentiation medium (DM). Cells were stained with myosin heavy chain (MyHC, red), and the nuclear dye Hoechst (blue). Scale bar: 200 mm.
(B) Fusion index of (A) representing the percent of total nuclei found in MyHC
+
cells with two or more nuclei (total nuclei assayed, n = 88,518).
(C) Representative qRT-PCR results showing the temporal gene-expression profiles of Myod,Myog,Mymk, and Mymx, normalized toGapdh, during myogene sis.
Values are expressed as fold change from the control at 0 h.
(D, F, and H) Representative images of myoblasts treated with DMSO (Ctrl) or 1mM ERKi in PM or DM for 24 h and stained for MyHC (red), and MYOG (green) (D);
MyHC (red) and Ki-67 (green) (F); and MyHC (red) and pH3 (green) (H). Nuclei are stained with DAPI (blue). Scale bar: 100 mm.
(E, G, and I) Percentage of MYOG, Ki-67, and pH3 positive nuclei, respectively. All data are representative of at least 3 biological repeats. Error bars indi-
cate SEM.
ll
OPEN ACCESS
Article
Developmental Cell 56, 3349–3363, December 20, 2021 3351
A
C
D E
B
(legend on next page)
ll
OPEN ACCESS Article
3352 Developmental Cell 56, 3349–3363, December 20, 2021
rapidly because of their inherent higher probability to interact
and fuse with neighboring cells. To test this, we performed
data-driven simulations (Figures 2C, S2B, and S2C). We consid-
ered two scenarios, one where all cells have an equal probability
to fuse (random simulation) and one where the probability to fuse
was dependent on cell size (weighted simulation). However,
neither of the simulations recapitulated our results, implying
that myotube growth is not stochastic in nature (Figures 2C,
S2B, and S2C). Time-lapse microscopy also revealed that, start-
ing at 8 h after ERKi treatment, myoblasts begin to display
concerted collective movement and an increase in actin-rich
membrane protrusions (Videos S1 and S3). Moreover, it showed
that fusion occurs at a single location, where a protrusion ex-
tends from the advancing myoblast to the receiving myotube
(observed in 85% of fusion events; n = 46) (Figure 2D; Video S4).
As Ca
2+
has long been implicated in processes specifically
associated with myoblast fusion, we visualized Ca
2+
dynamics
during ERKi-induced myogenesis by imaging myoblasts har-
vested from GCaMP6 Ca
2+
reporter mice. We observed that a
pulse of Ca
2+
in nascent myotubes precedes the phase of rapid
myotube growth, suggesting that Ca
2+
released from the ER in
early myotubes may facilitate secondary fusion and myotube
expansion (Figure 2E; Video S5). Taken together, these results
suggest that myotube growth in mammals is initiated by the gen-
eration of multinucleated founder cells (2–3 nuclei) that expand
by fusion of ‘‘advancing’’ myoblasts to the ‘‘receiving’’ myotube,
and that this process might be regulated by cytosolic Ca
2+
.
ERK1/2 inhibition initiates an RXR/RYR-dependent
fusion response
To better understand the role of cytosolic Ca
2+
in secondary
fusion, we examined the gene expression of various Ca
2+
chan-
nels. Ryanodine receptors (RYR1-3) are channels that mediate
the release of Ca
2+
stores from the sarcoplasmic reticulum
(SR) into the cytoplasm during excitation-contraction coupling
in both cardiac and skeletal muscle cells. The expression of
Ryr1 and Ryr3, as well as Ca
2+
-sensing channels such as
SERCA1/2 (Atp2a1 and Atp2a2), Orai1/2, and STIM1/2 were up-
regulated in ERKi-treated myoblast cultures (Figure 3A). Co-
treatment of cultures with ERKi and the RYR-specific antagonist
dantrolene (50 mM, RYRi) reduced fusion by 60% (Figures 3B and
3C) without affecting differentiation, measured by the fraction of
MYOG
+
nuclei (Figures 3B and 3D). Along the same line, myo-
blasts co-treated with ERKi and the Ca
2+
chelator BAPTA-AM
(10 mM) exhibited reduced fusion by 81% (Figures 3B and 3E),
without affecting myogenic differentiation (Figures 3B, 3F, and
S3). Taken together, these results imply that elevated levels of
cytosolic Ca
2+
are essential for myoblast fusion.
As we previously reported, ERK1/2 nuclear localization re-
presses myogenic differentiation, while sequestration of ERK in
the cytoplasm promotes differentiation (Michailovici et al.,
2014). We thus hypothesized that ERK1/2 may repress differen-
tiation through the phosphoinhibition of a nuclear transcription
factor. As RXR regulates myogenesis and is also shown to un-
dergo phosphoinhibition at S260 by ERK1/2, we asked whether
RXR might be a nuclear ERK1/2 target in proliferating myoblasts
upstream of Ryr1 and Ryr3. Co-treatment of myoblasts with
ERKi and the specific RXR antagonist HX531 (20 mm, RXRi) re-
sulted in the downregulation of Ryr1 and Ryr3 mRNA expression
(Figure 3G). RXR immunoprecipitated with ERK1/2 in myoblasts
grown in proliferation conditions, and this interaction was atten-
uated upon treatment with ERKi (Figure 3H). Co-treatment with
RXRi similarly led to inhibition of fusion by 47% at 24 h after treat-
ment (Figures 3B and 3I), without affecting differentiation, as
measured by the fraction of MYOG
+
nuclei (Figures 3B and 3J).
Consistently, treatment with RXRi and RYRi generated a similar
reduction in fusion in cultures grown in DM (Figure S4). More-
over, time-course experiments demonstrated a reduction in
phosphorylated RXR within 15 min of administration of ERKi,
coinciding with the reduction of ERK1/2 phosphorylation (Fig-
ure 3K). These data imply that in proliferating myoblasts, RXR
is directly regulated by ERK1/2 and that upon ERK inhibition,
phosphoinhibition of RXR is relieved, leading to the transactiva-
tion of Ryr1 and Ryr3 expression, which likely promotes Ca
2+
release from the ER, resulting in myoblast fusion with the
growing myotube.
Myotube expansion requires calcium-dependent
CaMKII activation
Next, we wondered if Ca
2+
-dependent phosphorylation and acti-
vation of cellular kinases might be involved in regulating fiber
growth through secondary fusion. We found that the Ca
2+
-depen-
dent enzyme CaMKII was activated by phosphorylation at the
T287 residue upon treatment of myoblasts with ERKi in PM, as
well as following treatment in DM for 24 h (Figures 4AandS5A).
CaMKII activation begins at 12 h following ERKi treatment, coin-
ciding with the increase in total and phosphorylated RYR protein
levels and with the onset of myotube expansion by secondary
Figure 2. Myotubes grow through recruitment of mononucleated myoblasts at a fusogenic synapse
(A) Hourly fusion index showing the distribution of mono-, bi-, tri-, and multinucleated (n R4) cells. Total number of nuclei assayed, n = 13,044.
(B) Representative frames from time-lapse microscopy of an individual growing myotube (Video S2). At time 0 a binucleated myotube labeled with a cytoplasmic
DsRed (purple) is approached by a mononucleated myoblast (yellow square) expressing a membrane -targeted GFP (white). When the cells fuse, cytoplasmic and
membrane mixing become apparent (t = 00:28). Scale bar: 50 mm. Yellow and white squares mark the fusion events shown in (D).
(C) Experimental data compared with simulated data in two stochastic fusion scenarios: equal probability of cells to fuse irrespective of their number of nuclei (R4
nuclei), and weighted probability, which considered the possibility that the probability of a cell to add nuclei was proportional to the number of nuclei within it (see
STAR Methods for full details).
(D) Two examples of ‘‘fusogenic synapses’’ from the expanding fiber in (B) (time: hh:mm). Scale bar: 10 mm. Left column: Z-projection of the confocal stack. A
protrusion extending from the myoblast to the myotube where fusion eventually occurs as can be seen by the simultaneous diffusion of the cytoplasmic marker
into the myoblast and the disappearance of the membrane marker from the protrusion between the two fusing cells (Video S4). Middle and right columns: focal
planes from two events where the fusion pore can be seen expanding. Cyan and yellow arrows point to the fusogenic synapses before and after fusion.
(E) Representative frames acquired of GCaMP6S Ca
2+
reporter fluorescence in a growing myotube undergoing expansion via fusion (Video S5). Arrows indicate a
myoblast and a small myotube before fusion and the initiation of fiber growth. Dashed arrow indicates a myoblast prior to and during fusion. Asterisk indicates
burst in GCaMP6S fluorescence. Scale bar: 50 mm. Time in (B), (D), and (E) (hh:mm).
ll
OPEN ACCESS
Article
Developmental Cell 56, 3349–3363, December 20, 2021 3353
AC
E
G
H
K
FB
IJ
Figure 3. ERK1/2 inhibition initiates an RXR/RYR-dependent fusion response
(A) qRT-PCR analysis of fold change in expression of Ca
2+
channels and sensors in DMSO (Ctrl) compared with ERKi-treated cells at 24 h; expression was
normalized to Hprt.
(B) Representative IF images of cells treated with DMSO (Ctrl), 1 mM ERKi, 50 mM dantrolene (RYRi), ERKi, and RYRi, 10 mM BAPTA-AM, ERKi and BAPTA-AM,
20 mM HX531(RXRi), or ERKi and RXRi at 24 h. The differentiation markers MyHC (red), MYOG (green), and nuclei (blue) are shown. White boxes indicate the
region enlarged on the right.
(legend continued on next page)
ll
OPEN ACCESS Article
3354 Developmental Cell 56, 3349–3363, December 20, 2021
fusion (Figures 4B, 2A, and S6). Moreover, CaMKII activation
following ERKi is dependent on the upstream activity of RYR,
RXR, and Ca
2+
(Figures S5B–S5D, respectively). Strikingly, co-
treatment with the CaMKII inhibitor KN93 (5 mM; CaMKIIi) sup-
pressed the formation of polynucleated myotubes but maintained
bi- and trinucleated MyHC
+
cells, without affecting differentiation
(Figures 4C–4F, S5E, and S5F). Bi- and trinucleated myotubes
were still apparent even at higher concentrations of CaMKIIi,
which began to show toxicity at 10 mM(Figures S5G–S5I). In addi-
tion, the CaMKII inhibitor tat-CN21 (a phosphomimetic peptide)
gave a similar fusion suppression phenotype (Figure S5J). Co-
treatment of ERKi with CaMKIIi did not affect cell-cycle arrest,
as measured by pH3 staining (Figure S5K) or expression of the
cell-cycle inhibitors p21 and p27,comparedwithERKialone(Fig-
ure S5L), nor did it affect cell motility, demonstrating that fusion
failure is not due to an effect on cell-cycle arrest or cell migration
(Figure S5M; Video S6).
To further evaluate if the effect of CaMKII inhibition was spe-
cific to myotube growth through secondary fusion, we adminis-
tered CaMKIIi at 12 h following treatment with ERKi, coinciding
with the time point at which its activation was observed. Late
addition of CaMKIIi resulted in a phenotype not significantly
different from its addition at time 0, indicating that CaMKII inhibi-
tion has no effect before secondary fusion begins (Figure S5N).
CaMKIIi also had a similar effect on myoblasts cultured in DM
for 48 h, showing that the effect of CaMKIIi is not dependent
on ERKi (Figure S5O). These results suggest that CaMKII activa-
tion is essential for myoblast-to-myotube fusion but not for
myoblast-to-myoblast fusion. Therefore, in the presence of
CaMKIIi, bi- and trinucleated myotubes form but fail to expand
into large multinucleated fibers. Consistently, both RYR and
phosphorylated CaMKII are primarily localized to myotubes
rather than to mononucleated MyHC cells, following ERKi treat-
ment (Figures S6A, S6B, and 4G, respectively).
To examine whether CaMKII activation is sufficient to induce
myoblast-to-myotube fusion independent of treatment with
ERKi, primary myoblasts were transduced with either empty
adenovirus vector (Ad-Ctrl), wildtype CaMKII (Ad-CaMK2d
WT
),
or phospho-null CaMKII (Ad-CAMK2d
T287V
), and induced to
differentiate in DM. We found, as expected, that following treat-
ment in DM for 72 h, exogenous CAMK2d
WT
was activated by
phosphorylation, yet CAMK2d
T287V
failed to undergo activation
(Figure 4H). Importantly, we observed that while expression of
CAMK2d
WT
enhanced formation of bi- and polynucleated
MyHC+ cells, expression of CAMK2d
T287V
did not; it rather sup-
pressed growth of multinucleated cells compared with the con-
trol (Figure 4I). Taken together, the results suggest that CaMKII
activation is sufficient to promote secondary (myoblast-to-myo-
tube) fusion and implies a role for CaMKII function in myotubes.
CaMKII interacts with and regulates MYMK and Rac1
during fusion
The expression of both Mymk and Mymx was elevated upon
treatment with ERKi (Figure 1C); However, the increase in
Mymk expression, but not of Mymx, was partially suppressed
upon co-treatment with ERKi and CaMKIIi (Figure 4F). Therefore,
we examined whether reduced fusion upon CaMKII inhibition
could be attributed to decreased Mymk expression. To assess
this, we overexpressed MYMK by retroviral transduction in pri-
mary myoblasts and subjected them to treatment with ERKi
and CaMKIIi. We found that ERKi-dependent fusion was
enhanced upon overexpression of MYMK (Figures 5A and 5B).
However, this effect was completely dependent on CaMKII ac-
tivity as large myotubes were lost upon co-treatment with
CaMKIIi, while the accumulation of mono-, bi-, and trinucleated
cells was similar to that of cells transduced with control retrovirus
(Figures 5A–5C).
These data suggested that CaMKII may interact with and regu-
late MYMK activity. To test this, we used a proximity ligation
assay (PLA), which demonstrated that CaMKII and MYMK PLA
signal mean fluorescent intensity was increased by 2.9-fold
following ERK inhibition (Figures 5C, 5D, and S7). Moreover,
the PLA signal was exclusive in myotubes and not in mononucle-
ated cells, similar to the expression pattern for RYR and p-CaM-
KII (Figures 5D, 4G, S7A, and S7B, respectively). Due to the in-
crease in actin-rich protrusions observed upon ERKi (Video
S3), we briefly explored potential interactions of CaMKII with
the actin reassembly machinery. Interestingly, the RhoGTPase
Rac1, which is required for fusion, was predicted as an in silico
CaMKII target at the serine 71 residue (Wang et al., 2020).
Indeed, we show that increased Rac1 phosphorylation at S71
is dependent on CaMKII activity following ERK inhibition (Figures
5E, S7C, and S7D). Moreover, we demonstrated a significant in-
crease in the PLA signal between CaMKII and Rac1 following
ERK inhibition (Figures 5F and 5G). Taken together, these results
suggest that Ca
2+
-dependent CaMKII activation is a down-
stream event to the activation of RXR and RYR, and that CaMKII
activity is essential in myotubes for their expansion by mediating
myoblast-to-myotube fusion, likely through regulation of MYMK
and Rac1.
CaMKII function during muscle regeneration and ERK-
CaMKII pathway conservation
To examine the role of CaMKII during muscle regeneration
in vivo, wild-type (WT) mice were subjected to cardiotoxin
(CTX) induced injuries, and tissues were collected on the day
of injection, and consecutively on days 2–8 post injury. We
observed an acute activation of ERK1/2 two days post CTX
injury, likely associated with increased proliferation of myoblasts
(C and D) Fusion index and quantification of percent of MYOG
+
nuclei, respectively, for the ERKi and RYRi co-treatment experiment. Total number of nuclei
assayed, n = 113,448.
(E and F) Fusion index and quantification of percentage of MYOG
+
nuclei, respectively, for the ERKi and BAPTA-AM co-treatment experiment. Total number of
nuclei assayed, n = 109,360.
(G) qRT-PCR analysis of Ryr1/3 gene expression following co-treatment with ERKi and RXRi.
(H) Co-immunoprecipitation of ERK1/2 with RXR.
(I and J) Fusion index and quantification of percent of MYOG
+
nuclei, respectively, for the ERKi and RXRi co-treatment experiment. Total number of nuclei, n =
106,116.
(K) Representative western blot (WB) showing inhibition of ERK1/2 and reduction in phosphorylated RXR within 15 min post addition of ERKi. All data are
representative of 3 biological repeats. Error Bars indicate SEM. Scale bars: 100 mm.
ll
OPEN ACCESS
Article
Developmental Cell 56, 3349–3363, December 20, 2021 3355
A
E
GH
I
F
BDE
Figure 4. Myotube expansion requires calcium-dependent CaMKII activation
(A) Representative WB of CaMKII activation (T287 phosphorylation) at 24 h after treatment with ERKi or DM.
(B) Representative WB of time-course experiments showing RYR (S2844) and CaMKII (T287) activation following ERKi treatment.
(C) Representative IF images of cells treated with DMSO (Ctrl), 1 mM ERKi, 5 mM KN93 (CaMKIIi), or co-treated with ERKi and CaMKIIi at 24 h. Cells were stained
for the differentiation markers MyHC (red), MYOG (green), and DAPI (blue). Indicated regions are enlarged on the bottom.
(D) Fusion index for (C); values are stratified by number of nuclei per MyHC
+
fiber. Total number of nuclei assayed n = 61,510.
(E) Quantification of MYOG
+
nuclei per field of (C). Total number of nuclei assayed n = 112,901.
(F) qRT-PCR gene-expression analysis of the experiment shown in (E); gene expression was normalized to Hprt. Values are expressed as fold change from Ctrl.
(G) Representative IF images showing p-CaMKII localization (green) primarily to myotubes, at 24 h post treatment with ERKi. Indicated region in the ERKi image is
enlarged and divided into individual channels on the right. Arrows indicate mononucleated MyHC
+
cells, which are negative for p-CaMKII, while the asterisk
shows a binucleated MyHC
+
cell, which is p-CaMKII
+
. Arrowhead shows a MyHC
+
cell that has already fused with a myotube and is p-CaMKII
+
.
(H) Representative WB of infection experiments showing activation state of exogenous wild-type CaMKII (Ad-CaMKII
WT
) or a phospho-null mutant
(Ad-CaMKII
T287V
) expressed in myoblasts 72 h following treatment with DM. Bands for endogenous and exogenous CaMKII are indicated.
(I) Fusion index for the CaMKII infection study at 72 h treatment in DM, presented as fold change from control virus. Total number of nuclei assayed n = 18,758.
Error bars indicate SEM. Scale bars: 100 mm.
ll
OPEN ACCESS Article
3356 Developmental Cell 56, 3349–3363, December 20, 2021
(Figure 6A). By the third day post injury, levels of CaMKII
increased in regenerating muscle and remained elevated
throughout the 8 days examined; this was accompanied by a
peak in CaMKII activation at 5 days post injury (Figure 6A).
Following these promising results, we sought to examine the
requirement for CaMKII during muscle regeneration. To accom-
plish this, we generated a tamoxifen-inducible and SC-specific
conditional double knockout mouse of the CaMKII dand g
isoforms (Figures 6B and 6C).
In initial studies, we found that CaMKII protein levels in
quiescent SCs are highly stable and not efficiently reduced
even 3 months following tamoxifen administration. Moreover,
KO in SCs would unlikely alter CaMKII protein levels in the
mature muscle fibers. To overcome this obstacle, we imple-
mented a repeat-injury model. We reasoned that the initial round
of regeneration would reduce the levels of the highly stable CaM-
KII protein in the SC pool and ultimately in the regenerated mus-
cle fibers, as the DNA content of the fusing KO myoblasts would
be integrated into the fibers. Pax7
CreERT/+
,CaMK2d
fl/fl
/g
fl/fl
(scDKO) or Pax7
+/+
, and CaMK2d
fl/fl
/g
fl/fl
(WT) 4-week-old mice
were given tamoxifen to induce Cre/Lox-based gene disruption.
When the mice were 12 weeks of age, CTX was administered,
and the mice were allowed to fully regenerate for 8 weeks. At
8 weeks post injury, mice were either sacrificed to harvest pri-
mary myoblasts from the injured leg (to assess function in vitro)
or subjected to a second CTX injury and sacrificed 14 days
post injury for histological analysis. Reduction in CaMKII levels
were indeed validated in scDKO myoblasts harvested 8 weeks
following the first injury (Figure 6D). A fusion index demonstrated
that such scDKO myoblasts exhibited a significant defect in
ERKi-induced secondary fusion compared with those isolated
from their WT littermates (Figures 6E and 6F). Specifically,
scDKO myoblasts exhibited a loss of the hyperfused myotubes
observed in the WT cultures and instead accumulated mononu-
cleated MyHC
+
cells and nascent myotubes (Figures 6E and 6F).
These results match and recapitulate the observations made on
myoblast cultures treated with CaMKIIi. Furthermore, scDKO
mice that received repeated injuries had significantly smaller
fiber cross-sectional area (851.4 mM
2
± 37.5) compared with their
WT counterparts (975 mM
2
± 25) (Figures 6G and 6H) and a trend
toward more centrally located nuclei (Figure 6I). Taken together,
the genetic loss of CaMK2d/gis sufficient to impair myoblast
fusion and muscle regeneration.
Finally, we tested whether this pathway is conserved beyond
mice. To this end, we treated primary chicken myoblasts with
ERKi in PM or with the conventional DM for 72 h. By 48 h, fusion
was highly elevated in the ERKi-treated cells as compared with
DM (fusion index = 64.6% and 8%, respectively) ( Figures 6J
and 6K). Moreover, ERKi treatment of proliferating chicken
A
B
C
F
D
E
G
Figure 5. CaMKII interacts with and regulates MYMK and Rac1
during fusion
(A) Representative IF images of myoblasts infected with control retrovi rus or
virus expressing Myomaker, and treated with DMSO (Ctrl), 1 mM ERKi, 5 mM
CaMKIIi, or co-treated with ERKi and CaMKIIi for 18 h.
(B) Stratified fusion index of (A).
(D) Representative images showing proximity ligation assay (PLA) between
CaMKII and MYMK for DMSO (Ctrl) or ERKi-treated myoblasts at 24 h post
treatment. Top panel shows the PLA signal (red), and bottom panel shows the
overlay of PLA signal (red), membrane marker (green), and nuclei (blue).
(E) Quantification of the PLA assay in (D) shown as the mean fluorescent in-
tensity normalized to nuclei number per field. (E) Representative WB analysis
of Rac1 S71 phosphorylation following treatment with ERKi and co-treatment
with CaMKIIi.
(F) Representative images showing results of the PLA between CaMKII and
Rac1 for DMSO (Ctrl) or ERKi-treated myoblasts at 24 h post treatment. Top
panel for each shows the PLA signal (red) and the bottom panel shows the
overlay of PLA signal (red), phalloidin (green), and nuclei (blue).
(G) Quantification of the results of the PLA assay, shown as the mean fluo-
rescent intensity normalized to nuclei number per field. All data are repre-
sentative of at least 3 biological repeats. Error bars indicate SEM. Scale
bars: 100 mm.
ll
OPEN ACCESS
Article
Developmental Cell 56, 3349–3363, December 20, 2021 3357
AD
E
F
B
C
G
J
HI
K
L
Figure 6. CaMKII function during muscle regeneration and ERK-CaMKII pathway conservation
(A) WB of analysis of indicated proteins from CTX-induced injured muscle. Line indicates where a lane was purposely removed.
(B) Schematic illustration of the SC-specific double CaMKII KO mouse model.
(C) Schematic illustration depicting the timeline of the repeat-injury experimental design.
(D) WB validation of CaMKII depletion in WT or scDKO primary myoblasts isolated for 2 weeks following initial injury.
(E) IF staining of WT or scDKO primary myoblasts following ERKi-induced fusion at 24 h post treatment. Insets are enlarged to the right.
(F) Fusion index comparison between WT (n = 4) and scDKO (n = 4) primary myoblasts stratified by number of nuclei per fiber. Total number of nuclei assayed,
n = 12,743.
(G) Representative field of WT and scDKO muscle 14 days after CTX-induce d reinjury.
(H) Quantification of myofiber cross-sectional areas of WT (n = 4) and scDKO (n = 4) mice 14 days following reinjury.
(I) Average percentage of central nuclei in WT (n = 4) and scDKO (n = 4) mice 14 days following reinjury. At least 9,000 fibers per mouse were measured for (H)
and (I).
(J) Representative IF staining of primary chicken myoblasts over 72 h of treatment either with ERKi in proliferation medium, or in conventional DM.
(K) Fusion index for the 48-h time point of (J).
(L) Representative WB analysis of CaMKII activation in chicken myoblasts, following treatment with ERKi or co-treatment with CaMKIIi. Error bars indicate SEM.
All scale bars, 100 mm.
ll
OPEN ACCESS Article
3358 Developmental Cell 56, 3349–3363, December 20, 2021
myoblasts results in activation of CaMKII (Figure 6L), showing
evolutionary conservation in at least two vertebrate lineages.
DISCUSSION
In this study, we demonstrate that ERK1/2 represses processes
leading to both differentiation and secondary fusion (Figure 7).
We show that ERKi induces robust differentiation and fusion within
24 h, without requiring low serum conditions. ERKi results in
reduced RXR phospho-inhibition and in the induction of RXR-
dependent RYR expression in nascent myotubes. Whether RXR
directly regulates RYR remains to be further explored. Ultimately,
RYR accumulation leads to Ca
2+
-dependent activation of CaMKII
in the myotube and to CaMKII-dependent myoblast-to-myotube
fusion likely via the interactions with MYMK and Rac1 (Figure 7).
In addition, we demonstrate a requirement for CaMKII in muscle
regeneration after injury.
The discovery of a signaling cascade that enhances myoblast
differentiation and fusion has implications for the ever-growing
Figure 7. Schematic representation of the
ERK1/2-CaMKII secondary fusion pathway
Schematic of the ERK-CaMKII signaling pathway
during myoblast differentiation and fusion: (1) In
proliferating myoblasts ERK1/2 suppresses MYOG
and p21/p27 activation. (2) Upon ERK1/2 inhibition,
p21/p27 are expressed and cells exit the cell cycle;
simultaneously, MYOG is upregulated and cells
differentiate. (3) During the differentiation process,
ERK1/2 inhibition results in reduced phosphoinhi-
bition of RXR leading to RYR1/3 upregulation and
accumulation in early myotubes. RYR activity pro-
motes in Ca
2+
-dependent CaMKII activation and
CaMKII-dependent myotube driven asymmetric
fusion, likely through CaMKII regulation of MYMK
and Rac1.
field of cultivated meat, which builds
upon the techniques used for decades of
culturing myoblasts (Choi et al., 2021;
Post et al., 2020). Here we show that the
ERK1/2-CaMKII pathway is conserved in
chicken myoblasts, suggesting that it
may be conserved in other vertebrates.
The cultivated meat industry is actively
seeking ways to increase production effi-
ciency in order to reach price-parity with
the current meat industry. Therefore, tak-
ing advantage of processes that can
speed up and enhance efficiency of
myoblast differentiation and fusion would
facilitate this goal.
The Ca
2+
channels RYR1 and RYR3 are
differentially expressed in skeletal muscle
during late development and in different
muscle types (Bertocchini et al., 1997;
Conti et al., 1996;Tarroni et al., 1997),
and mediate fetal myoblast myogenesis
in vitro (Pisaniello et al., 2003). Here, we
demonstrate that elevated expression of
Ryr1 and Ryr3 during myogenesis are dependent on the activity
of ERK and more directly downstream of RXR activity. The delay
in the upregulation in RYR protein levels compared with the inhi-
bition of RXR, which occurs within minutes of ERK inhibition,
may imply that the regulation of Ryr transcription via RXR activity
is indirect and that there is yet another intermediate regulator of
Ryr expression downstream of RXR. As RXR inhibition did not
change the number of MYOG positive nuclei, the upregulation
of RYR may be dependent on RXR-mediated regulation of
MYOD and MYOG function and not expression.
Ca
2+
has long been implicated in processes regulating
myoblast differentiation and fusion (Knudsen and Horwitz,
1977,1978;Shainberg et al., 1969). Upon ERK inhibition, the
activated form of RYR accumulates at the onset of myotube
growth. As we show that RYR activity is upstream to CaMKII
activation, RYR activation through S2844 phosphorylation is
likely regulated by yet another unidentified kinase. Accumulation
and phosphorylation of RYR in the cytoplasm of early myotubes
likely results in ‘‘leakage’’ of Ca
2+
from the ER, as previously
ll
OPEN ACCESS
Article
Developmental Cell 56, 3349–3363, December 20, 2021 3359
reported (Marx et al., 2000;Reiken et al., 2003). Using live imag-
ing of transgenic myoblasts, we observed an acute and persis-
tent increase in the GCaMP6 signal in early myotubes at the
onset of secondary fusion and myotube expansion consistent
with the activation of CaMKII at the onset of fiber growth.
The fact that ERK1/2 mediates signaling from growth factors
and their cognate receptors implies that fusion during muscle
development and regeneration is also regulated via long-distance
signaling, consistent with recent studies that demonstrated the
role of TGF-beta signaling in repressing myoblast fusion (Girardi
et al., 2021;Melendez et al., 2021). We show that there is an acute
activation of ERK following muscle injury. The direct signal that
mediates theactivation of ERK in muscle tissue postinjury remains
unclear. Fibroblast growth factors (FGFs) are potent regulators of
myoblast proliferation in vitro and in vivo, mediated through activa-
tion of ERK1/2 (Knight and Kothary, 2011;Pawlikowski et al.,
2017). FGF-6 was reported to be elevated in regenerating muscle
tissue, and theloss of FGF-6 results in a regeneration defect (Floss
et al., 1997), which worsens in FGF-2 andFGF-6 double KO (Neu-
haus et al., 2003). Therefore, transient upregulation of FGFs during
regeneration may facilitate myoblast proliferation and repression
of fusion throughERK activity in vivo, and their eventual downregu-
lation may lead to initiation of CaMKII- dependent fusion pro-
cesses following myoblast cell-cycle exit.
Our study provides direct evidence that fusion in mammalian
muscle occurs at a single membrane protrusion extending from
an ‘‘advancing’’ myoblast to a ‘‘receiving’’ myotube (Lipton and
Konigsberg, 1972;Shilagardi et al., 2013). Live imaging revealed
that nascent myotubes (2–3 nuclei) are evident as early as 12 h
post treatment with the ERK1/2 inhibitor. RYR upregulation and
activation, as well as Ca
2+
-dependent CaMKII activation also
occur after 12 h after treatment with ERKi, concurrent with a
concerted increase in myoblast-to-myotube fusion events, lead-
ing to rapid growth of the myotube. While the direct role of CaMKII
during secondary fusion is not fully understood, its activation pre-
cedes fusion and growth of the myotube by a short interval. This
temporal link is consistent with the putative interactions of CaMKII
with MYMK and Rac1, which are essential for the membrane and
cytoskeleton rearrangements needed for fusion (Millay et al.,
2013;Vasyutina et al., 2009). Consistently, elevated Rac1 serine
71 phosphorylation following ERK inhibition, a site previously
identified for switching the function of Rac1 from being prolamel-
lipodial to a more filopodial-promoting phenotype (Schwarz et al.,
2012), is dependent on CaMKII activation. Moreover, MYMK
activity in myotubes appears to depend on CaMKII activity. There-
fore, one possible role of CaMKII during fusion might be to regu-
late the preparation of the postsynapse on the receiving myotube
side through regulation of Myomaker and Rac1.
In summary, we have characterized a pleiotropic role for ERK
signaling in muscle biology in the direct and independent repres-
sion of cell-cycle exit, differentiation, and secondary fusion and
have identified CaMKII as a potent regulator of myoblast fusion
with myotubes. These findings and methodologicaladvancements
will surely have profoundand long-lastingimplications forthe fields
of muscle biology, regenerative medicine, and cultivated meat.
Limitations of the study
While we implicate CaMKII in regeneration in vivo, a limitation of
our study is that the inducible KO of CaMKIId/gisoforms was
performed in SCs rather than in myofibers. To compensate for
this, we knocked-down CaMKII in muscle fibers by adopting a
double injury model. After the first round of injury, nuclei bearing
CaMK2d/gKO DNA are incorporated into the regenerated mus-
cle, thus creating a myofiber CaMKII knockdown setting for the
next round of injury.
The in vitro experiments demonstrating the role of CaMKII dur-
ing fusion were carried out using a chemical inhibitor of CaMKII.
Therefore, we cannot rule out a possible role of CaMKII during
primary fusion, as it is possible that the chemical inhibitor did
not completely inhibit CaMKII activity. However, this is unlikely
given the immunofluorescence data showing the myotube spe-
cific localization of RYR and the interaction of CaMKII with
MYMK as evident by PLA. Taken together with the observation
that CaMKII activation is only evident upon formation of nascent
myotubes and not in the mononucleated myocytes, and similarly
that KO myoblasts are still able to fuse to form bi- and tri-
nucleated cells in vitro, we conclude that the observed effect
on regeneration is likely due to an impairment of CaMKII activity
in myofibers or de novo myotubes in vivo, which fail to fuse with
the existing myofiber.
STAR+METHODS
Detailed methods are provided in the online version of this paper
and include the following:
dKEY RESOURCES TABLE
dRESOURCE AVAILABILITY
BLead contact
BMaterial availability
BData and code availability
dEXPERIMENTAL MODEL AND SUBJECT DETAILS
BAnimal ethics statement
BIn vivo experimental animal models
BGenetic models for primary myoblast cultures and
isolation technique
dMETHOD DETAILS
BCTX induced injuries
BIn vitro fusion assays of primary myoblast cultures
BStatistical analysis
SUPPLEMENTAL INFORMATION
Supplemental information can be found online at https://doi.org/10.1016/j.
devcel.2021.11.022.
ACKNOWLEDGMENTS
CaMK2d
fl/fl
/g
fl/fl
mice were kindly provided by Eric Olson and Johannes Backs.
Histology sections were prepared by Calanit Raanan. This study was sup-
ported by grants to E.T. from the AFM (#21655), the Weizmann Institute Hellen
and Martin Kimmel Stem Cell grant, the European Research Council (ERC StG
#281289, ERC AdG #788194), the Israel Science Foundation (ISF), and
Minerva Foundation with funding from the Federal German Ministry for Educa -
tion and Research (to E.T. and O.A.). This project also received funding from
the European Research Council (ERC StG # 851080 to O.A.). O.A. also ac-
knowledges funding from the David Barton Center for Research on the Chem-
istry of Life and the Ruth and Herman Albert Scholarship Program for New Sci-
entists as well as the Estate of Fannie Sherr. O.A. is an incumbent of the Miriam
Berman Presidential Development Chair.
ll
OPEN ACCESS Article
3360 Developmental Cell 56, 3349–3363, December 20, 2021
AUTHOR CONTRIBUTIONS
T.E., E.T., and O.A. conceived and designed the experiments. T.E., with help
from G.Z., E.A., and S.S., carried out most of the experiments and analyzed
the data. Specifically, G.Z. performed and analyzed the live-cell imaging
data with assistance from N.S. and S.S. T.E., S.S., and E.A. performed qRT-
PCR experiments. T.E., G.Z., and S.S. performed and analyzed the experi-
ments comparing ERKi in PM and DM. T.E. and G.Z. performed immunohisto-
chemistry. Y.Z. and A.Z. performed the simulations. T.E. and K.U. performed
CTX injuries, and T.E. carried out all follow-up studies. E.S. contributed to
experimental design and critical review of the manuscript. D.M. contributed
to the design of the in vivo model and associated experiments. E.T. and O.A.
supervised the project. T.E, E.T., and O.A. wrote the manuscript with editing
contributions from all the authors.
DECLARATION OF INTERESTS
T.E, E.T., and O.A. hold a patent related to the scientific findings presented in
this manuscript and are the founders of ProFuse Technology. T.E. is the CTO
and E.T. and O.A. are the scientific advisors of ProFuse Technology.
Received: December 8, 2020
Revised: July 28, 2021
Accepted: November 21, 2021
Published: December 20, 2021
REFERENCES
Abmayr, S.M., and Pavlath, G.K. (2012). Myoblast fusion: lessons from flies
and mice. Development 139, 641–656.
Alric, S., Froeschle
´, A., Piquemal, D., Carnac, G., and Bonnieu, A. (1998).
Functional specificity of the two retinoic acid receptor RAR and RXR families
in myogenesis. Oncogene 16, 273–282.
Alter, J., Rozentzweig, D., and Bengal, E. (2008). Inhibition of myoblast differ-
entiation by tumor necrosis factor ais mediated by c-Jun N-terminal kinase 1
and leukemia inhibitory factor. J. Biol. Chem. 283, 23224–23234.
Avinoam, O., Fridman, K., Valansi, C., Abutbul, I., Zeev-Ben-Mordehai, T.,
Maurer, U.E., Sapir, A., Danino, D., Gr€
unewald, K., White, J.M., and
Podbilewicz, B. (2011). Conserved eukaryotic fusogens can fuse viral enve-
lopes to cells. Science 332, 589–592.
Bayer, K.U., Lo
¨hler, J., and Harbers, K. (1996). An alternative, nonkinase prod-
uct of the brain-specifically expressed Ca2+/calmodulin-dependent kinase II
alpha isoform gene in skeletal muscle. Mol. Cell. Biol. 16, 29–36.
Beckett, K., and Baylies, M.K. (2007). 3D analysis of founder cell and fusion
competent myoblast arrangements outlines a new model of myoblast fusion.
Dev. Biol. 309, 113–125.
Bennett, A.M., and Tonks, N.K. (1997). Regulation of distinct stages of skeletal
muscle differentiation by mitogen-activated protein kinases. Science 278,
1288–1291.
Bertocchini, F., Ovitt, C.E., Conti, A., Barone, V., Scho
¨ler, H.R., Bottinelli, R.,
Reggiani, C., and Sorrentino, V. (1997). Requirement for the ryanodine recep-
tor type 3 for efficient contraction in neonatal skeletal muscles. EMBO J 16,
6956–6963.
Campbell, J.S., Wenderoth, M.P., Hauschka, S.D., and Krebs, E.G. (1995).
Differential activation of mitogen-activated protein kinase in response to basic
fibroblast growth factor in skeletal muscle cells. Proc. Natl. Acad. Sci. USA 92,
870–874.
Chal, J., and Pourquie
´, O. (2017). Making muscle: skeletal myogenesis in vivo
and in vitro. Development 144, 2104–2122.
Chen, E.H. (2011). Invasive podosomes and myoblast fusion. Curr. Top.
Membr. 68, 235–258.
Chen, E.H., and Olson, E.N. (2005). Unveiling the mechanisms of cell-cell
fusion. Science 308, 369–373.
Choi, K.-H., Yoon, J.W., Kim, M., Lee, H.J., Jeong, J., Ryu, M., Jo, C., and Lee,
C.-K. (2021). Muscle stem cell isolation and in vitro culture for meat production:
A methodological review. Compr. Rev. Food Sci. Food Saf. 20, 429–457.
Constantin, B., Cognard, C., and Raymond, G. (1996). Myoblast fusion re-
quires cytosolic calcium elevation but not activation of voltage-dependent cal-
cium channels. Cell Calcium 19, 365–374.
Conti, A., Gorza, L., and Sorrentino, V. (1996). Differential distribution of ryano-
dine receptor type 3 (RyR3) gene product in mammalian skeletal muscles.
Biochem. J. 316, 19–23.
Desgeorges, T., Liot, S., Lyon, S., Bouvie
`re, J., Kemmel, A., Trignol, A.,
Rousseau, D., Chapuis, B., Gondin, J., Mounier, R., et al. (2019). Open-
CSAM, a new tool for semi-automated analysis of myofiber cross-sectional
area in regenerating adult skeletal muscle. Skelet. Muscle 9,2.
Dinev, D., Jordan, B.W., Neufeld, B., Lee, J.D., Lindemann, D., Rapp, U.R., and
Ludwig, S. (2001). Extracellular signal regulated kinase 5 (ERK5) is required for
the differentiation of muscle cells. EMBO Rep 2, 829–834.
Dumont, N.A., and Rudnicki, M.A. (2017). Characterizing satellite cells and
myogenic progenitors During skeletal muscle regeneration. Methods Mol.
Biol. 1560, 179–188.
Eilers, W., Gevers, W., van Overbeek, D., de Haan, A., Jaspers, R.T., Hilbers,
P.A., van Riel, N., and Fl€
uck, M. (2014a). CaMKII content affects contractile,
but not mitochondrial, characteristics in regenerating skeletal muscle. BMC
Physiol 14, 1215.
Eilers, W., Jaspers, R.T., de Haan, A., Ferrie
´, C., Valdivieso, P., and Fl€
uck, M.
(2014b). Muscle-type specific autophosphorylation of CaMKII isoforms after
paced contractions. Biomed Res Int 2014, 1–20.
Floss, T., Arnold, H.H., and Braun, T. (1997). A role for FGF-6 in skeletal muscle
regeneration. Genes Dev 11, 2040–2051.
Froeschle
´, A., Alric, S., Kitzmann, M., Carnac, G., Aurade
´, F., Rochette-Egly,
C., and Bonnieu, A. (1998). Retinoic acid receptors and muscle b-HLH pro-
teins: partners in retinoid-induced myogenesis. Oncogene 16, 3369–3378.
Girardi, F., Taleb, A., Ebrahimi, M., Datye, A., Gamage, D.G., Peccate, C.,
Giordani, L., Millay, D.P., Gilbert, P.M., Cadot, B., and Le Grand, F. (2021).
TGFbsignaling curbs cell fusion and muscle regeneration. Nat. Commun.
12, 750.
Harel, I., Nathan, E., Tirosh-Finkel, L., Zigdon, H., Guimara
˜es-Camboa, N.,
Evans, S.M., and Tzahor, E. (2009). Distinct origins and genetic programs of
head muscle satellite cells. Dev. Cell 16, 822–832.
Herna
´ndez, J.M., and Podbilewicz, B. (2017). The hallmarks of cell-cell fusion.
Development 144, 4481–4495.
Herna
´ndez-Herna
´ndez, J.M., Garcı
´a-Gonza
´lez, E.G., Brun, C.E., and
Rudnicki, M.A. (2017). The myogenic regulatory factors, determinants of mus-
cle development, cell identity and regeneration. Semin. Cell Dev. Biol.
72, 10–18.
Hindi, S.M., Tajrishi, M.M., and Kumar, A. (2013). Signaling mechanisms in
mammalian myoblast fusion. Sci. Signal. 6, re2.
Horsley, V., Jansen, K.M., Mills, S.T., and Pavlath, G.K. (2003). IL-4 acts as a
myoblast recruitment factor during mammalian muscle growth. Cell 113,
483–494.
Jones, N.C., Fedorov, Y.V., Rosenthal, R.S., and Olwin, B.B . (2001). ERK1/2 is
required for myoblast proliferation but is dispensable for muscle gene expres-
sion and cell fusion. J. Cell. Physiol. 186, 104–115.
Khilji, S., Hamed, M., Chen, J., and Li, Q. (2020). Dissecting myogenin-medi-
ated retinoid X receptor signaling in myogenic differentiation. Commun. Biol.
3, 315.
Kim, J.H., and Chen, E.H. (2019). The fusogenic synapse at a glance. J. Cell
Sci. 132, jcs213124.
Kim, J.H., Ren, Y., Ng, W.P., Li, S., Son, S., Kee, Y.S., Zhang, S., Zhang, G.,
Fletcher, D.A., Robinson, D.N., and Chen, E.H. (2015). Mechanical tension
drives cell membrane fusion. Dev. Cell 32, 561–573.
Knight, J.D.R., and Kothary, R. (2011). The myogenic kinome: protein kinases
critical to mammalian skeletal myogenesis. Skelet. Muscle 1,29.
Knudsen, K.A., and Horwitz, A.F. (1977). Tandem events in myoblast fusion.
Dev. Biol. 58, 328–338.
Knudsen, K.A., and Horwitz, A.F. (1978). Differential inhibition of myoblast
fusion. Dev. Biol. 66, 294–307.
ll
OPEN ACCESS
Article
Developmental Cell 56, 3349–3363, December 20, 2021 3361
Kreusser, M.M., Lehmann, L.H., Keranov, S., Hoting, M.-O., Oehl, U.,
Kohlhaas, M., Reil, J.-C., Neumann, K., Schneider, M.D., Hill, J.A., et al.
(2014). Cardiac CaM kinase II genes dand gcontribute to adverse remodeling
but redundantly inhibit calcineurin-induced myocardial hypertrophy.
Circulation 130, 1262–1273.
Le May, M., Mach, H., Lacroix, N., Hou, C., Chen, J., and Li, Q. (2011).
Contribution of retinoid X receptor signaling to the specification of skeletal
muscle lineage. J. Biol. Chem. 286, 26806–26812.
Lee, D.M., and Chen, E.H. (2019). Drosophila myoblast fusion: invasion and
resistance for the ultimate union. Annu. Rev. Genet. 53, 67–91.
Lehka, L., and Re
˛dowicz, M.J. (2020). Mechanisms regulating myoblast fusion:
a multilevel interplay. Semin. Cell Dev. Biol. 104, 81–92.
Leikina, E., Gamage, D.G., Prasad, V., Goykhberg, J., Crowe, M., Diao, J.,
Kozlov, M.M., Chernomordik, L.V., and Millay, D.P. (2018). Myomaker and
Myomerger work independently to control distinct steps of membrane remod-
eling during myoblast fusion. Dev. Cell 46, 767–780.e7.
Lipton, B.H., and Konigsberg, I.R. (1972). A fine-structural analysis of the
fusion of myogenic cells. J. Cell Biol. 53, 348–364.
Luo, J., Deng, Z.-L., Luo, X., Tang, N., Song, W.-X., Chen, J., Sharff, K.A., Luu,
H.H., Haydon, R.C., Kinzler, K.W., et al. (2007). A protocol for rapid generation
of recombinant adenoviruses using the AdEasy system. Nat. Protoc. 2,
1236–1247.
Macoritto, M., Nguyen-Yamamoto, L., Huang, D.C., Samuel, S., Yang, X.F.,
Wang, T.T., White, J.H., and Kremer, R. (2008). Phosphorylation of the human
retinoid X receptor alpha at serine 260 impairs coactivator(s) recruitment and
induces hormone resistance to multiple ligands. J. Biol. Chem. 283,
4943–4956.
Marx, S.O., Reiken, S., Hisamatsu, Y., and Jayaraman, T. (2000). PKA phos-
phorylation dissociates FKBP12. 6 from the calcium release channel (ryano-
dine receptor): defective regulation in failing hearts. Cell 101, 365–376.
Matsushima-Nishiwaki, R., Okuno, M., Adachi, S., Sano, T., Akita, K.,
Moriwaki, H., Friedman, S.L., and Kojima, S. (2001). Phosphorylation of reti-
noid X receptor alpha at serine 260 impairs its metabolism and function in hu-
man hepatocellular carcinoma. Cancer Res. 61, 7675–7682.
Melendez, J., Sieiro, D., Salgado, D., Morin, V., Dejardin, M.-J., Zhou, C.,
Mullen, A.C., and Marcelle, C. (2021). TGFbsignalling acts as a molecular
brake of myoblast fusion. Nat. Commun. 12, 749.
Michailovici, I., Harrington, H.A., Azogui, H.H., Yahalom-Ronen, Y., Plotnikov,
A., Ching, S., Stumpf, M.P.H., Klein, O.D., Seger, R., and Tzahor, E. (2014).
Nuclear to cytoplasmic shuttling of ERK promotes differentiation of muscle
stem/progenitor cells. Development 141, 2611–2620.
Millay, D.P., Gamage, D.G., Quinn, M.E., Min, Y.-L., Mitani, Y., Bassel-Duby,
R., and Olson, E.N. (2016). Structure-function analysis of myomaker domains
required for myoblast fusion. Proc. Natl. Acad. Sci. USA 113, 2116–2121.
Millay, D.P., O’Rourke, J.R., Sutherland, L.B., Bezprozvannaya, S., Shelton,
J.M., Bassel-Duby, R., and Olson, E.N. (2013). Myomaker is a membrane acti-
vator of myoblast fusion and muscle formation. Nature 499, 301–305.
Millay, D.P., Sutherland, L.B., Bassel-Duby, R., and Olson, E.N. (2014).
Myomaker is essential for muscle regeneration. Genes Dev. 28, 1641–1646.
Moradi, F., Copeland, E.N., Baranowski, R.W., Scholey, A.E., Stuart, J.A., and
Fajardo V, A. (2020). Calmodulin-binding proteins in muscle: a minireview on
nuclear receptor interacting protein, neurogranin, and growth-associated pro-
tein 43. Int J Mol Sci. 21, 1016.
Morris, E.J., Jha, S., Restaino, C.R., Dayananth, P., Zhu, H., Cooper, A., Carr,
D., Deng, Y., Jin, W., Black, S., et al. (2013). Discovery of a novel ERK inhibitor
with activity in models of acquired resistance to BRAF and MEK inhibitors.
Cancer Discov. 3, 742–750.
Murphy, M.M., Lawson, J.A., Mathew, S.J., Hutcheson, D.A., and Kardon, G.
(2011). Satellite cells, connective tissue fibroblasts and their interactions are
crucial for muscle regeneration. Development 138, 3625–3637.
Nakanishi, K., Kakiguchi, K., Yonemura, S., Nakano, A., and Morishima, N.
(2015). Transient Ca2+ depletion from the endoplasmic reticulum is critical
for skeletal myoblast differentiation. FASEB J. 29, 2137–2149.
Neuhaus, P., Oustanina, S., Loch, T., Kr€
uger, M., Bober, E., Dono, R., Zeller,
R., and Braun, T. (2003). Reduced mobility of fibroblast growth factor (FGF)-
deficient myoblasts might contribute to dystrophic changes in the musculature
of FGF2/FGF6/mdx triple-mutant mice. Mol. Cell. Biol. 23, 6037–6048.
Nissan, M.H., Rosen, N., and Solit, D.B. (2013). ERK pathway inhibitors: how
low should we go? Cancer Discov. 3, 719–721.
Ojuka, E.O., Goyaram, V., and Smith J, A. (2012). The role of CaMKII in regu-
lating GLUT4 expression in skeletal muscle. Am J Physiol Endocrinol Metab
303, E3221–E331.
O
¨nel, S.-F., and Renkawitz-Pohl, R. (2009). FuRMAS: triggering myoblast
fusion in Drosophila. Dev. Dyn. 238, 1513–1525.
Pawlikowski, B., Vogler, T.O., Gadek, K., and Olwin, B.B. (2017). Regulation of
skeletal muscle stem cells by fibroblast growth factors. Dev. Dyn. 246,
359–367.
Pisaniello, A., Serra, C., Rossi, D., Vivarelli, E., Sorrentino, V., Molinaro, M., and
Bouche
´, M. (2003). The block of ryanodine receptors selectively inhibits fetal
myoblast differentiation. J. Cell Sci. 116, 1589–1597.
Post, M.J., Levenberg, S., Kaplan, D.L., Genovese, N., Fu, J., Bryant, C.J.,
Negowetti, N., Verzijden, K., and Moutsatsou, P. (2020). Scientific, sustainabil-
ity and regulatory challenges of cultured meat. Nat. Food 1, 403–415.
Quinn, M.E., Goh, Q., Kurosaka, M., Gamage, D.G., Petrany, M.J., Prasad, V.,
and Millay, D.P. (2017). Myomerger induces fusion of non-fusogenic cells and
is required for skeletal muscle development. Nat. Commun. 8, 15665.
Rau, A., Buttgereit, D., Holz, A., Fetter, R., Doberstein, S.K., Paululat, A.,
Staudt, N., Skeath, J., Michelson, A.M., and Renkawitz-Pohl, R. (2001). rolling
pebbles (rols) is required in Drosophila muscle precursors for recruitment of
myoblasts for fusion. Development 128, 5061–5073.
Reiken, S., Lacampagne, A., Zhou, H., Kherani, A., Lehnart, S.E., Ward, C.,
Huang, F., Gaburjakova, M., Gaburjakova, J., Rosemblit, N., et al. (2003).
PKA phosphorylation activates the calcium release channel (ryanodine recep-
tor) in skeletal muscle: defective regulation in heart failure. J. Cell Biol. 160,
919–928.
Richter, E.A., and Hargreaves, M. (2013). Exercise, GLUT4, and skeletal mus-
cle glucose uptake. Physiol Rev. 93, 999–1017.
Riedl, J., Crevenna, A.H., Kessenbrock, K., Yu, J.H., Neukirchen, D., Bista, M.,
Bradke, F., Jenne, D., Holak, T.A., Werb, Z., et al. (2008). Lifeact: a versatile
marker to visualize F-actin. Nat. Methods 5, 605–607.
Rochlin, K., Yu, S., Roy, S., and Baylies, M.K. (2010). Myoblast fusion: when it
takes more to make one. Dev. Biol. 341, 66–83.
Rose, A.J., Frøsig, C., Kiens, B., Wojtaszewski, J.F., and Richter E, A. (2007).
Effect of endurance exercise training on Ca2+–calmodulin-dependent protein
kinase II expression and signalling in skeletal muscle of humans. J Physiol 583,
785–795.
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.
Sarbassov, D.D., and Peterson, C.A. (1998). Insulin receptor substrate-1 and
phosphatidylinositol 3-kinase regulate extracellular signal-regulated kinase-
dependent and -independent signaling pathways during myogenic differentia-
tion. Mol. Endocrinol. 12, 1870–1878.
Scata, K.A., Bernard, D.W., Fox, J., and Swain, J.L. (1999). FGF rece ptor avail-
ability regulates skeletal myogenesis. Exp. Cell Res. 250, 10–21.
Schejter, E.D. (2016). Myoblast fusion: experimental systems and cellular
mechanisms. Semin. Cell Dev. Biol. 60, 112–120.
Schmidt, M., Sch€
uler, S.C., H€
uttner, S.S., von Eyss, B., and von Maltzahn, J.
(2019). Adult stem cells at work: regenerating skeletal muscle. Cell. Mol. Life
Sci. 76, 2559–2570.
Schneider, C.A., Rasband, W.S., and Eliceiri, K.W. (2012). NIH image to
ImageJ: 25 years of image analysis. Nat. Methods 9, 671–675.
Schwarz, J., Proff, J., H€
avemeier, A., Ladwein, M., Rottner, K., Barlag, B., Pich,
A., Tatge, H., Just, I., and Gerhard, R. (2012). Serine-71 phosphorylation of
Rac1 modulates downstream signaling. PLoS One 7, e44358.
ll
OPEN ACCESS Article
3362 Developmental Cell 56, 3349–3363, December 20, 2021
Segal, D., Dhanyasi, N., Schejter, E.D., and Shilo, B.-Z. (2016). Adhesion and
fusion of muscle cells are promoted by filopodia. Dev. Cell 38, 291–304.
Segale
´s, J., Perdiguero, E., and Mun
˜oz-Ca
´noves, P. (2016). Regulation of
muscle stem cell functions: a focus on the p38 MAPK signaling pathway.
Front. Cell Dev. Biol. 4,91.
Shainberg, A., Yagil, G., and Yaffe, D. (1969). Control of myogene sis in vitro by
Ca2+ concentration in nutrittonal medium. Exp. Cell Res. 58, 163–167.
Shi, C., Cai, Y., Li, Y., Li, Y., Hu, N., Ma, S., Hu, S., Zhu, P., Wang, W., and Zhou,
H. (2018). Yap promotes hepatocellular carcinoma metastasis and mobiliza-
tion via governing cofilin/F-actin/lamellipodium axis by regulation of JNK/
Bnip3/SERCA/CaMKII pathways. Redox Biol. 14, 59–71.
Shilagardi, K., Li, S., Luo, F., Marikar, F., Duan, R., Jin, P., Kim, J.H., Murnen,
K., and Chen, E.H. (2013). Actin-propelled invasive membrane protrusions
promote fusogenic protein engagement during cell-cell fusion. Science 340,
359–363.
Stringer, C., Wang, T., Michaelos, M., and Pachitariu, M. (2021). Cellpose: a
generalist algorithm for cellular segmentation. Nat. Methods 18, 100–106.
Sunadome, K., Yamamoto, T., Ebisuya, M., Kondoh, K., Sehara-Fujisawa, A.,
and Nishida, E. (2011). ERK5 regulates muscle cell fusion through Klf transcrip-
tion factors. Dev. Cell 20, 192–205.
Tarroni, P., Rossi, D., Conti, A., and Sorrentino, V. (1997). Expression of the
ryanodine receptor type 3 calcium release channel during development and
differentiation of mammalian skeletal muscle cells. J. Biol. Chem. 272,
19808–19813.
Tiffin, N., Adi, S., Stokoe, D., Wu, N.-Y., and Rosenthal, S.M. (2004). Akt phos-
phorylation is not sufficient for insulin-like growth factor-stimulated myogenin
expression but must be accompanied by down-regulation of mitogen-acti-
vated protein kinase/extracellular signal-regulated kinase phosphorylation.
Endocrinology 145, 4991–4996.
Vasyutina, E., Martarelli, B., Brakebusch, C., Wende, H., and Birchmeier, C.
(2009). The small G-proteins Rac1 and Cdc42 are essential for myoblast fusion
in the mouse. Proc. Natl. Acad. Sci. USA 106, 8935–8940.
Vest, R.S., Davies, K.D., O’Leary, H., Port, J.D., and Bayer, K.U. (2007). Dual
mechanism of a natural CaMKII inhibitor. Mol. Biol. Cell 18, 5024–5033.
Wang, C., Xu, H., Lin, S., Deng, W., Zhou, J., Zhang, Y., Shi, Y., Peng, D., and
Xue, Y. (2020). GPS 5.0: an update on the prediction of kinase-specific phos-
phorylation sites in proteins. Genomics Proteomics Bioinformatics 18, 72–80.
Wu, Z., Woodring, P.J., Bhakta, K.S., Tamura, K., Wen, F., Feramisco, J.R.,
Karin, M., Wang, J.Y., and Puri, P.L. (2000). p38 and extracellular signal-regu-
lated kinases regulate the myogenic program at multiple steps. Mol. Cell. Biol.
20, 3951–3964.
Xie, S.-J., Li, J.-H., Chen, H.-F., Tan, Y.-Y., Liu, S.-R., Zhang, Y., Xu, H., Yang,
J.-H., Liu, S., Zheng, L.-L., et al. (2018). Inhibition of the JNK/MAPK signaling
pathway by myogenesis-associated miRNAs is required for skeletal muscle
development. Cell Death Differ. 25, 1581–1597.
Yang, W., Chen, Y., Zhang, Y., Wang, X., Yang, N., and Zhu, D. (2006).
Extracellular signal-regulated kinase 1/2 mitogen-activated protein kinase
pathway is involved in myostatin-regulated differentiation repression. Cancer
Res. 66, 1320–1326.
Yohe, M.E., Gryder, B.E., Shern, J.F., Song, Y.K., Chou, H.C., Sindiri, S.,
Mendoza, A., Patidar, R., Zhang, X., Guha, R., et al. (2018). MEK inhibition in-
duces MYOG and remodels super-enhancers in RAS-driven rhabdomyosar-
coma. Sci Transl Med 10, 828–851.
Zhu, G.-H., Huang, J., Bi, Y., Su, Y., Tang, Y., He, B.-C., He, Y., Luo, J., Wang,
Y., Chen, L., et al. (2009). Activation of RXR and RAR signaling promotes
myogenic differentiation of myoblastic C2C12 cells. Differentiation 78,
195–204.
ll
OPEN ACCESS
Article
Developmental Cell 56, 3349–3363, December 20, 2021 3363
STAR+METHODS
KEY RESOURCES TABLE
REAGENT or RESOURCE SOURCE IDENTIFIER
Antibodies
Mouse monoclonal anti-MyHC (MF-20) DSHB Cat# MF 20; RRID:AB_2147781
Mouse monoclonal anti-MyHC (MY-32) Abcam Cat# ab51263; RRID:AB_2297993
Mouse monoclonal anti-MYOG Santa Cruz Biotechnology Cat# sc-13137; RRID:AB_627979
Rabbit polyclonal anti-PH3 Abcam Cat# ab4729;
RRID:AB_880448
Rabbit monoclonal anti-Ki-67 Cell Marque Cat#275R; RRID:AB_1158033
Mouse monoclonal anti-RYR Abcam Cat# ab2868; RRID:AB_2183051
Rabbit polyclonal anti-p-RYR Abcam Cat# ab59225; RRID:AB_946327
Rabbit polyclonal anti-p-CAMKII MERCK Cat# SAB4504356
Rabbit polyclonal anti-p-CAMKII Abcam Cat# ab182647
Rabbit monoclonal anti-CaMKII Abcam Cat# ab52476; RRID:AB_868641
Rabbit polyclonal anti-CaMKII Cell Signaling Cat# 3362; RRID:AB_2067938
Mouse monoclonal anti-CaMKII Santa Cruz Biotechnology Cat# sc-5306; RRID:AB_626788
Rabbit polyclonal anti-ERK1/2 MERCK Cat# M7927; RRID:AB_260665
Rabbit polyclonal anti-ERK1/2 MERCK Cat# M5670; RRID:AB_477216
Mouse monoclonal anti-p-ERK1/2 MERCK Cat# M9692; RRID:AB_260729
Mouse monoclonal anti-Rac1 Millipore Cat# 05-389; RRID:AB_309712)
Rabbit polyclonal anti-p-Rac1 Millipore Cat# 07-896-I; RRID:AB_612043
Mouse monoclonal anti-Vinculin Benny Geiger, Weizmann Institute
of Science
Rabbit monoclonal anti-GAPDH Abcam Cat# ab181602; RRID:AB_2630358
Rabbit polyclonal anti-RXRa Santa Cruz Biotechnology Cat# sc-553; RRID:AB_2184874
Rabbit polyclonal anti- p-RXR Affinity Biosciences Cat# AF8214; RRID:AB_2840276
Rabbit polyclonal anti-TMEM8C MERCK Cat# HPA051846
RRID:AB_2681636
Chemicals, peptides, and recombinant proteins
SCH772984 Cayman Chemicals Cat# 19166
HX-531 Cayman Chemicals Cat# 20762
Dantrolene Cayman Chemicals Cat# 14326
KN93 Cayman Chemicals Cat# 13319
Tat-scramble (Myr-YGRKKRRQRRRLSGPIIPRRD
GRKQRKEDVVK
Peptide 2.0
Tat-CN21 (Myr-YGRKKRRQRRRKRPPKL
GQIGRSKRVVIEDDR
Peptide 2.0
tamoxifen SIGMA Cat# T5648
Cardiotoxin (CTX) Lotaxan Cat# L8102
Critical commercial assays
Duolink Proximity Ligation Assay MERCK Cat# DUO92013
Cat# DUO92005; RRID:AB_2810942
Cat# DUO92001; RRID:AB_2810939
Deposited data
DOI: https://zenodo.org/badge/latestdoi/284677675
Experimental models: Organisms/strains
Mouse: B6.Cg-Pax7
tm1(cre/ERT2)Gaka
/J The Jackson laboratory 017763
Mouse: CaMK2D
flox
/G
flox
Eric Olson/Johannes Backs
(Continued on next page)
ll
OPEN ACCESS Article
e1 Developmental Cell 56, 3349–3363.e1–e6, December 20, 2021
RESOURCE AVAILABILITY
Lead contact
Request for reagents and data should be directed to and will be fulfilled by the lead contact, Eldad Tzahor (eldad.tzahor@weizmann.
ac.il)
Material availability
All plasmids generated in this study are available upon request to the lead contact.
Data and code availability
dAll data reported in this paper will be shared by the lead contact upon request.
dAll original code and related data has been deposited at https://github.com/assafZaritskyLab/MyocytesFusionSimulations
GeneratorAnalyzer#readme and is publicly available as of the date of publication. DOIs are listed in the key resources table.
dAny additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.
EXPERIMENTAL MODEL AND SUBJECT DETAILS
Animal ethics statement
All experiments were approved by the Animal Care and Use Committee of the Weizmann Institute of Science (IACUC application #
00720120-4 and 13780519-1). The study is compliant with all of the relevant ethical regulations regarding animal research. The mice
Continued
REAGENT or RESOURCE SOURCE IDENTIFIER
Mouse: B6J.Cg-Gt(ROSA)26Sor
tm96(CAG-GCaMP6s)Hze
/
MwarJ
The Jackson laboratory 028866
Mouse: ROSA26-tdTomato Weizmann Institute mouse repository
Mouse: nuclear reporter nTnG - B6N.129S6-Gt(ROSA)
26Sor
tm1(CAG-tdTomato*,-EGFP*)Ees
/J
The Jackson laboratory 023537
Mouse: membrane reporter mTmG - STOCK Gt(ROSA)
26Sor
tm4(ACTB-tdTomato,-EGFP)Luo
/J
The Jackson laboratory 007576
Mouse: LifeActGFP Weizmann Institute
mouse repository
Mouse: c57/bl6OlaHsd Envigo
Oligonucleotides
Primers for qRT-PCR (see Table S1) This paper
Primers for cloning (see Table S2) This paper
Recombinant DNA
RedTrack-CMV-EGFP-FLAG-CAMK2D
WT
(Ad-CaMK2D
WT
)
This paper
RedTrack-CMV-EGFP-FLAG-CAMK2D
T287V
(Ad-CaMK2D
T287V
)
This paper
RedTrack-CMV (Ad-Ctrl) Addgene Cat# 50957
pBabe-MYMK-CFPnls This paper
pBabe-CFPnls This paper
pBabe-dsRed plasmid This paper
Software and algorithms
Open-CSAM, semi-automated analysis
tool with ImageJ
ImageJ v1.52 software NIH RRID:SCR_003070
Image Lab software Bio-Rad RRID:SCR_014210
StepOne software Applied Biosystems RRID:SCR_014281
NIS-Elements imaging software ver.5.11.00 Nikon RRID:SCR_014329
VisiView software Visitron Systems GmbH
Cellpose software RRID:SCR_021716
ll
OPEN ACCESS
Article
Developmental Cell 56, 3349–3363.e1–e6, December 20, 2021 e2
were given ad libitum access to water and food and monitored daily for health and activity. Mice belonging to different experimental
groups were caged together in 12-h light/dark cycles and treated the same.
In vivo experimental animal models
To generate satellite cell specific and tamoxifen inducible CaMK2d/gdouble KO mice were, Pax7-Cre
ERT
mice (Murphy et al., 2011)
The Jackson laboratory, stock no. 017763) with double floxed CaMK2d
fl/fl
/g
fl/fl
mice (Kreusser et al., 2014). Female Pax7
CreERT/+
;
CaMK2d
fl/fl
/g
fl/fl
(scDKO) or Pax7
+/+
;CaMK2d
fl/fl
/g
fl/fl
(WT) littermates received intraperitoneal tamoxifen administration beginning
at weening (4 weeks of age) for 6 consecutive days, followed by weekly boosters until 12 weeks of age. Then these mice underwent
CTX induced injuries described below. 7-week-old Female Wildtype c57/bl6 mice were purchased from ENVIGO, and used to
evaluate ERK and CaMKII protein levels following CTX-induced injuries.
Genetic models for primary myoblast cultures and isolation technique
Nuclear and membrane reporter mice were bred inhouse by crossing nTnG
+/+
and mTmG
+/+
mice (The Jackson laboratory, stock no
023537, 007576 respectively). Actin/nuclear reporter mice were bred inhouse by crossing LifeAct-GFP mice (Riedl et al., 2008) with
nTnG
+/+
mice. Validation of of transgene expression was performed through examination of ear notch samples under a fluorescence
microscope. Ca
2+
reporter mice were bred inhouse by crossing Pax7-Cre
ERT+/+
(The Jackson laboratory, stock no. 017763) with
GCaMP6s
flstop/flstop
mice (The Jackson laboratory, stock no. and 028866), tdTomato reporter mice were bred inhouse by crossing
Pax7-Cre
ERT+/+
with tdTomato
flstop/flstop
. Genotyping was performed on every litter.
Primary mouse myoblasts were isolated from gastrocnemius muscle of female mice, or primary chicken myoblasts were isolated
from breast and leg muscles of post-mortem P1 chick breast and leg muscles. Briefly, muscle tissues were incubated in Trypsin B
(Biological Industries, Israel) and subjected to mechanical dissociation with a serological pipet. Supernatants were strained and
centrifuged. Pellets were resuspended in proliferation media and plated on 10% Matrigel-coated plates at 37and 5% CO
2
(Harel
et al., 2009). For all in vitro experiments, proliferation medium was Bio-Amf2 (Biological Industries, Israel) and Differentiation medium
(DM) was DMEM:F12 supplemented with 2% horse serum with 1% pen/strep mix. Myoblasts were maintained in proliferation media
until reaching approximately 80% confluency and then detached with Trypsin C (Biological Industries, Israel) and subjected to two
rounds of pre-plating on uncoated plates to reduce the number of fibroblasts, then seeded for specific experiments. Myoblast iso-
lations from Pax7-Cre
ERT+/+
;GCaMP6s
flstop/flstop
mice, and Pax7-Cre
ERT+/+
;tdTomato
flstop/flstop
mice were treated with 5uM of
Tamoxifen in culture for 24 hours immediately following harvesting, and fresh proliferation media with 5uM tamoxifen was replaced
after 24 hours. Then fresh media was replaced daily without tamoxifen. All in vitro experiments with primary myoblasts were done on
cells limited to the first and second passage.
METHOD DETAILS
CTX induced injuries
Mice were anesthetized with isoflurane and injected in the right gastrocnemius muscle with CTX dissolved in PBS at 10 sites (3ul per
site) at 10mm, using a Hamilton syringe. All injuries were performed on female mice. For mice that received a repeat injury: following
the first injury, mice were maintained for an additional 8 weeks and then injured again in the right gastrocnemius, as described above.
In vitro fusion assays of primary myoblast cultures
Primary myoblasts were plated at a density of 8x10
3
per well in 10% Matrigel-coated 96-well plates in proliferation medium for 24
hours. The following day, proliferation media was replaced either with proliferation media containing DMSO (Ctrl) or 1mM ERK1/2 in-
hibitor (ERKi; SCH772984, Cayman Chemicals), 20mM RXR antagonist (RXRi; HX-531, Cayman Chemicals), 50mM Ryanodine
receptor antagonist (RYRi; Dantrolene, Cayman Chemicals), 5mM CaMKII inhibitor (CaMKIIi; KN93, Cayman Chemicals), 50 mM
Tat-scramble, and 50 mM Tat- ((Vest et al., 2007) peptide 2.0), or with DM. Inhibitors were used at a the highest concentration before
becoming toxic, as determined by a dose response experiments.
Immunofluorescence staining
First passage primary myoblasts isolated from various strains (as indicated in figure legends) were plated in 96-well plates or cham-
ber slides and treated as described above. The cells were fixed with ice cold 4%PFA in PBS for 10 minutes, permeabilized with 0.5%
Triton X-100 in PBS for 6 minutes, and blocked in PBS with 0.025% Tween20, 10% normal horse serum and 10% normal goat serum
for 1 hour at room temperature. Primary antibody incubation was done in blocking buffer overnight at 4 degrees, with the following
antibodies: Myosin Heavy Chain (MyHC, MF20, DSHB hybridoma supernatant 1:10, or MY-32 ABCAM ab51263 1:400), Myogenin
(MYOG sc-13137 SCBT 1:200), pHistone 3 (PH3, ab47297 ABCAM 1:1000), Ki-67 (Cell Marque #275R), RYR (ab2868 ABCAM
1:100), and pCaMKII (MERCK SAB4504356 1:100). Cells were washed 3 times in PBS with 0.025% Tween20 and then incubated
with appropriate secondary antibodies in PBS for 1 hour. Where indicated, nuclei were either labeled with DAPI (MERCK D9542,
5ug/ml) or Hoechst 33342 (Thermo scientific #62249, 1:2000). Cells were imaged using the Nikon Eclipse Ti2 microscope (further
described in microscopy section). All analysis was performed on at least 1000 nuclei. For fixed cells following the time-course
with ERKi or DM (Figure 1B), images were captured with an inverted Olympus IX83 microscope (details in microscopy section).
All imaging analysis were performed on at least 1000 cells.
ll
OPEN ACCESS Article
e3 Developmental Cell 56, 3349–3363.e1–e6, December 20, 2021
Generation of retroviruses and transduction for live-cell imaging
The pBabe-puro and pBabe-GFPfarn plasmids were purchased from Addgene (Plasmid #1764 and #21836, respectively). The
pBABE-CFP-NLS plasmid was constructed by replacing the PuroR gene from the pBABE-Puro with the coding DNA sequence of
cyan fluorescent protein (CFP) fused to a tandem repeat of a nuclear localization signal (NLS) at the C-terminal (2 x PKKKRKV). For-
ward and Reverse DNA primers used for restriction free (RF) cloning of CFP-2xNLS from pCDNA3.1-CFPnls (Avinoam et al., 2011)
into a pBABE vector are listed in Table S2. pBABE-dsRed was constructed in a similar manner (see Table S2 for primer information).
4hrs prior to transfection, 3310
6
cells Platinum E Cells (Cell Biolabs) were seeded in 100-mm culture dish. 10mg of appropriate retro-
viral plasmid DNA (indicated in figure/video legends) was transfected using FuGENE 6 (Roche). Viral suspension was collected from
the conditioned media 48hrs post transfection. The media was centrifuged (1000 RCF/10mins) to remove cell debris. The clarified
viral suspension was used to transduce primary myoblasts. First passage primary myoblasts were seeded at 30,000 cells per well
of a 6-well plate, 48 hrs prior to transduction using Polybrene (6mg/mL) (Merck: #TR 1003-G) as a transduction reagent. 1.5hrs after
infection, viral suspension was removed, cells were washed with PBS, and fresh Bioamf-2 culture media was added to cells. 24hrs
following transfection, cells were trypsinized and seeded in 8-chamber slide (Ibidi #80826) at a density of 20,000/well and allowed to
attach. The following day, proliferation media was replaced with the appropriate treatment condition and imaging began (time of initi-
ation and duration are shown in figure legends).
Spinning-disc confocal microscopy
Live cell imaging (37C, with 5% CO2) was performed using Olympus IX83 fluorescence microscope controlled via VisiView software
(Visitron Systems GmbH) and equipped with CoolLED pE-4000 light source (CoolLED Ltd., UK), an PLAPON60XOSC2 NA 1.4 oil
immersion objective, and a Prime 95B sCMOS camera (Photometrics). Fluorescence excitation and emission were detected using
filter-sets 488 nm and 525/50 nm for GFP, 561nm and 609/54 nm for mCherry.
Cell Discoverer 7-Zeiss microscopy
Fixed samples (Figure 1B) were imaged using Cell discoverer 7-Zaiss inverted in widefield mode with s CMOS 702 camera Carl Zeiss
Ltd. Images were acquired using a ZEISS Plan-APOCHROMAT 20x / 0.95 Autocorr Objective. ZEN blue software 3.1 was used for
image acquisition using AF647 for the acquisition of the MyHC signal and DAPI for the nuclei. If necessary, linear adjustments to
brightness and contrast were applied using ImageJ v1.52 software (Schneider et al., 2012).
Nikon Eclipse Ti2 microscopy
Fixed samples (Figures 2 and 3) were imaged using the Nikon Eclipse Ti2 microscope and NIS-Elements imaging software
ver.5.11.00. using a 10x objective for the acquisition of MyHC, MYOG, KI-67, pH3 and DAPI staining. If necessary, linear adjustment
to brightness and contrast were applied using Photoshop. Live-imaging of tdTomato expressing myoblasts (Videos S1 and S2) were
imaged using the Nikon Eclipse Ti2 microscope and NIS-elements software, using a 10x objective. linear adjustments to brightness
and contrast were applied using ImageJ v1.52 software (Schneider et al., 2012).
Quantification of fusion index, MYOG nuclear localization, and migration rate
Following immunostaining and imaging, a fusion index was quantified by manually identifying nuclei found in a MyHC positive cell
with at least 2 nuclei. Then the values were expressed as a percentage of the total nuclei per field. Briefly, in Figures where fusion
index is stratified into subgroups of fiber size, the nuclei number in MyHC positive cell was manually quantified in a given field
and stratified into groups of mononucleated, bi-nucleated myotubes, myotubes with 3-10 nuclei and myotubes with greater than
10 nuclei. For myotube growth curves, LifeAct-EGFP; nTnG reporter primary myoblasts underwent time-lapse imaging beginning
at 8 hours after treatment and followed until 23 hours. Fields were analyzed hourly, and nuclei per cell was quantified and stratified
into mononucleated, bi-nucleated, trinucleated and cells with R4 nuclei. In later experiments nuclei were segmented and count using
the Cellpose software (Stringer et al., 2021) together with a home-made python script to match the nuclei to the cells. Nuclei positive
after MYOG immunofluorescence staining were segmented and overlapped computationally over an image of the total segmented
nuclei for each field, and the percent of MYOG positive out of the total was nuclei calculated. Cell migration rate was calculated by
tracking the nuclei and calculating their displacement in x and y between time frames using a home-made script.
Data-driven cell fusion simulations
For each experiment we defined a matched ‘‘shadow’’ simulation that compared the experimental fusion dynamics to a scenario
where cell-cell fusion occurred randomly. The input for the ‘‘shadow’’ simulation was the observed distribution of multinucleated cells
in each time frame. This included the number of cells with a single, pair, triplet or quartette-or-more nuclei that were manually anno-
tated with a time resolution of 60 minutes intervals between consecutive measurements. The estimated number of fusion events
per time interval was calculated as the difference between the weighted accumulated number of multinucleated cells P
i=4
i=2
½ðCtðiÞ
Ct1ðiÞÞ ði1Þ, where i is the number of nuclei in a multinucleated cell, t is the time interval and Ct(i) is the number of cells with
i nuclei at time interval t. We assumed that the number of cells remain constant throughout the experiment. The input for the simu-
lation included (1) N - the number of nuclei determined at the onset of the experiment, where each of the cells had exactly one nu-
cleus. And (2) N_fusion - the list of estimated fusion events per time interval. For each time interval t, we simulated N_fusion(t) fusion
events by randomly selecting two cells and fusing them, generating one cell with the joint number of nuclei for the next simulation
round. For each time interval, we recorded the probability of a nucleus to be part of a 4-nuclei cells, i.e., what is the fraction of nuclei
in a multinucleated cell that contains 4 or more nuclei. This fraction was used as a measure to compare experiments to simulations.
Due to annotation limitations, we considered multinucleated cells that contained 4 nuclei. This means that a multinucleated cell with
ll
OPEN ACCESS
Article
Developmental Cell 56, 3349–3363.e1–e6, December 20, 2021 e4
more than 4 nuclei was annotated as a 4-nuclei cell. On the one hand, this limitation had implications in the calculations of the esti-
mated number of fusions - which was a lower bound to the true number of fusion events. On the other hand, the calculated probability
for a nucleus to take part in a 4-nucleated cell was also a lower bound to the true probability. This double lower bound effect is
expected to cancel each other and also takes place only in the later stages of an experiment.
Statistical significance for each experiment was calculated using a Bootstrapping approach. For each experiment we performed
1000 simulations. For each time interval in each simulation, we recorded whether the probability of a nucleus to be in a 4-multinucle-
ated cell was equal or exceeded the experimental observation. The p-value was defined as the probability for a simulation to exceed
the experiment with this measure. We used a cutoff threshold %0.05 (50 simulations out of 1000 for each experiment) to reject the null
hypothesis of random fusions. Importantly, this assessment provides a p-value for each time interval in each experiment. As a more
realistic scenario we considered the possibility that the probability of selecting a cell for fusion was proportional to the number of nuclei
within it. This followed the simplistic assumption that the area of a n-nucleated cell is n times the size of a single-nucleated cell. Thus,
simulating the situation where a cell fuses randomly, but its chance of bumping-and-fusing into another cell is dependent on its area.
Quantitative real-time PCR (qRT-PCR)
Total RNA was isolated using Tri-Reagent (MERCK) according to the manufacturer’s instructions. cDNA was synthesized with the
High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems) according to the manufacturer’s instructions. qRT-PCR was
performed with SYBR green PCR Master Mix (Applied Biosystems) using the StepOnePlus Real-time PCR system (Applied Bio-
systems). Values for specific genes were normalize to either Gapdh or Hprt housekeeping control as indicated in Figure legend.
Expression was calculated using the ddCT method. Primer sequences are listed in Table S1.
Western Blot analysis
Cultured cells and whole tissues extracts were prepared with RIPA buffer supplemented with protease inhibitor cocktail (MERCK
P8340), and phosphatase inhibitor cocktails (MERCK P5726 and P0044). Western blotting was performed using the Mini-PROTEAN
Tetra Cell electrophoresis system, and transferred to PVDF membranes. The following primary antibodies concentrations were used
p-CAMKII 1:1000 (Abcam ab182647), CaMKII 1:1000 (Cell Signaling 3362), GAPDH 1:10,000 (Abcam ab181602), p-ERK1/2 1:20,000
(MERCK M9692), ERK1/2 1:40,000 (MERCK M5670), p-RXR 1:1000 (Affinity Biosciences), RXR antibody 1:200 (SCBT sc-553), p-
RYR 1:2000(Abcam ab59225), RYR 1:1000 (ab2868), p-Rac1 1:1000 (Millipore 07-896-I) Rac1 1:1000 (Millipore 05-389), (and Vinculin
(provided by Benny Geiger, Weizmann Institute of Science). Horseradish peroxidase conjugated secondary anti-mouse, anti-rabbit
or anti-goat was used to detect proteins (Jackson Immunology). Western blots were imaged using the Chemidoc Multiplex system
(Bio-rad) and Image Lab software (Bio-rad).
Co-immunoprecipitation (Co-IP)
Primary myoblasts derived from gastrocnemius muscle were pooled from 10 mice and plated on 15cm dishes and allowed to adhere
for 24 hours. The following day, Bio-Amf2 media was replaced supplemented either with DMSO or 1mm SCH772984. Cells were
treated for 4 hours, and then nuclear lysates were prepared according to the instructions of the Universal Magnetic Co-IP KIT (Active
Motif cat#54002). 1mg of protein was used to immunoprecipitate ERK1/2using 2ug of ERK1/2Antibody (MERCK M7927). Rabbit IgG
was used as a control. Reactions were resuspended in 2x Sample buffer with DTT and loaded onto a 12% Tris-glycine SDS-page gel.
1% of original volume of lysate loaded into IP reaction was loaded into the gel as input control. Membranes were blotted with RXR
antibody (SCBT sc-553).
Cloning and expression of CaMKII adenovirus for fusion assay
CaMKII-dcDNA was PCR amplified from mouse primary myoblasts using primers, CAMK2D-F and CAMK2D-R (all cloning primer
sequences are available in Table S2), designed against published CaMKII-dsequences, and ligated into the PGEM-T-easy cloning
system (Promega), and sequence validated. The T287V mutation was introduced by PCR assembly. A 909bp upstream PCR frag-
ment was amplified with primer sequences designed to incorporate a XhoI site and FLAG tag at the N-terminus of CAMK2D and
a the T287V mutation, using primers XhoI-FLAG-CAMK2D-F and CAMK2D-T287V-IN-R. The 640bp downstream PCR fragment
was similarly amplified with a primer to introduce the T287V mutation and a BamHI site using the primers CAMK2D-T287V-IN-F:
and CAMK2D-BamHI-R. Both PCR fragments were used as template for an assembly PCR reaction with XhoI-FLAG-CAMK2D-F
and CAMK2D-BamHI-R primers to generate a 1525 bp product, which was ligated back into PGEM. Similarly, the WT CAMK2D
was amplified with the same primers to incorporate the FLAG-tag and ligated back into PGEM. The 1525bp FLAG-CAMK2D
WT
and FLAG-CAMK2D
T287V
fragments were digested out of PGEM with BaMHI and XhoI and ligated into pEGFP-C1 (Clontech). A
2865 bp product EGFP-FLAG-CAMK2D
WT
or EGFP-FLAG- CAMK2D
T287V
was digested out using KPNI and ECORV and inserted
into RedTrackCMV (addgene plasmid #50957). RedTrack-CMV-EGFP-FLAG-CAMK2D
WT
(Ad-CaMK2D
WT
), RedTrack-CMV-
EGFP-FLAG-CAMK2D
T287V
(Ad-CaMK2D
T287V
), and empty RedTrack-CMV (Ad-Ctrl), vector were used as template to grow adeno-
virus using the Adeasy system as previously described (Luo et al., 2007). Myoblasts were infected with crude adenoviral lysate at an
MOI of 100 at the time of plating (reverse infection) in BioAmf2 media. Following overnight incubation, the cells were washed once
with warm DM and were incubated for 72 hours in DM and number of nuclei per fiber was quantified.
Myomaker plasmid construct and overexpression fusion assay
To generate pBabe-Mymk-CFPnls, the CDS sequence of murine MYMK (Millay et al., 2013) was subcloned in the MCS region of
pBabe-CFPnls plasmid using restriction free cloning. Primer sequences are provided in in Table S2. Retroviruses were generated
as described above. Myoblasts were seeded at 7x10
3
per well of 96 well. The following morning cells were infected with viral
prep supernatants together with polybrene (6ng/mL) for 1 hour, then replaced with fresh growth media, then after 8 hours the media
was changed according to indicated conditions. Cells were fixed and stained at 18 hours post treatment.
ll
OPEN ACCESS Article
e5 Developmental Cell 56, 3349–3363.e1–e6, December 20, 2021
Histology and CSA quantification
14 days post CTX induced reinjury, muscles were excised and fixed in 4% PFA, embedded in paraffin, and sectioned. Muscles were
cut transversely in the center and cut into serial sections at 0.3mm intervals. For analysis of muscle fiber cross-sectional area (CSA),
sections were permeabilized and stained with WGA and DAPI. The entire muscle transverse section of WT and scDKO mice taken at
identical locations within the muscle were imaged using the Nikon at 10x. CSA was quantified using the Open-CSAM, semi-auto-
mated analysis tool with ImageJ (Desgeorges et al., 2019). Each field was evaluated for accuracy and manually corrected. At least
9,000 fibers/mouse were measured.
Proximity ligation assay
Primary myoblasts were isolated from Wiltype or mTmG expressing mice as described above. Following treatment and fixation
with 4% PFA, PLA was performed using the Duolink Proximity Ligation Assay (MERCK) according to manufacturer’s instructions.
Validation studies with individual antibodies were performed (not shown) to demonstrate specificity of the PLA signal. For the
CaMKII:MYMK PLA, rabbit-anti-TMEM8C (MERCK HPA051846,1:50) and mouse-anti-CaMKII (SCBT sc-5306, 1:50) were used.
For the Rac1:CaMKII PLA, rabbit anti-CaMKII (AB52476 1:100) and mouse-anti-Rac1(Millipore 05-389 1:100) were used. Where
indicated, phalloidin-488 was used (ab176753) and DAPI (MERCK D9542, 5ug/ml
Statistical analysis
Sample size was chosen empirically following previous experience in the assessment of experimental variability. Generally, all
experiments were carried out with nR3 biological replicates. The analyzed animal numbers or cells per groups are described in
the respective figure legends. All animals were matched by age and gender, and cells harvested from mice of similar age. Animals
were genotyped before and after completion of the experiment and were caged together and treated in the same way. Statistical
analysis was carried out using Prism software. Whenever comparing between two conditions, data was analyzed with two tailed
student’s t-test. If comparing more than two conditions, ANOVA analysis with multiple comparisons was executed. In all Figures,
measurements are reported as mean of multiple biological repeats, and the error bars denote SEM, unless otherwise specified in
the figure legend. Throughout the study, threshold for statistical significance was considered for p-values%0.05, denoted by one
asterisk (*), two (**) if P%0.01, three (***) if P <0.001 and four (****) if P%0.001.
ll
OPEN ACCESS
Article
Developmental Cell 56, 3349–3363.e1–e6, December 20, 2021 e6
