ReportIntraflagellar Transport Complex B Proteins
Regulate the Hippo Effector Yap1 during
CardiogenesisGraphical AbstractHighlightsd IFT proteins restrict proepicardial and myocardial
development
d IFT proteins act independently of primary cilia in this process
d IFT proteins modulate BMP signaling by tuning Yap1-Tead
activity
d IFT88 and IFT20 interact with YAP1 in the cytoplasm to set
Yap1-Tead activityPeralta et al., 2020, Cell Reports 32, 107932
July 21, 2020 ª 2020 The Authors.
https://doi.org/10.1016/j.celrep.2020.107932Authors
Marina Peralta, Laia Ortiz Lopez,
Katerina Jerabkova, ...,
Benedicte Delaval, Sigole`ne M. Meilhac,
Julien Vermot
Correspondence
jvermot@imperial.ac.uk
In Brief
Peralta et al. show that intraflagellar
transport (IFT) complex B proteins (Ift88,
Ift54, and Ift20) modulate the Hippo
pathway effector Yap1 in zebrafish and
mouse. This cytoplasmic interaction is
key to restrict proepicardial and
myocardial development.ll
OPEN ACCESS
llReport
Intraflagellar Transport Complex B Proteins
Regulate the Hippo Effector Yap1
during Cardiogenesis
Marina Peralta,1,2,3,4 Laia Ortiz Lopez,1,2,3,4,13 Katerina Jerabkova,1,2,3,4,13 Tommaso Lucchesi,5,6,7,13 Benjamin Vitre,8
Dong Han,5,6 Laurent Guillemot,5,6 Chaitanya Dingare,9 Izabela Sumara,1,2,3,4 Nadia Mercader,10,11 Virginie Lecaudey,9
Benedicte Delaval,8 Sigole`ne M. Meilhac,5,6 and Julien Vermot1,2,3,4,7,12,14,*
1Institut de Ge´ne´tique et de Biologie Mole´culaire et Cellulaire, Illkirch, France
2Centre National de la Recherche Scientifique, UMR7104, Illkirch, France
3Institut National de la Sante´ et de la Recherche Me´dicale, U964, Illkirch, France
4Universite´ de Strasbourg, Illkirch, France
5Imagine–Institut Pasteur, Laboratory of Heart Morphogenesis, Paris, France
6INSERM UMR1163, Universite´ Paris Descartes, Paris, France
7Sorbonne Universite´, Colle`ge Doctoral, F-75005, Paris, France
8Centre de Recherche en Biologie Cellulaire de Montpellier (CRBM), CNRS, Universite´ de Montpellier, Montpellier, France
9Institute for Cell Biology and Neurosciences, Goethe University of Frankfurt, Frankfurt, Germany
10Institute of Anatomy, University of Bern, Bern, Switzerland
11Centro Nacional de Investigaciones Cardiovasculares (CNIC), Madrid, Spain
12Department of Bioengineering, Imperial College London, London, UK
13These authors contributed equally
14Lead Contact
*Correspondence: jvermot@imperial.ac.uk
https://doi.org/10.1016/j.celrep.2020.107932SUMMARYCilia and the intraflagellar transport (IFT) proteins involved in ciliogenesis are associatedwith congenital heart
diseases (CHDs). However, the molecular links between cilia, IFT proteins, and cardiogenesis are yet to be
established. Using a combination of biochemistry, genetics, and live-imaging methods, we show that IFT
complex B proteins (Ift88, Ift54, and Ift20)modulate the Hippo pathway effector YAP1 in zebrafish andmouse.
We demonstrate that this interaction is key to restrict the formation of the proepicardium and the myocar-
dium. In cellulo experiments suggest that IFT88 and IFT20 interact with YAP1 in the cytoplasm and function-
ally modulate its activity, identifying a molecular link between cilia-related proteins and the Hippo pathway.
Taken together, our results highlight a noncanonical role for IFT complex B proteins during cardiogenesis and
shed light on a mechanism of action for ciliary proteins in YAP1 regulation.INTRODUCTION
Primary cilia are immotile microtubule-based organelles, well
known for being both chemical and/or mechanical sensors (Ber-
bari et al., 2009; Ferreira et al., 2019). Disruption of cilia function
causes multiple human syndromes known as ciliopathies (Reiter
and Leroux, 2017). Cilia are required for cardiac development
and mutations in cilia-related proteins have been linked to
congenital heart diseases (CHDs) (Klena et al., 2016; Li et al.,
2015; San Agustin et al., 2016; Slough et al., 2008). Nevertheless,
the specific role of cilia and cilia-related proteins during cardio-
genesis is still unclear.
Intraflagellar transport (IFT) proteins are required for the trans-
port of cilia components along the axoneme and are thus essen-
tial for cilia formation and function (Rosenbaum and Witman,
2002). The role of IFT proteins during ciliogenesis is highly
conserved across organisms (Taschner et al., 2012). The IFTma-This is an open access article under the CC BY-Nchinery is composed of two biochemically distinct subcom-
plexes, IFT-A and IFT-B (Taschner et al., 2012). The IFT complex
B member IFT20, which localizes inside the cilium and at the
Golgi complex in mammalian cells, participates in the sorting
and/or transport of membrane proteins for the cilia (Follit et al.,
2006). Within the IFT complex B, IFT20 interacts with IFT54/Eli-
psa (Omori et al., 2008; Zhu et al., 2017) and IFT88, which is
essential for flagellar assembly in Chlamydomonas and ciliogen-
esis in vertebrates (Pazour et al., 2000). Mutations in some IFT
proteins have been identified in CHDs (Li et al., 2015). IFT pro-
teins also display noncanonical, cilia-independent functions
(Hua and Ferland, 2018; Vertii et al., 2015). IFT88, for example,
is needed for spindle orientation and organization, cleavage
furrow ingression, or extra centrosome clustering in dividing cells
(Delaval et al., 2011; Taulet et al., 2017, 2019; Vitre et al., 2020)
and regulates G1-S phase transition in nonciliated cells (Robert
et al., 2007). IFT20, together with IFT88 and IFT54, plays a roleCell Reports 32, 107932, July 21, 2020 ª 2020 The Authors. 1
C-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
(legend on next page)
2 Cell Reports 32, 107932, July 21, 2020
Report
ll
OPEN ACCESS
Report
ll
OPEN ACCESSin the establishment of the immune synapse in T lymphocytes
lacking cilia (Finetti et al., 2009; Galgano et al., 2017). These ob-
servations suggest that IFT proteins could play a key role in em-
bryonic cardiogenesis through both cilia-dependent and -inde-
pendent functions.
The Hippo signaling mediators, YAP1/WWTR1 (TAZ), consti-
tute a key signaling pathway for the regulation of cardiac devel-
opment (Fukui et al., 2018; Lai et al., 2018; Ragni et al., 2017; Xin
et al., 2011) and cardiac regeneration (Bassat et al., 2017; Leach
et al., 2017) in vertebrates. For example, YAP1/WWTR1 (TAZ) are
required for epicardium and coronary vasculature development
(Singh et al., 2016) in mice. In addition, Yap1 signaling has
been reported to regulate the number of atrial cardiomyocytes
derived from Islet1 (Isl1)-positive (+) secondary heart field
(SHF) cells in the zebrafish (Fukui et al., 2018). Changes in cell
shape, substrate stiffness, and tension forces activate a phos-
phorylation-independent YAP1/WWTR1 (TAZ) modulation (Elo-
segui-Artola et al., 2017), which is mediated by the Motin family
(AMOT, AMOTL1, and AMOTL2) (Bratt et al., 2002; Zheng et al.,
2009). Motin proteins bind to YAP/WWTR1 (TAZ), sequestering
them in the cytoplasm in several cellular contexts (Agarwala
et al., 2015; DeRan et al., 2014; Nakajima et al., 2017; Wang
et al., 2011; Zhao et al., 2011). While it is known that ciliary pro-
teins from the Nephrocystin family modulate the transcriptional
activity of YAP1/WWTR1 (TAZ) (Frank et al., 2013; Grampa
et al., 2016; Habbig et al., 2012) and that the Hippo kinases
MST1/2-SAV1 promote ciliogenesis (Kim et al., 2014), the
connection between the Hippo pathway and cilia function re-
mains unclear.
Cardiogenesis involves an interplay between multiple tissue
layers. The epicardium is the outermost layer covering the heart.
This cardiac cell layer plays an essential role in myocardial matu-
ration and coronary vessel formation during development
(Ma¨nner et al., 2005; Moore et al., 1999; Wu et al., 2013) and
has a crucial role during regeneration (Gonza´lez-Rosa et al.,
2012; Kikuchi and Poss, 2012; Limana et al., 2011). Epicardial
cells derive from the proepicardium (PE), an extracardiac tran-
sient cluster of heterogeneous cells (Katz et al., 2012). The PE,
the sinoatrial node, and the atrial myocardium all derive from
the SHF (Buckingham et al., 2005; Mommersteeg et al., 2010;
van Wijk et al., 2009). In zebrafish, the transcription factor Isl1Figure 1. IFT Complex B Proteins Regulate Proepicardial Developmen
(A) Schematic representation of the zebrafish heart at 55 hpf in frontal and lateral
red, the avcPE in light green, and the epicardial cells in dark green.
(B) Percentage of avcPE cluster number in iguana (n = 43), ift88 (n = 50), and elips
between 50 and 57 hpf (Mann-Whitney test, iguana p value < 0.0001; ift88 p valu
(C) High-speed avcPE imaging (digital red mask to improve visualization) of igua
(D) Graphs show avcPE cell number quantified on iguana (n = 20), ift88 (n = 18 and
12; n = 15; n = 28 n = 18, respectively) (t test, iguana p value, 0.001; ift88 p value, 0.
whole-mount immunofluorescence (IF) of hearts using anti-myosin heavy chain a
(E) Percentage of avcPE cluster number on iguana
/
, ift88
/
, epi:GFP (n = 9),
(F) Graph shows avcPE cell number quantified in wild-type (n = 13), double-hete
Wallis, p value, 0.014). 3D projections of whole-mount IF of hearts. MHC (red) an
(G) Schematic representation of an E9.5 mouse embryo. Heart tube (HT) in dark
(H) Left-side graph shows quantification of PE volume (in cubic micrometers) in Ift2
KO, 9.063 106 mm3 ± 1.463 106) (t test, p value, 0.008) and that of Ift88KO (n = 4)
KO, 7.05 3 106 mm3 ± 1.2 3 106) (t test, p value, 0.003). In the lower panel, 3D p
marked using anti-Wt1 antibody. Thewhite dotted shapes enclose the PE area. In
the top. Arrowheads mark avcPE, and asterisk shows lack of avcPE. V, ventriclemarks a subset of SHF cells that give rise to part of the atrial
myocardium (de Pater et al., 2009; Witzel et al., 2012). In mice,
the PE is a single cell cluster located close to the venous pole
of the heart (Katz et al., 2012), while in zebrafish it is composed
of two cell clusters, avcPE (the main source of cells) and vpPE
(Peralta et al., 2013). PE cells give rise to the epicardium, part
of the coronary vasculature, and intracardiac fibroblasts
(Acharya et al., 2012; Katz et al., 2012; Mikawa and Fischman,
1992; Mikawa and Gourdie, 1996). The secreted signaling mole-
cules of the bone morphogenetic protein (BMP) family are indis-
pensable for PE formation (Ishii et al., 2010; Liu and Stainier,
2010; Schlueter et al., 2006).
Despite the increasing evidence for the role of IFT proteins in
cell signaling, it has been difficult to pinpoint the exact function
of cilia-related proteins outside the cilium. Without this key infor-
mation, the question of whether cilia-related proteins can affect
the developmental program independently of their cilia function
remains unresolved. Here, we provide a combination of in vivo
and in vitro analyses of IFT protein function showing that IFT
complex B proteins can modulate the Hippo pathway effector
Yap1. In particular, we show that IFT88 is required to restrict
PE and myocardium development through cytoplasmic activity.
RESULTS
IFT Complex B Proteins Regulate PE Development in
Zebrafish and Mouse
Considering the importance of cilia during cardiogenesis, we as-
sessed the role of several IFT complex B proteins during PE
development. In order to benefit from live imaging and genetics,
we used zebrafish as a model organism. We performed live-im-
aging focusing on the main PE cell source located near the atrio-
ventricular canal (avcPE) in ift88 (Tsujikawa and Malicki, 2004)
and elipsa/ift54 (Omori et al., 2008) mutants (Figures 1A–1D). Be-
tween 50 and 57 hours post fertilization (hpf), we found that ift88
mutants display multiple avcPE clusters (either two or three),
when controls have only one (Video S1). Using a wilms tumor 1
a (wt1a) enhancer trap line that marks proepicardial and epicar-
dial cells (Et(
26.5Hsa.WT1-gata2:EGFP)cn1; hereafter, epi:GFP)
(Peralta et al., 2013), we quantified avcPE and epicardial cell
number and found that ift88
/
 have increased avcPE andt in Zebrafish and Mice
views. The dorsal pericardium (DP) is shown in gray, the myocardium (myoc) in
a/ift54 (n = 38) mutants and their controls (n = 24; n = 43; n = 40, respectively),
e < 0.0001; elipsa p value, 0.014).
na and ift88 mutants and their control. Scale bars: 20 mm.
29), and elipsa (n = 25) mutants in epi:GFP background and their controls (n =
0002; and p value, 0.015, respectively; elipsa p value, 0.0005). 3D projections of
ntibody (MHC) (red) and GFP (green) expression.
and controls (n = 23) at 55 hpf (Mann-Whitney test, p value, 0.059).
rozygous controls (n = 10), and double ift88; iguana mutants (n = 9) (Kruskal-
d GFP (green) expression.
gray and PE in white. Section of the PE represented inside the yellow box.
0 KO (n = 4) and control mice (n = 3) (control, 5.353 106 mm3 ± 3.633 105; Ift20
and control embryo (n = 4) on the right (control, 2.783 106 mm3 ± 1.33 106; Ift88
rojections of whole-mount IF performed on control and Ift20 KO embryos. PE
all graphs, red bars indicatemean ±SD. In all panels, ventral views, anterior is to
; At, atrium.
Cell Reports 32, 107932, July 21, 2020 3
(legend on next page)
4 Cell Reports 32, 107932, July 21, 2020
Report
ll
OPEN ACCESS
Report
ll
OPEN ACCESSepicardial cell numbers compared to their controls at 50 and 55
hpf, respectively (Figures S1B and S2A). Similarly, we found that
elipsa
/
 also showed bigger avcPE compared to controls (Fig-
ures S1C and S2A00). Taken together, these data indicate that
ift88 and elipsa/ift54 are required to restrict PE size.
To determine whether primary cilia are required for PE forma-
tion, we took advantage of the iguana/dzip (Tay et al., 2010)
mutant, a well-established cilia mutant lacking primary cilia
(Kim et al., 2010). We confirmed the absence of primary cilia in
the pericardial cavity of the iguana
/
 using the cilia reporter
actb2:Mmu.Arl13b-GFP (Borovina et al., 2010; Figures S2C–
S2E; Video S2). Zebrafish ift genes are expressed maternally
(Sun et al., 2004; Figure S2E), and complete removal of both zy-
gotic andmaternal expression of ift88 leads to the ablation of pri-
mary cilia resembling the iguana mutant phenotype (Huang and
Schier, 2009). Live imaging revealed that a significant fraction of
iguana mutants lacked an avcPE between 50 and 57 hpf (Fig-
ure 1B; Video S3). At 55 hpf, iguana
/
 displayed decreased
avcPE cell numbers (Figures 1D, S1A, and S2A0). Since the
iguana mutants presented a phenotype opposite to that of the
ift88mutants, we analyzed the ift88
/
; iguana
/
; epi:GFP dou-
blemutant to elucidate whether the differences could be due to a
new cilia-independent function. The ift88
/
; iguana
/
; double
mutant showsmultiple avcPE cluster formation and an increased
avcPE cell number, reminiscent of ift88 loss of function (Figures
1E, 1F, and S1D). We conclude that Ift88 modulates the PE cell
number independently of its cilia function.
We assessed whether the role of IFT complex B proteins dur-
ing PE formation was conserved in mammals. We analyzed the
PE in Ift20 and Ift88 knockout (KO) mice at mouse embryonic
day 9.5 (E9.5) (Figures S3A and S3B). First, we performed immu-
nofluorescence (IF) using an anti-Arl13b antibody to quantify the
percentage of ciliated PE cells in Ift20, Ift88 KO, and wild-type
mice (Figures S3C–S3C00). We found that, while over 60% of
the wild-type PE cells were ciliated, in Ift88 KO mice only 3%
of cells were ciliated and Ift20 KO embryos lack cilia. We
compared the PE volume of Ift20 and Ift88 KOmice to their con-Figure 2. BMP Signaling Is Increased in ift88, elipsa/ift54, and Ift20 Mu
(A) Graphs show total myocardial (myoc) and atrial-myoc cell number quantified in
0.0034; atrial myoc p value < 0.0001).
(A0) 3D projections of whole-mount immunofluorescence (IF) of hearts using myo
(B) Graphs show total myoc and atrial-myoc cell number quantified in elipsa (n = 14
myoc p value, 0.017).
(C) Graphs show total myoc and atrial-myoc cell number quantified in yap1 (n = 14
myoc p value, 0.0009).
(D) Graph shows number of p-smad1/5+ cells in the myoc and dorsal pericardium
p value, 0.009) and DP (t test, p value, 0.036).
(D0) 3D projections of whole-mount IF of hearts. MHC (red), epi:GFP (green), and p
p-smad1/5 (same embryos as in A0).
(E) Graph shows number of p-smad1/5+ cells in themyoc andDP quantified in elip
and DP (t test, p value, 0.0004).
(F) Graph shows number of p-smad1/5+ cells in themyoc andDPquantified in yap1
0.0001) and DP (t test, p value, 0.599).
(G) Graph shows that the percentage of p-smad1/5/9+ PE cells is higher in Ift20 K
1,414 nuclei analyzed) mice (Chi-square test of homogeneity = 51.593; p value,
(G0 ) Control and Ift20 KO IF confocal cryosections labeled with WT1 (red), p-sma
area. Arrowheadsmark p-smad1/5/9 andWt1 double-positive cells. Individual cha
visualization of signal intensity, where green is the minimum and red is the maxim
SD. V, ventricle; At, atrium. Ventral views, anterior is to the top. p-smad, p-smadtrols, using an anti-Wt1 antibody (a well-established PE marker;
Carmona et al., 2001). Mutant embryos showed increased PE
volume (Figures 1G and 1H; Videos S5 and S6). These results
suggest that IFT protein function is conserved during PE forma-
tion in mouse and zebrafish.
To test whether IFT complex B affected other cardiac tissues,
we focused on the myocardium. Cell quantification showed
increased myocardial cell number in ift88
/
 compared to their
controls at 50 hpf (Figures 2A and 2A0) and 55 hpf (Figure S4A).
Interestingly, the increment was mostly affecting the atrium.
Similar data were obtained with the elipsamutants at 55 hpf (Fig-
ure 2B). We concluded that the absence of IFT complex B pro-
teins also affects the myocardium.
To determine whether Isl1+ SHF cells were altered at earlier
stages leading to the increment in avcPE and atrial myocardial
cells, we quantified Isl1+ cells in ift88 and amotl2a mutants at
36 hpf. We did not find differences between ift88
/
, amotl2a
/
,
and their controls (Figure S1E).
BMP Signaling Is Increased in ift88, elipsa/ift54, and
Ift20 Mutants during PE Development
As it is known in zebrafish that overexpressing BMP increases
PE size (Liu and Stainier, 2010), we investigated whether ift88
loss influences BMP signaling. We first assessed bmp4 expres-
sion by in situ hybridization (ISH) in ift88 and elipsa mutants. We
found that ift88
/
 display ectopic expression of bmp4 in the
dorsal pericardium (DP) at 48 hpf (Figure S4B). To validate that
the upregulation is functional in vivo, we quantified cellular
BMP activity by using p-smad1/5 as a readout and confirmed
that the absence of Ift88 leads to increased BMP signaling activ-
ity in the DP and myocardium, especially in the venous pole (Fig-
ures 2D, 2D0, and S4C). By 55 hpf, bmp4 expression at the atrio-
ventricular canal myocardium is reduced and the expression at
the venous pole is almost undetectable in controls. By contrast,
ift88 mutants presented strong bmp4 expression in both do-
mains (Figure S4B). Myocardial p-smad1/5 was also increased
in ift88
/
 compared to controls, but not in the DP (Figurestants
ift88 (n = 16) mutants and controls (n = 11) at 50 hpf (t test, total myoc p value,
sin heavy chain antibody (MHC) (red).
) mutants and controls (n = 15) at 55 hpf (t test, total myoc p value, 0.0026; atrial
) mutants and controls (n = 17) at 55 hpf (t test, total myoc p value, 0.0001; atrial
(DP) quantified in ift88mutants (n = 8) and controls (n = 7) at 50 hpf. Myoc (t test,
-smad1/5 (yellow). Arrowsmark dorsal pericardial cells positive for epi:GFP and
samutants (n = 14) and controls (n = 10) at 55 hpf.Myoc (t test, p value < 0.0001)

/
; tcf21:nls-GFP (n = 12) and controls (n = 11) at 55 hpf. Myoc (t test, p value <
O (n = 4 embryos: 1,862 nuclei analyzed) compared to control (n = 3 embryos:
6.829e-13 on 1 degree of freedom).
d1/5/9 (yellow), and Hoechst (blue) at E9.5. Yellow dotted lines enclose the PE
nnels are displayed for p-smad1/5/9 (signal is shown as ice LUT to facilitate the
um), Hoechst (white), and WT1 (white). In all graphs, red bars indicate mean ±
1/5 in (D0) and p-smad1/5/9 in (G0).
Cell Reports 32, 107932, July 21, 2020 5
(legend on next page)
6 Cell Reports 32, 107932, July 21, 2020
Report
ll
OPEN ACCESS
Report
ll
OPEN ACCESSS4C and S4D). We reached similar conclusions when analyzing
the elipsa mutants at 55 hpf (Figures 2E, S4B, and S4C). To
confirm that the regulation of BMP signaling by IFT is conserved
in vertebrates, we analyzed p-smad1/5/9 in Ift20 and Ift88 KO
and control mice at E9.5 and found that only the Ift20 mutants
displayed a higher percentage of p-smad+ cells compared to
controls (Figures 2G, 2G0, and S3D). Together, these results sug-
gest that Ift88 and Ift54 modulate BMP signaling activity in the
myocardium and the pericardium of the zebrafish, and that
IFT20 plays a similar role in the mouse PE.
Yap1-Tead Activity Is Increased in ift88, elipsa/ift54,
and Ift20 Mutants during PE Development
Yap1 participates in the regulation of Bmp signaling during SHF
development (Fukui et al., 2018). This led us to assess whether
Yap1/Wwtr1-Tead is active during PE formation using the
4xGTIIC:d2GFP reporter line (Miesfeld and Link, 2014).We found
that the reporter was active in the PE cluster, myocardium, peri-
cardium, and epicardial cells at 55 hpf (Figure S5A). We per-
formed time-lapse analysis to study the dynamics of Yap1/
Wwtr1-Tead activity during PE development (Video S4). GFP
quantification in PE and pericardial cells showed higher Yap1/
Wwtr1-Tead activity (GFP average intensity) in the PE cells
than in the pericardial cells (75% of the embryos) (Figure S5B;
Table S1).
To test whether pericardial cells have differential Yap1 activity
at earlier stages, before avcPE formation, we analyzed
4xGTIIC:d2GFP reporter expression at 36 hpf (Figure S5D). WeFigure 3. Yap1-Tead Activity Is Increased in ift88, elipsa/ift54, and Ift2
(A) Graphs show number of Yap1+ cells in the myocardium (myoc) and dorsal per
and n = 7, respectively). ift88mutants show increased Yap1+myoc cell numbers (t
p value, 0.07). elipsa mutants show higher Yap1+ myoc cell numbers (t test, p va
(A0) Control and ift88
/
, epi:GFP immunofluorescence (IF) confocal sections lab
(white), and DAPI (blue) at 55 hpf. Individual channel is displayed for Yap1 (signal i
is the minimum and red is the maximum). Yellow arrows mark nuclear Yap1-pos
(B) Graph shows avcPE cell number quantified in control (n = 9), ift88
/
; epi:GFP
20) embryos. Control embryos treated with Verteporfin showed smaller avcPE com
epi:GFP embryos presented lower avcPE cell numbers than nontreated ift88
/

bigger avcPE compared to untreated (t test, p value, 0.01) and treated controls (
(C) Graph shows number of p-smad1/5+ cells in the myoc quantified in control (n
(n = 5) and control (n = 6) embryos from 31 to 55 hpf. Control embryos treated
untreated controls (t test, p value, 0.0219). Verteporfin-treated ift88
/
; epi:GFP
embryos (t test, p value, 0.0173). ift88
/
; epi:GFP embryos showedmore p-smad
(t test, p value < 0.0001).
(D) Percentage of avcPE cluster number in amotl2a+/+ (n = 26) and amotl2a
/
 (n
(E)Whole-mount bmp4 in situ hybridization (ISH) in amotl2a+/+ (n = 18/25) and amo
Scale bars: 20 mm.
(F) Graph shows number of p-smad1/5+ cells in the myoc quantified in Verteporfin
amotl2a
/
 (n = 12) embryos. Treated embryos showed decreased p-smad1/5+
(F0) Graph shows avcPE cell number quantified in Verteporfin (20 mM)-treated am
embryos. Treated embryos showed decreased avcPE cell numbers compared to
(G) Whole-mount bmp4 ISH in XAV939 (10 mM)-treated amotl2a
/
 (n = 16) from
decreased (n = 8/16) or absent (n = 7/16) bmp4 expression at the atrioventricula
(H) Graphs show percentages of YAP1+ PE cells and double-YAP1-AMOTL1-posi
embryos: 929 nuclei analyzed) mice at E9.5. The percentage of nuclear YAP1+ PE c
freedom) and nuclear YAP1-AMOTL1+ cells (Chi-square test of homogeneity = 12
control mice.
(H0) Control and Ift20 KO IF confocal sections labeled with TBX18 (red), YAP1 (ye
(H00) Zoomed region shows the difference between nuclear YAP1+ cells (yellow arro
cell nuclei. YAP1 signal is shown as fire LUT to facilitate the visualization of signal
red bars indicate mean ± SD. In all panels, ventral views, anterior is to the top. Vobserved that some pericardial cells show stronger GFP inten-
sity (hence higher levels of active Yap1) than others. Consistently
with our data at later stages (Video S4; Figure S5B), pericardial
cells at the area that will give rise to the avcPE show stronger
GFP intensity. Additionally, we obtained similar data with Ctgf
antibody, a well-characterized Yap1 target gene (Zhao et al.,
2008; Figure S5E).
To test whether the increased avcPE cell number was due to
abnormal Yap1 activity, we performed IF to quantify the number
of nuclear Yap1+ cells in ift88 and elipsamutants (Figures 3A, 3A0
and S5F–S5G0). At 55 hpf, ift88
/
 showedmoremyocardial with
nuclear Yap1+ cells than controls, especially in the atrium
(Figure 3A-A’). Similar data were obtained with the elipsa mu-
tants (Figure 3A). To further validate the link between increased
PE size and Yap1-Tead activity, we treated the ift88
/
 with
the drug Verteporfin, which binds to YAP and changes its confor-
mation, blocking its interaction with TEAD (Liu-Chittenden et al.,
2012). To assess Verteporfin specificity, we performed time-
lapse imaging on Verteporfin-treated 4xGTIIC:d2GFP embryos.
We measured the Yap/Wwtr1-Tead activity (GFP average inten-
sity) in the same PE and pericardial cells at several time points:
before adding Verteporfin (t0), after 2-h treatment with Vertepor-
fin, and 45 min after washing out (Figure S5C). We found that
Yap/Wwtr1-Tead activity is significantly decreased with Verte-
porfin, confirming the specificity of the drug on the embryo.
We next treated ift88
/
 and control (ift88+/+ and +/
) embryos
with Verteporfin (Figure 3B). Control embryos treated with Verte-
porfin showed smaller avcPE compared to untreated controls.0 Mutants during Proepicardium Development
icardium (DP) quantified in ift88 (n = 13), elipsa (n = 6), and their controls (n = 12
test, p value, 0.03) and a tendency toward higher Yap1+ DP cell numbers (t test,
lue, 0.036).
eled with anti-myosin heavy chain antibody (MHC) (red), GFP (green), -Yap1
s shown as ice LUT to facilitate the visualization of signal intensity, where green
itive atrial myocardial cells. White arrowheads mark the avcPE.
(n = 8) and Verteporfin (5 mM)-treated ift88
/
; epi:GFP (n = 16) and control (n =
pared to untreated controls (t test, p value, 0.04). Verteporfin-treated ift88
/
;
; epi:GFP embryos (t test, p value, 0.027). ift88
/
; epi:GFP embryos showed
t test, p value, 0.0001).
= 7), ift88
/
; epi:GFP (n = 5), and Verteporfin (20 mM)-treated ift88
/
; epi:GFP
with Verteporfin showed decreased p-smad1/5+ cell numbers compared to
embryos presented less p-smad1/5+ cells than non-treated ift88
/
; epi:GFP
1/5+ cells compared to untreated (t test, p value < 0.0001) and treated controls
= 38) embryos between 50 and 57 hpf (Mann-Whitney test, p value, 0.08).
tl2a
/
 (n = 13/17) embryos (55 hpf). Yellow arrow shows bmp4 overexpression.
(20 mM)-treated amotl2a
/
 (n = 19) embryos from 31 to 55 hpf and untreated
cell numbers compared to untreated ones (t test, p value, 0.0148).
otl2a
/
 (n = 13) embryos from 31 to 55 hpf and untreated amotl2a
/
 (n = 10)
untreated ones (t test, p value, 0.0059).
31 to 55 hpf and untreated embryos (n = 11). Treated embryos showed either
r canal myoc and the venous pole. Scale bars: 20 mm.
tive PE cells in Ift20 KO (n = 5 embryos: 1,196 nuclei analyzed) and control (n = 4
ells (Chi-square test of homogeneity = 25.354; p value, 4.77E-07 on 1 degree of
,025; p value, 5.25E-04 on 1 degree of freedom) were higher in Ift20 KO than in
llow), and Hoechst (blue) at E9.5. The white dotted lines enclose the PE area.
ws) and YAP1-negative cells (yellow asterisks). Hoechst signal (blue) highlights
intensity, where blue is the minimum and yellow is the maximum. In all graphs,
, ventricle; At, atrium.
Cell Reports 32, 107932, July 21, 2020 7
Figure 4. IFT88 and IFT20 Are Physically Associated to YAP1, and IFT88 Modulates YAP1 Activity
(A) Co-IP experiment using HeLa cells transfected with IFT88-GFP, HA-Amotl1, and YAP1-Myc (l.e., long exposure; s.e., short exposure). Co-IP experiment using
HEK293 cells transfected with Flag-Amotl1 and IFT20-GFP. Endogenous levels of Yap1 are monitored.
(B) Schematic representation of the IFT88 auxin-inducible degron (AID) system.
(C) Western blot analysis of IFT88 AID DLD-1 cells after 2-h auxin treatment.
(C0) Western blot analysis of IFT88 and YAP1 degradation after auxin treatment (0, 0.5, and 2 h).
(D) Graph shows the increase in normalized YAP1 nuclear signal in cells treated with auxin (6 h) (Mann-Whitney, p value < 0.0001) (controls: 2 replicates, n = 204
cells; Auxin 2 h: 2 replicates, n = 261 cells).
(E) Graph shows the increase in YAP/WWTR1 (TAZ) nuclear signal in IFT88-siRNA (48 h)-treated MDCK cells (n = 5 replicates, average cell number analyzed for
each condition = 382; t test, p value, 0.004). Box and whiskers (5th–95th percentile). Outliers are represented as red dots (NT-siRNA) or blue squares (IFT88-
siRNA).
(legend continued on next page)
8 Cell Reports 32, 107932, July 21, 2020
Report
ll
OPEN ACCESS
Report
ll
OPEN ACCESSVerteporfin-treated ift88
/
 embryos presented fewer avcPE cell
numbers than nontreated ift88
/
 embryos. Interestingly, non-
treated controls and treated ift88 mutants did not show signifi-
cant differences in avcPE cell number. Thus, the increase in
avcPE cell number induced in the absence of Ift88 was rescued
by Verteporfin treatment. These results suggest that Ift88 re-
quires Yap1 activity to modulate the PE size restriction.
To explore a possible role of Yap1 in the regulation of
BMP signaling by IFT, we treated ift88
/
 and control (ift88+/+
and +/
) embryos with Verteporfin (Figure 3C) and quantified
BMP activity. Control embryos treated with Verteporfin showed
decreased p-smad1/5+myocardial cell number compared to un-
treated controls. Verteporfin-treated ift88
/
 embryos presented
fewer p-smad1/5+ myocardial cells than nontreated ift88
/
 em-
bryos. Remarkably, nontreated controls and treated ift88
/
 did
not show significant differences. Thus, the increase in BMP ac-
tivity induced in the absence of Ift88 was rescued by Verteporfin
treatment. Likewise, ISH performed in elipsa
/
 and control (eli-
psa+/+ and +/
) embryos treated with Verteporfin displayed
decreased bmp4 expression (Figure S6C). Furthermore, treat-
ment with the drug XAV939, a tankyrase inhibitor that sup-
pressed YAP-TEAD activity (Wang et al., 2015), also led to a
reduction of bmp4 expression (Figure S5H). These results sug-
gest that Ift88 and Ift54 require Yap1 activity to modulate BMP
signaling activity in the myocardium of the zebrafish. To confirm
the involvement of an increased Yap1 activity in the ift88
/

phenotype, we analyzed the avcPE in angiomotin like 2a
(amotl2a) mutants (Figures 3D and 3E). In zebrafish, Amotl2a
physically interacts with Yap1 leading to its cytoplasmic reten-
tion in a way that amotl2amutants show upregulated Yap1 activ-
ity (nuclear Yap1) (Agarwala et al., 2015; Nakajima et al., 2017).
Accordingly, amotl2a
/
 presented multiple avcPE formation
and increased bmp4 expression when compared to amotl2a+/+
embryos at 55 hpf. When treated with Verteporfin, amotl2a
/

showed decreased myocardial p-smad1/5+ and avcPE cell
numbers compared to untreated amotl2a
/
 embryos (Figures
3F and 3F0). Likewise, bmp4 expression at the atrioventricular
canal myocardium and at the venous pole were reduced in
amotl2a
/
 embryos treated with XAV939 (Figure 3G). Thus, Ver-
teporfin and XAV939 treatments rescued the increase in BMP
signaling induced in the absence of Amotl2a, suggesting it acts
through Yap1 activity. We next analyzed the yap1mutants (Agar-
wala et al., 2015). At 55 hpf, yap1
/
 showed decreasedmyocar-
dial cell number compared to their controls (Figure 2C). Besides,
fewer myocardial cells were p-smad1/5+ (Figure 2F). Surpris-
ingly, avcPE cell number was similar between yap1
/
 and their
controls (Figures S2B and S2B0). Altogether, these results sug-
gest that Ift88 and Ift54 modulate BMP signaling activity by tun-
ing Yap1-Tead activity in the myocardium of the zebrafish. To
assess whether the upregulation of Yap1 activity observed in
IFT zebrafish mutants was conserved in vertebrates, we
analyzed YAP1 localization in the PE of Ift20 KO and control
mice (Figures 3H–3H00). We quantified the proportion of nuclear(E0 ) Immunofluorescence (IF) confocal images of MDCK cells treated with NT- or
(F) Graph shows the increase in YAP/WWTR1 (TAZ) nuclear signal in IFT88-siRN
each condition = 252; t test, p value, 0.02). Box andwhiskers (5th–95th percentile).
(F0) IF confocal images of HeLa cells treated with NT- or IFT88-siRNA (48 h). DAPYAP1+ cells (i.e., cells with higher signal intensity inside the nu-
cleus than in the cytoplasm). The percentage of nuclear YAP1+
PE cells was higher in Ift20 KO than in control mice. In mouse
myocardium, YAP1 and AMOTL1 translocate to the nucleus
together to modulate cell response (Ragni et al., 2017). To
assess AMOTL1 localization in PE cells, we performed AMOTL1
IF. Consistent with the results obtained with an anti-YAP1 anti-
body, the percentage of nuclear AMOTL1+ cells (i.e., cells with
higher signal intensity inside the nucleus than in the cytoplasm)
was increased in the Ift20 KO when compared to control mice
(Figure S5I). We also found that the mutants showed an increase
in the percentage of YAP1-AMOTL1 double-positive cells (Fig-
ure 3H). Together, these data show that the increased PE size
in Ift20 KO mice is associated with increased YAP1 activity.
IFTs Interact with YAP1 and Regulate Its Activity
Taking advantage of cultured cells, we next explored the mech-
anism by which IFT proteins could regulate YAP1 activity. IF per-
formed in HeLa cells transfected with IFT88-GFP showed co-
localization between IFT88 and YAP1 in the cytoplasm (Figures
S6A–S6C). To confirm a potential physical interaction between
IFT proteins with YAP1, we performed co-immunoprecipitation
(co-IP) experiments using IFT88-GFP and YAP1-Myc in HeLa
cells but never obtained any clear interaction. Considering that
YAP1 often necessitates the scaffold protein Angiomotin-like 1
(Amotl1) (Ragni et al., 2017), we next performed co-IP experi-
ment using HA-Amotl1, IFT88-GFP, and YAP1-Myc in HeLa
cells. The experiments revealed a clear interaction between
YAP1, Amotl1, and IFT88 (Figure 4A). Similarly, physical interac-
tion between endogenous YAP1, Flag-Amotl1, and IFT20-GFP
was observed in transfected HEK293 cells (Figure 4A). These re-
sults reveal that IFT proteins, YAP1, and Amotl1 could function
as part of a complex involved in the functional modulation of
YAP1 activity in vivo.
To functionally assess the impact of endogenous IFT88 deple-
tion on YAP1 activity, we used a DLD-1 IFT88-auxin-inducible
degron (AID) cell line, in which a rapid degradation of IFT88 pro-
tein can be induced by auxin treatment (Figure 4B). Of note,
DLD-1 cells do not grow cilia, allowing us to explore the cyto-
plasmic, cilia-independent function of IFT88 (Lancaster et al.,
2011).
We observed that 2 h of auxin treatment led to IFT88 and YAP1
degradation. The rapid co-degradation further suggested that
both proteins physically interact (Figure 4C). We confirmed this
finding by performing shorter auxin treatments to study the pro-
gressive degradation of IFT88 and YAP1 (Figure 4C0). Impor-
tantly, as previously observed in vivo, longer depletion of endog-
enous IFT88 after 6 h of auxin treatment led to an increase in
nuclear YAP1 measured by IF (Figure 4D). Consistently, the nu-
clear/cytoplasmic YAP1 ratio was also increased (Figure S7A).
Additionally, we confirmed these results using HeLa and
MDCK cell lines where IFT88 function was inactivated by
IFT88-siRNA. The IFT88-siRNA efficiency was validated by IFIFT88-siRNA (48 h). DAPI (blue) and YAP/WWTR1 (TAZ) (white inverted LUT).
A (48 h)-treated HeLa cells (n = 5 replicates, average cell number analyzed for
Outliers are represented as red dots (NT-siRNA) or blue squares (IFT88-siRNA).
I (blue) and YAP/WWTR1 (TAZ) (white inverted LUT).
Cell Reports 32, 107932, July 21, 2020 9
Report
ll
OPEN ACCESSand western blot (Figures S7E–S7G). After 48 h of IFT88-siRNA
treatment, nuclear YAP1 signal was measured by IF (Figures
4E–4F0). The inactivation of IFT88 was accompanied with
increased nuclear YAP1 (Figures 4E and 4F). Altogether, these
data indicate that IFT88 can interact with YAP1 in the cytoplasm
and is involved inmodulating the activation of the Hippo pathway
effector YAP1.
To confirm that increased nuclear YAP1 is specific of IFT pro-
tein function, we performed IFT88 and IFT20 overexpression as-
says. While the transfection of pEGFP C1 did not alter the nu-
clear signal between GFP-positive and GFP-negative cells
(endogenous control) (Figure S6D), IFT88-GFP and IFT20-GFP
overexpression caused a decrease in nuclear signal (Figures
S7B–S7D). Together, these data confirm that IFT proteins modu-
late YAP1 activation.
DISCUSSION
IFT proteins have long been associated with ciliary functions in
developmental processes. For example, ift88 mutants display
a number of phenotypes reminiscent of ciliary defects such as
abnormal patterning of the neural tube, defects in the Hedgehog
pathway and left-right patterning (Huang and Schier, 2009). Ift88
has also been associated with planar cell polarity (Cao et al.,
2010) and cell division (Delaval et al., 2011; Taulet et al., 2017,
2019; Vitre et al., 2020). Our observations provide evidence for
a role of IFT complex B proteins in cardiogenesis. Mechanisti-
cally, IFT proteins are best known for their function in ciliary
transport (Ocbina et al., 2011). Here, we describe an unexpected
interaction between the ciliary machinery proteins and a potent
mechanosensing pathway, the Hippo pathway. The hippo
effector YAP1 is known to have essential roles in cancer (Zanco-
nato et al., 2016), regeneration (Bassat et al., 2017; Leach et al.,
2017; Oh et al., 2018), and organ size control (Artap et al., 2018;
Thompson and Sahai, 2015; Yu et al., 2015). Several mecha-
nisms have been shown to regulate the shuttling of YAP1 into
the nucleus, including phosphorylation by Hippo kinases.
Recent studies show that YAP1 is mechanosensitive and that
force applied to the nucleus can directly drive YAP1 nuclear
translocation (Elosegui-Artola et al., 2017; Sun et al., 2014).
Additionally, Angiomotin (AMOT) has been shown to interact
physically with YAP1 and act as a buffering factor sequestering
YAP1 in the cytoplasm (Zhao et al., 2011). Nevertheless,
AMOTL1 has also been shown to co-localize with YAP1 in the
nucleus (Ragni et al., 2017; Yi et al., 2013), demonstrating that
YAP1 subcellular localization is highly regulated by Motin family
proteins. Our results indicate that IFT complex B proteins are
also involved in regulating YAP1 localization. We demonstrate
that IFT88 interacts biochemically with YAP1 and both co-
localize in the cytoplasm. We did not study TAZ, the other Hippo
effector that is known to act with YAP1 (Piccolo et al., 2014), and
we cannot draw general conclusions on the role of IFT88 on all
the known Hippo effectors. Nevertheless, our work suggests
alternative ways to interpret Ift88 mutant phenotypes, which
are often interpreted based on polarity or cilia function issues,
and, more generally, phenotypes of other mutants with abnormal
IFT complex B proteins. Our working model is that Ift88 partici-
pates in sequestering YAP1 away from the nucleus using its10 Cell Reports 32, 107932, July 21, 2020cargo transport activity. Other cilia-related proteins, such as ki-
nesin2 and IFT complex A proteins, have been shown to promote
nuclear localization of b-catenin during Wnt signaling in
Drosophila (Vuong et al., 2018), further suggesting that proteins
identified for their ciliary transport functions are not always
limited to that function. Future work will reveal the mechanism
by which Ift88 limits nuclear translocation of YAP1 and address
whether IFT complex A proteins play a role.
Importantly, while our study points toward a noncanonical func-
tion for IFT complex B proteins, our results do not exclude a role
for primary cilia in PE formation. We found that iguana/dzip mu-
tants display an avcPE phenotype, suggesting that primary cilia
function is required for PE morphogenesis. The HH pathway is
often associatedwith ciliary function. To date, a number of studies
suggest the proepicardium and epicardium formation are not
regulated by Hedgehog signaling (Rudat et al., 2013; Sugimoto
et al., 2017). Indeed, dissection of zebrafish shha function in the
PE and epicardium using a tcf21:CreER, a well-established PE
and epicardial tissue driver (Robb et al., 1998), does not affect
PE formation (Sugimoto et al., 2017). Besides, expression of
Shh, Dhh, Ihh, and Ptch1 was neither detected in the mouse PE
nor in the epicardium at subsequent stages (Rudat et al., 2013).
Thus, primary cilia certainly operate independently of the HH
pathway andYap1,which is not altered indzipmutants, in the pro-
cess. Our work further highlights the important role of the BMP
pathway in PE formation (Andre´s-Delgado et al., 2019; Liu and
Stainier, 2010). Primary cilia modulate BMP pathway and, more
generally, TGF beta signaling pathways (Mo¨nnich et al., 2018; Vil-
lalobos et al., 2019). In endothelial cells, primary cilia modulate
angiogenesis by altering BMP signaling (Vion et al., 2018). We
speculate the situation is different in PE cells, where BMP is acti-
vated downstream of YAP1, and where IFT88 helps to limit YAP1
and BMP signaling. Similarly, Hippo signaling determines the
number of atrial myocardial cells that originate from the SHF by
modulating BMP signaling (Fukui et al., 2018). Interestingly, lack
of Ift88 did not affect Isl1 expression at early stages, suggesting
Ift88 modulates Yap1-BMP axis at later stages during PE and
atrial myocardium development. While yap1 zebrafish mutants
display decreased myocardial cell number, they surprisingly
display normal avcPE cell number. The zebrafish yap1 mutants
are able to develop and to reach adulthood, which can be due
to the compensation between Yap1 and Wwtr1 previously
described (Miesfeld et al., 2015). We speculate that similar
compensation may be at work in the PE as well. Importantly, we
found phenotypic difference between mouse and fish, especially
regarding the heart size, which is not affected in mouse. More
work will be needed to understand these interspecific differences
as well as the potential mechanisms of compensation involved if
any. Zebrafish, ift88 and elipsa, and mouse, Ift20 and Ift88, mu-
tants are lethal at 6 dpf and E10.5–E11.5, respectively, making
impossible the study of cardiovascular defects at later stages
without conditional gene inactivation approaches.
In summary, our study reports the role of IFT complex B pro-
teins during PE development by modulating YAP1 activity inde-
pendently of any cilia function. Linking IFT with YAP1 activity
might have important implications for understanding the etiology
of ciliopathies during cardiogenesis and for the interpretation of
ciliary defects in IFT mutants.
Report
ll
OPEN ACCESSSTAR+METHODS
Detailed methods are provided in the online version of this paper
and include the following:
d KEY RESOURCES TABLE
d RESOURCE AVAILABILITYB Lead Contact
B Materials Availability
B Data and Code Availability
d EXPERIMENTAL MODEL AND SUBJECT DETAILS
B Zebrafish (ZF) Models
B Mouse Models
d METHOD DETAILS
B In Vivo Imaging
B ZF Treatments
B ZF Immunofluorescence
B In Situ Hybridization (ISH)
B Whole Mount Immunofluorescence in the Mouse
B Immunofluorescence on Cryosections in the Mouse
B Cell Culture, siRNAs and Transfection
B Generation of DLD-1 IFT88-AID Targeted Cells
B Immunofluorescence on Cells
B Lysates and Immunoblotting
d IMMUNOPRECIPITATION (IP) ASSAYS
B HEK293 Cells
B HeLa Cells
B ZF avcPE Cell Quantification
B ZF Myocardial Cell Quantification
B Mice PE Volume Analysis
B Mice Yap1- and Amotl1-Positive Cell Quantifications
B Nuclear Yap Quantifications on Cells
B Nuclear/cytoplasmic Yap1 Ratio
d QUANTIFICATION AND STATISTICAL ANALYSISSUPPLEMENTAL INFORMATION
Supplemental Information can be found online at https://doi.org/10.1016/j.
celrep.2020.107932.
ACKNOWLEDGMENTS
We thank P. Lamperti, E. Steed, A. Bhat, D. Riveline, S. Harlepp, J. Godin, A.
Benmerah, and the Vermot laboratory for discussion and thoughtful comments
on the manuscript, in particular R. Chow for her help with editing. We thank S.
Roth for technical help. We thank G. Pazour, M. Faucourt, and N. Spassky for
advice, for the gift of plasmids, and for providing the Ift20 and Ift88 mutant
mouse lines, and C. Cimper, L. Bombardelli, and C. Shea for technical assis-
tance. We are grateful to the IGBMC fish facility, the IGBMC imaging center,
and the imaging platforms of the SFR Necker. This project has received fund-
ing from the European Union’s Horizon 2020 Research and Innovation Pro-
gramme under the Marie Sklodowska-Curie Grant Agreement No. 708312
(M.P.) and from the European Research Council (ERC) under the European
Union’s Horizon 2020 Research and Innovation Programme: GA No. 682938
(J.V.). This work was supported by FRM (DEQ20140329553), by ANR (ANR-
15-CE13-0015–liveheart, ANR- SNF310030E-164245-forcinregeneration),
and by the Grant ANR-10-LABX-0030-INRT, a French State fund managed
by the Agence Nationale de la Recherche under the frame program Investisse-
ments d’Avenir labeled ANR-10-IDEX-0002-02. B.D.’s teamwas supported by
ANR-12-CHEX-005 and CNRS. S.M.M.’s teamwas supported by core funding
from the Institut Imagine, Institut Pasteur, Inserm, Universite´ Paris Descartes,and a grant from the AFM-Te´le´thon (Trampoline 18727). T.L. was funded by the
ED515 (1691/2014). L.O.L. is supported by the European Commission (H2020-
MSCA-ITN-2016 European Industrial Doctorate 4DHeart 722427).AUTHOR CONTRIBUTIONS
Conceptualization, M.P. and J.V.; Data Curation, M.P., L.O.L., K.J., and T.L.;
Formal Analysis, M.P.; Funding Acquisition, M.P., B.D., S.M.M., and J.V.;
Investigation, M.P., L.O.L., K.J., T.L., B.V., D.H., and L.G.; Methodology,
M.P., K.J., T.L., B.V., B.D., S.M.M., and J.V.; Project Administration, M.P.
and J.V.; Resources, C.D., V.L., I.S., B.D., S.M.M., and J.V.; Supervision,
J.V.; Validation, M.P., B.D., S.M.M., and J.V.; Visualization, M.P.; Writing –
Original Draft, M.P. and J.V.; Writing – Review & Editing, M.P., L.O.L., K.J.,
T.L., D.H., C.D., B.V., N.M., V.L., B.D., S.M.M., and J.V.DECLARATION OF INTERESTS
The authors declare no competing interests.
Received: October 24, 2019
Revised: April 30, 2020
Accepted: June 29, 2020
Published: July 21, 2020REFERENCES
Acharya, A., Baek, S.T., Huang, G., Eskiocak, B., Goetsch, S., Sung, C.Y.,
Banfi, S., Sauer, M.F., Olsen, G.S., Duffield, J.S., et al. (2012). The bHLH tran-
scription factor Tcf21 is required for lineage-specific EMT of cardiac fibroblast
progenitors. Development 139, 2139–2149.
Agarwala, S., Duquesne, S., Liu, K., Boehm, A., Grimm, L., Link, S., Ko¨nig, S.,
Eimer, S., Ronneberger, O., and Lecaudey, V. (2015). Amotl2a interacts with
the Hippo effector Yap1 and the Wnt/b-catenin effector Lef1 to control tissue
size in zebrafish. eLife 4, e08201.
Andre´s-Delgado, L., Ernst, A., Galardi-Castilla, M., Bazaga, D., Peralta, M.,
M€unch, J., Gonza´lez-Rosa, J.M., Marques, I., Tessadori, F., de la Pompa,
J.L., et al. (2019). Actin dynamics and the Bmp pathway drive apical extrusion
of proepicardial cells. Development 146, dev174961.
Artap, S., Manderfield, L.J., Smith, C.L., Poleshko, A., Aghajanian, H., See, K.,
Li, L., Jain, R., and Epstein, J.A. (2018). Endocardial Hippo signaling regulates
myocardial growth and cardiogenesis. Dev. Biol. 440, 22–30.
Bassat, E., Mutlak, Y.E., Genzelinakh, A., Shadrin, I.Y., Baruch Umansky, K.,
Yifa, O., Kain, D., Rajchman, D., Leach, J., Riabov Bassat, D., et al. (2017).
The extracellular matrix protein agrin promotes heart regeneration in mice. Na-
ture 547, 179–184.
Berbari, N.F., O’Connor, A.K., Haycraft, C.J., and Yoder, B.K. (2009). The pri-
mary cilium as a complex signaling center. Curr. Biol. 19, R526–R535.
Boin, A., Couvelard, A., Couderc, C., Brito, I., Filipescu, D., Kalamarides, M.,
Bedossa, P., De Koning, L., Danelsky, C., Dubois, T., et al. (2014). Proteomic
screening identifies a YAP-driven signaling network linked to tumor cell prolif-
eration in human schwannomas. Neuro-oncol. 16, 1196–1209.
Borovina, A., Superina, S., Voskas, D., and Ciruna, B. (2010). Vangl2 directs
the posterior tilting and asymmetric localization of motile primary cilia. Nat.
Cell Biol. 12, 407–412.
Bratt, A., Wilson, W.J., Troyanovsky, B., Aase, K., Kessler, R., Van Meir, E.G.,
and Holmgren, L. (2002). Angiomotin belongs to a novel protein family with
conserved coiled-coil and PDZ binding domains. Gene 298, 69–77.
Buckingham, M., Meilhac, S., and Zaffran, S. (2005). Building the mammalian
heart from two sources of myocardial cells. Nat. Rev. Genet. 6, 826–835.
Cao, Y., Park, A., and Sun, Z. (2010). Intraflagellar transport proteins are
essential for cilia formation and for planar cell polarity. J. Am. Soc. Nephrol.
21, 1326–1333.Cell Reports 32, 107932, July 21, 2020 11
Report
ll
OPEN ACCESSCarmona, R., Gonza´lez-Iriarte, M., Pe´rez-Pomares, J.M., and Mun˜oz-Cha´puli,
R. (2001). Localization of theWilm’s tumour proteinWT1 in avian embryos. Cell
Tissue Res. 303, 173–186.
Cong, L., Ran, F.A., Cox, D., Lin, S., Barretto, R., Habib, N., Hsu, P.D., Wu, X.,
Jiang,W., Marraffini, L.A., and Zhang, F. (2013). Multiplex genome engineering
using CRISPR/Cas systems. Science 339, 819–823.
de Pater, E., Clijsters, L., Marques, S.R., Lin, Y.F., Garavito-Aguilar, Z.V., Ye-
lon, D., and Bakkers, J. (2009). Distinct phases of cardiomyocyte differentia-
tion regulate growth of the zebrafish heart. Development 136, 1633–1641.
Delaval, B., Bright, A., Lawson, N.D., and Doxsey, S. (2011). The cilia protein
IFT88 is required for spindle orientation in mitosis. Nat. Cell Biol. 13, 461–468.
DeRan, M., Yang, J., Shen, C.-H., Peters, E.C., Fitamant, J., Chan, P., Hsieh,
M., Zhu, S., Asara, J.M., Zheng, B., et al. (2014). Energy stress regulates hippo-
YAP signaling involving AMPK-mediated regulation of angiomotin-like 1 pro-
tein. Cell Rep. 9, 495–503.
Elosegui-Artola, A., Andreu, I., Beedle, A.E.M., Lezamiz, A., Uroz, M., Kosmal-
ska, A.J., Oria, R., Kechagia, J.Z., Rico-Lastres, P., Le Roux, A.-L., et al. (2017).
Force triggers YAP nuclear entry by regulating transport across nuclear pores.
Cell 171, 1397–1410.e14.
Ferreira, RR, Fukui, H, Chow, R, Vilfan, A, Vermot, J (Eds.), 2019 Jul. The cilium
as a force sensor-myth versus reality. J Cell Sci 30132, jcs213496.
Finetti, F., Paccani, S.R., Riparbelli, M.G., Giacomello, E., Perinetti, G., Pazour,
G.J., Rosenbaum, J.L., and Baldari, C.T. (2009). Intraflagellar transport is
required for polarized recycling of the TCR/CD3 complex to the immune syn-
apse. Nat. Cell Biol. 11, 1332–1339.
Follit, J.A., Tuft, R.A., Fogarty, K.E., and Pazour, G.J. (2006). The intraflagellar
transport protein IFT20 is associatedwith theGolgi complex and is required for
cilia assembly. Mol. Biol. Cell 17, 3781–3792.
Frank, V., Habbig, S., Bartram, M.P., Eisenberger, T., Veenstra-Knol, H.E.,
Decker, C., Boorsma, R.A.C., Go¨bel, H., N€urnberg, G., Griessmann, A., et al.
(2013). Mutations in NEK8 link multiple organ dysplasia with altered Hippo sig-
nalling and increased c-MYC expression. Hum. Mol. Genet. 22, 2177–2185.
Fukui, H., Miyazaki, T., Chow, R.W.-Y., Ishikawa, H., Nakajima, H., Vermot, J.,
and Mochizuki, N. (2018). Hippo signaling determines the number of venous
pole cells that originate from the anterior lateral plate mesoderm in zebrafish.
eLife 7, e29106.
Galgano, D., Onnis, A., Pappalardo, E., Galvagni, F., Acuto, O., and Baldari,
C.T. (2017). The T cell IFT20 interactome reveals new players in immune syn-
apse assembly. J. Cell Sci. 130, 1110–1121.
Gonza´lez-Rosa, J.M., Peralta, M., and Mercader, N. (2012). Pan-epicardial
lineage tracing reveals that epicardiumderived cells give rise tomyofibroblasts
and perivascular cells during zebrafish heart regeneration. Dev. Biol. 370,
173–186.
Grampa, V., Delous, M., Zaidan, M., Odye, G., Thomas, S., Elkhartoufi, N., Fil-
hol, E., Niel, O., Silbermann, F., Lebreton, C., et al. (2016). Novel NEK8 muta-
tions cause severe syndromic renal cystic dysplasia through YAP dysregula-
tion. PLoS Genet. 12, e1005894.
Habbig, S., Bartram, M.P., Sa¨gm€uller, J.G., Griessmann, A., Franke, M.,
M€uller, R.-U., Schwarz, R., Hoehne, M., Bergmann, C., Tessmer, C., et al.
(2012). The ciliopathy disease protein NPHP9 promotes nuclear delivery and
activation of the oncogenic transcriptional regulator TAZ. Hum. Mol. Genet.
21, 5528–5538.
Haycraft, C.J., Zhang, Q., Song, B., Jackson, W.S., Detloff, P.J., Serra, R., and
Yoder, B.K. (2007). Intraflagellar transport is essential for endochondral bone
formation. Development 134, 307–316.
He, M., Subramanian, R., Bangs, F., Omelchenko, T., Liem, K.F., Jr., Kapoor,
T.M., and Anderson, K.V. (2014). The kinesin-4 protein Kif7 regulates mamma-
lian Hedgehog signalling by organizing the cilium tip compartment. Nat. Cell
Biol. 16, 663–672.
Holland, A.J., Fachinetti, D., Han, J.S., and Cleveland, D.W. (2012). Inducible,
reversible system for the rapid and complete degradation of proteins in
mammalian cells. Proc. Natl. Acad. Sci. USA 109, E3350–E3357.12 Cell Reports 32, 107932, July 21, 2020Hua, K., and Ferland, R.J. (2018). Primary cilia proteins: ciliary and extraciliary
sites and functions. Cell. Mol. Life Sci. 75, 1521–1540.
Huang, P., and Schier, A.F. (2009). Dampened Hedgehog signaling but normal
Wnt signaling in zebrafish without cilia. Development 136, 3089–3098.
Ishii, Y., Garriock, R.J., Navetta, A.M., Coughlin, L.E., and Mikawa, T. (2010).
BMP signals promote proepicardial protrusion necessary for recruitment of
coronary vessel and epicardial progenitors to the heart. Dev. Cell 19, 307–316.
Jonassen, J.A., San Agustin, J., Follit, J.A., and Pazour, G.J. (2008). Deletion of
IFT20 in the mouse kidney causes misorientation of the mitotic spindle and
cystic kidney disease. J. Cell Biol. 183, 377–384.
Katz, T.C., Singh, M.K., Degenhardt, K., Rivera-Feliciano, J., Johnson, R.L.,
Epstein, J.A., and Tabin, C.J. (2012). Distinct compartments of the proepicar-
dial organ give rise to coronary vascular endothelial cells. Dev. Cell 22,
639–650.
Kikuchi, K., and Poss, K.D. (2012). Cardiac regenerative capacity and mecha-
nisms. Annu. Rev. Cell Dev. Biol. 28, 719–741.
Kikuchi, K., Gupta, V., Wang, J., Holdway, J.E., Wills, A.A., Fang, Y., and Poss,
K.D. (2011). tcf21+ epicardial cells adopt non-myocardial fates during zebra-
fish heart development and regeneration. Development 138, 2895–2902.
Kim, H.R., Richardson, J., van Eeden, F., and Ingham, P.W. (2010). Gli2a pro-
tein localization reveals a role for Iguana/DZIP1 in primary ciliogenesis and a
dependence of Hedgehog signal transduction on primary cilia in the zebrafish.
BMC Biol. 8, 65.
Kim, M., Kim, M., Lee, M.-S., Kim, C.-H., and Lim, D.-S. (2014). The MST1/2-
SAV1 complex of the Hippo pathway promotes ciliogenesis. Nat. Commun. 5,
5370.
Klena, N., Gabriel, G., Liu, X., Yagi, H., Li, Y., Chen, Y., Zahid, M., Tobita, K.,
Leatherbury, L., Pazour, G., and Lo, C.W. (2016). Role of cilia and left-right
patterning in congenital heart disease. Etiology and Morphogenesis of
Congenital Heart Disease: From Gene Function and Cellular Interaction to
Morphology (Springer), p. 1.
Lai, J.K.H., Collins, M.M., Uribe, V., Jime´nez-Amilburu, V., G€unther, S., Mai-
schein, H.-M., and Stainier, D.Y.R. (2018). The Hippo pathway effector
Wwtr1 regulates cardiac wall maturation in zebrafish. Development 145,
dev159210.
Lallemand, Y., Luria, V., Haffner-Krausz, R., and Lonai, P. (1998). Maternally
expressed PGK-Cre transgene as a tool for early and uniform activation of
the Cre site-specific recombinase. Transgenic Res. 7, 105–112.
Lancaster, M.A., Schroth, J., and Gleeson, J.G. (2011). Subcellular spatial
regulation of canonical Wnt signalling at the primary cilium. Nat. Cell Biol.
13, 700–707.
Leach, J.P., Heallen, T., Zhang, M., Rahmani, M., Morikawa, Y., Hill, M.C., Se-
gura, A., Willerson, J.T., and Martin, J.F. (2017). Hippo pathway deficiency re-
verses systolic heart failure after infarction. Nature 550, 260–264.
Li, Y., Klena, N.T., Gabriel, G.C., Liu, X., Kim, A.J., Lemke, K., Chen, Y., Chat-
terjee, B., Devine, W., Damerla, R.R., et al. (2015). Global genetic analysis in
mice unveils central role for cilia in congenital heart disease. Nature 521,
520–524.
Limana, F., Capogrossi, M.C., and Germani, A. (2011). The epicardium in car-
diac repair: from the stem cell view. Pharmacol. Ther. 129, 82–96.
Liu, J., and Stainier, D.Y.R. (2010). Tbx5 and Bmp signaling are essential for
proepicardium specification in zebrafish. Circ. Res. 106, 1818–1828.
Liu-Chittenden, Y., Huang, B., Shim, J.S., Chen, Q., Lee, S.-J., Anders, R.A.,
Liu, J.O., and Pan, D. (2012). Genetic and pharmacological disruption of the
TEAD-YAP complex suppresses the oncogenic activity of YAP. Genes Dev.
26, 1300–1305.
Ma¨nner, J., Schlueter, J., and Brand, T. (2005). Experimental analyses of the
function of the proepicardium using a new microsurgical procedure to induce
loss-of-proepicardial-function in chick embryos. Dev. Dyn. 233, 1454–1463.
Miesfeld, J.B., and Link, B.A. (2014). Establishment of transgenic lines to
monitor and manipulate Yap/Taz-Tead activity in zebrafish reveals both evolu-
tionarily conserved and divergent functions of the Hippo pathway. Mech. Dev.
133, 177–188.
Report
ll
OPEN ACCESSMiesfeld, J.B., Gestri, G., Clark, B.S., Flinn, M.A., Poole, R.J., Bader, J.R., Be-
sharse, J.C., Wilson, S.W., and Link, B.A. (2015). Yap and Taz regulate retinal
pigment epithelial cell fate. Development 142, 3021–3032.
Mikawa, T., and Fischman, D.A. (1992). Retroviral analysis of cardiac morpho-
genesis: discontinuous formation of coronary vessels. Proc. Natl. Acad. Sci.
USA 89, 9504–9508.
Mikawa, T., and Gourdie, R.G. (1996). Pericardial mesoderm generates a pop-
ulation of coronary smooth muscle cells migrating into the heart along with
ingrowth of the epicardial organ. Dev. Biol. 174, 221–232.
Mommersteeg, M.T.M., Domı´nguez, J.N., Wiese, C., Norden, J., de Gier-de
Vries, C., Burch, J.B.E., Kispert, A., Brown, N.A., Moorman, A.F.M., and Chris-
toffels, V.M. (2010). The sinus venosus progenitors separate and diversify from
the first and second heart fields early in development. Cardiovasc. Res. 87,
92–101.
Mo¨nnich, M., Borgeskov, L., Breslin, L., Jakobsen, L., Rogowski, M., Doganli,
C., Schrøder, J.M., Mogensen, J.B., Blinkenkjær, L., Harder, L.M., et al. (2018).
CEP128 localizes to the subdistal appendages of themother centriole and reg-
ulates TGF-b/BMP signaling at the primary cilium. Cell Rep. 22, 2584–2592.
Moore, A.W., McInnes, L., Kreidberg, J., Hastie, N.D., and Schedl, A. (1999).
YAC complementation shows a requirement for Wt1 in the development of
epicardium, adrenal gland and throughout nephrogenesis. Development
126, 1845–1857.
Nakajima, H., Yamamoto, K., Agarwala, S., Terai, K., Fukui, H., Fukuhara, S.,
Ando, K.,Miyazaki, T., Yokota, Y., Schmelzer, E., et al. (2017). Flow-dependent
endothelial YAP regulation contributes to vessel maintenance. Dev. Cell 40,
523–536.e6.
Ocbina, P.J.R., Eggenschwiler, J.T., Moskowitz, I., and Anderson, K.V. (2011).
Complex interactions between genes controlling trafficking in primary cilia.
Nat. Genet. 43, 547–553.
Oh, S.-H., Swiderska-Syn, M., Jewell, M.L., Premont, R.T., and Diehl, A.M.
(2018). Liver regeneration requires Yap1-TGFb-dependent epithelial-mesen-
chymal transition in hepatocytes. J. Hepatol. 69, 359–367.
Omori, Y., Zhao, C., Saras, A., Mukhopadhyay, S., Kim,W., Furukawa, T., Sen-
gupta, P., Veraksa, A., and Malicki, J. (2008). Elipsa is an early determinant of
ciliogenesis that links the IFT particle to membrane-associated small GTPase
Rab8. Nat. Cell Biol. 10, 437–444.
Pazour, G.J., Dickert, B.L., Vucica, Y., Seeley, E.S., Rosenbaum, J.L., Witman,
G.B., andCole, D.G. (2000).Chlamydomonas IFT88 and itsmouse homologue,
polycystic kidney disease gene tg737, are required for assembly of cilia and
flagella. J. Cell Biol. 151, 709–718.
Pei, Z., Bai, Y., and Schmitt, A.P. (2010). PIV5 M protein interaction with host
protein angiomotin-like 1. Virology 397, 155–166.
Peralta, M., Steed, E., Harlepp, S., Gonza´lez-Rosa, J.M., Monduc, F., Ariza-
Cosano, A., Corte´s, A., Rayo´n, T., Go´mez-Skarmeta, J.-L., Zapata, A., et al.
(2013). Heartbeat-driven pericardiac fluid forces contribute to epicardium
morphogenesis. Curr. Biol. 23, 1726–1735.
Piccolo, S., Dupont, S., and Cordenonsi, M. (2014). The biology of YAP/TAZ:
hippo signaling and beyond. Physiol. Rev. 94, 1287–1312.
Ragni, C.V., Diguet, N., Le Garrec, J.-F., Novotova, M., Resende, T.P., Pop, S.,
Charon, N., Guillemot, L., Kitasato, L., Badouel, C., et al. (2017). Amotl1 medi-
ates sequestration of the Hippo effector Yap1 downstream of Fat4 to restrict
heart growth. Nat. Commun. 8, 14582.
Reiter, J.F., and Leroux, M.R. (2017). Genes and molecular pathways under-
pinning ciliopathies. Nat. Rev. Mol. Cell Biol. 18, 533–547.
Robb, L., Mifsud, L., Hartley, L., Biben, C., Copeland, N.G., Gilbert, D.J., Jen-
kins, N.A., and Harvey, R.P. (1998). epicardin: a novel basic helix-loop-helix
transcription factor gene expressed in epicardium, branchial arch myoblasts,
and mesenchyme of developing lung, gut, kidney, and gonads. Dev. Dyn. 213,
105–113.
Robert, A., Margall-Ducos, G., Guidotti, J.-E., Bre´gerie, O., Celati, C., Bre´chot,
C., and Desdouets, C. (2007). The intraflagellar transport component IFT88/
polaris is a centrosomal protein regulating G1-S transition in non-ciliated cells.
J. Cell Sci. 120, 628–637.Rosenbaum, J.L., and Witman, G.B. (2002). Intraflagellar transport. Nat. Rev.
Mol. Cell Biol. 3, 813–825.
Rudat, C., Norden, J., Taketo, M.M., and Kispert, A. (2013). Epicardial function
of canonical Wnt-, Hedgehog-, Fgfr1/2-, and Pdgfra-signalling. Cardiovasc.
Res. 100, 411–421.
San Agustin, J.T., Klena, N., Granath, K., Panigrahy, A., Stewart, E., Devine,
W., Strittmatter, L., Jonassen, J.A., Liu, X., Lo, C.W., and Pazour, G.J.
(2016). Genetic link between renal birth defects and congenital heart disease.
Nat. Commun. 7, 11103.
Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M., Pietzsch,
T., Preibisch, S., Rueden, C., Saalfeld, S., Schmid, B., et al. (2012). Fiji: an
open-source platform for biological-image analysis. Nat. Methods 9, 676–682.
Schlueter, J., Ma¨nner, J., and Brand, T. (2006). BMP is an important regulator
of proepicardial identity in the chick embryo. Dev. Biol. 295, 546–558.
Singh, A., Ramesh, S., Cibi, D.M., Yun, L.S., Li, J., Li, L., Manderfield, L.J., Ol-
son, E.N., Epstein, J.A., and Singh,M.K. (2016). Hippo signalingmediators Yap
and Taz are required in the epicardium for coronary vasculature development.
Cell Rep. 15, 1384–1393.
Slough, J., Cooney, L., and Brueckner, M. (2008). Monocilia in the embryonic
mouse heart suggest a direct role for cilia in cardiac morphogenesis. Dev. Dyn.
237, 2304–2314.
Sugimoto, K., Hui, S.P., Sheng, D.Z., and Kikuchi, K. (2017). Dissection of ze-
brafish shha function using site-specific targeting with a Cre-dependent ge-
netic switch. eLife 6, e24635.
Sun, Z., Amsterdam, A., Pazour, G.J., Cole, D.G., Miller, M.S., and Hopkins, N.
(2004). A genetic screen in zebrafish identifies cilia genes as a principal cause
of cystic kidney. Development 131, 4085–4093.
Sun, Y., Yong, K.M.A., Villa-Diaz, L.G., Zhang, X., Chen, W., Philson, R., Weng,
S., Xu, H., Krebsbach, P.H., and Fu, J. (2014). Hippo/YAP-mediated rigidity-
dependent motor neuron differentiation of human pluripotent stem cells.
Nat. Mater. 13, 599–604.
Taschner, M., Bhogaraju, S., and Lorentzen, E. (2012). Architecture and func-
tion of IFT complex proteins in ciliogenesis. Differentiation 83, S12–S22.
Taulet, N., Vitre, B., Anguille, C., Douanier, A., Rocancourt, M., Taschner, M.,
Lorentzen, E., Echard, A., and Delaval, B. (2017). IFT proteins spatially control
the geometry of cleavage furrow ingression and lumen positioning. Nat. Com-
mun. 8, 1928.
Taulet, N., Douanier, A., Vitre, B., Anguille, C., Maurin, J., Dromard, Y., Geor-
get, V., and Delaval, B. (2019). IFT88 controls NuMA enrichment at k-fibers
minus-ends to facilitate their re-anchoring into mitotic spindles. Sci. Rep. 9,
10311.
Tay, S.Y., Yu, X., Wong, K.N., Panse, P., Ng, C.P., and Roy, S. (2010). The
iguana/DZIP1 protein is a novel component of the ciliogenic pathway essential
for axonemal biogenesis. Dev. Dyn. 239, 527–534.
Thisse, C., and Thisse, B. (2008). High-resolution in situ hybridization to whole-
mount zebrafish embryos. Nat Protoc 3, 59–69.
Thompson, B.J., and Sahai, E. (2015). MST kinases in development and dis-
ease. J. Cell Biol. 210, 871–882.
Tsujikawa, M., and Malicki, J. (2004). Intraflagellar transport genes are essen-
tial for differentiation and survival of vertebrate sensory neurons. Neuron 42,
703–716.
van Wijk, B., van den Berg, G., Abu-Issa, R., Barnett, P., van der Velden, S.,
Schmidt, M., Ruijter, J.M., Kirby, M.L., Moorman, A.F.M., and van den Hoff,
M.J.B. (2009). Epicardium and myocardium separate from a common precur-
sor pool by crosstalk between bone morphogenetic protein- and fibroblast
growth factor-signaling pathways. Circ. Res. 105, 431–441.
Vertii, A., Bright, A., Delaval, B., Hehnly, H., and Doxsey, S. (2015). New fron-
tiers: discovering cilia-independent functions of cilia proteins. EMBO Rep. 16,
1275–1287.
Villalobos, E., Criollo, A., Schiattarella, G.G., Altamirano, F., French, K.M.,May,
H.I., Jiang, N., Nguyen, N.U.N., Romero, D., Roa, J.C., et al. (2019). Fibroblast
primary cilia are required for cardiac fibrosis. Circulation 139, 2342–2357.Cell Reports 32, 107932, July 21, 2020 13
Report
ll
OPEN ACCESSVion, A.-C., Alt, S., Klaus-Bergmann, A., Szymborska, A., Zheng, T., Perovic,
T., Hammoutene, A., Oliveira, M.B., Bartels-Klein, E., Hollfinger, I., et al.
(2018). Primary cilia sensitize endothelial cells to BMP and prevent excessive
vascular regression. J. Cell Biol. 217, 1651–1665.
Vitre, B., Taulet, N., Guesdon, A., Douanier, A., Dosdane, A., Cisneros, M.,
Maurin, J., Hettinger, S., Anguille, C., Taschner, M., et al. (2020). IFT proteins
interact with HSET to promote supernumerary centrosome clustering in
mitosis. EMBO Rep. 21, e49234.
Vuong, L.T., Iomini, C., Balmer, S., Esposito, D., Aaronson, S.A., and Mlodzik,
M. (2018). Kinesin-2 and IFT-A act as a complex promoting nuclear localization
of b-catenin during Wnt signalling. Nat. Commun. 9, 5304.
Wang, W., Huang, J., and Chen, J. (2011). Angiomotin-like proteins associate
with and negatively regulate YAP1. J. Biol. Chem. 286, 4364–4370.
Wang, W., Li, N., Li, X., Tran, M.K., Han, X., and Chen, J. (2015). Tankyrase in-
hibitors target YAP by stabilizing angiomotin family proteins. Cell Rep. 13,
524–532.
Witzel, H.R., Jungblut, B., Choe, C.P., Crump, J.G., Braun, T., and Dobreva, G.
(2012). The LIM protein Ajuba restricts the second heart field progenitor pool
by regulating Isl1 activity. Dev. Cell 23, 58–70.
Wu, S.-P., Dong, X.-R., Regan, J.N., Su, C., and Majesky, M.W. (2013). Tbx18
regulates development of the epicardium and coronary vessels. Dev. Biol. 383,
307–320.
Xin, M., Kim, Y., Sutherland, L.B., Qi, X., McAnally, J., Schwartz, R.J., Richard-
son, J.A., Bassel-Duby, R., and Olson, E.N. (2011). Regulation of insulin-like14 Cell Reports 32, 107932, July 21, 2020growth factor signaling by Yap governs cardiomyocyte proliferation and em-
bryonic heart size. Sci. Signal. 4, ra70.
Yi, C., Shen, Z., Stemmer-Rachamimov, A., Dawany, N., Troutman, S., Showe,
L.C., Liu, Q., Shimono, A., Sudol, M., Holmgren, L., et al. (2013). The p130 iso-
form of angiomotin is required for Yap-mediated hepatic epithelial cell prolifer-
ation and tumorigenesis. Sci. Signal. 6, ra77.
Yu, F.-X., Zhao, B., and Guan, K.-L. (2015). Hippo pathway in organ size con-
trol, tissue homeostasis, and cancer. Cell 163, 811–828.
Zanconato, F., Cordenonsi, M., and Piccolo, S. (2016). YAP/TAZ at the roots of
cancer. Cancer Cell 29, 783–803.
Zhao, B., Ye, X., Yu, J., Li, L., Li, W., Li, S., Yu, J., Lin, J.D., Wang, C.Y., Chin-
naiyan, A.M., et al. (2008). TEADmediates YAP-dependent gene induction and
growth control. Genes Dev. 22, 1962–1971.
Zhao, B., Li, L., Lu, Q., Wang, L.H., Liu, C.-Y., Lei, Q., and Guan, K.-L. (2011).
Angiomotin is a novel Hippo pathway component that inhibits YAP oncopro-
tein. Genes Dev. 25, 51–63.
Zheng, Y., Vertuani, S., Nystro¨m, S., Audebert, S., Meijer, I., Tegnebratt, T.,
Borg, J.-P., Uhle´n, P., Majumdar, A., and Holmgren, L. (2009). Angiomotin-
like protein 1 controls endothelial polarity and junction stability during sprout-
ing angiogenesis. Circ. Res. 105, 260–270.
Zhu, X., Liang, Y., Gao, F., and Pan, J. (2017). IFT54 regulates IFT20 stability
but is not essential for tubulin transport during ciliogenesis. Cell. Mol. Life
Sci. 74, 3425–3437.
Report
ll
OPEN ACCESSSTAR+METHODSKEY RESOURCES TABLEREAGENT or RESOURCE SOURCE IDENTIFIER
Antibodies
anti-myosin heavy chain DSHB Cat. # MF20; RRID:AB_2147781
anti-GFP AVES Cat. # GFP 1020; RRID:AB_10000240
anti-phospho-Smad 1/5 (Ser463/465) Cell signaling Cat. # 9516S; RRID:AB_491015
anti-phospho- Smad1/5/9 Cell signaling Cat. #13820; RRID:AB_2493181
anti-Tbx18 Santa Cruz Cat. # sc-17869; RRID:AB_2200374
anti-Yap1 Santa Cruz Cat. # sc-101199; RRID:AB_1131430
anti-Yap1 Lecaudey laboratory N/A
anti-Wt1 Santa Cruz Cat. # sc-192; RRID:AB_632611
anti- Amotl1 Sigma Cat. # HPA001196; RRID:AB_1078147
anti-g-tubulin Santa Cruz Cat. # sc-17787; RRID:AB_628417
anti-a-tubulin Sigma Cat. # T6199; RRID:AB_477583
anti-IFT88 Euromedex Cat. # 13967-1-AP; RRID:AB_2121979
anti-Islet1 Genetex Cat. # GTX128201; RRID:AB_11179180
anti-Yap1 Cell signaling Cat. # 4912; RRID:AB_2218911
anti-Yap/Taz Cell signaling Cat. # D24E4; RRID:AB_2799044
goat anti-chicken Alexa Fluor 488 IgY (H+L) Invitrogen Cat. # A-11039; RRID:AB_142924
goat anti-mouse IgG Cy3 conjugate (H+L) Life technologies Cat. # M30010; RRID:AB_2536619
goat anti-rabbit Alexa Fluor 647 ThermoFisher Cat. # A-21244; RRID:AB_2535812
40,6-diamidino-2-phenylindole (DAPI) Invitrogen Cat. # D1306; RRID:AB_2629482
anti-GFP ThermoFisher Cat. # CAB421; RRID AB_10709851
anti-FLAG Sigma Cat. # F7425; RRID:AB_439687
anti-Ift20 Proteintech Cat. # 13615-1-AP; RRID:AB_2280001
anti-GFP Abcam Cat. # ab290; RRID:AB_303395
anti-HA Roche Cat. # 11867423001; RRID:AB_390918
Chemicals, Peptides, and Recombinant Proteins
1-phenyl-2-thiourea (PTU) Sigma-Aldrich Cat. # P7629
Tricaine/MS-222 Sigma-Aldrich Cat. # A-5040
UltraPureTM Low Melting Point Agarose Invitrogen Cat. # 16520-050
Verteporfin Sigma-Aldrich Cat. # SML0534
XAV939 Tocris Bioscience Cat. # 3748
Phosphate buffered saline (PBS) Sigma-Aldrich Cat. # P4417
Indole-3-acetic acid sodium salt
(Heteroauxin)
Sigma-Aldrich Cat. # I5148
Lipofectamine 2000 Transfection Reagent ThermoFisher SCIENTIFIC Cat. # 11668030
cOmpleteTM Protease Inhibitor Cocktail Roche Cat. # 11697498001
Opti-MEM GIBCO Cat. # 51985-026
Oligofectamine Invitrogen Cat. # 12252-01
Hoechst ThermoFisher Cat. # 62249
Experimental Models: Cell Lines
HEK293 cells Q-BIOgene AES0503
MDCK cells Sumara laboratory N/A
HeLa cells Sumara laboratory N/A
DLD-1 IFT88-AID Vitre et al., 2020 N/A
(Continued on next page)
Cell Reports 32, 107932, July 21, 2020 e1
Continued
REAGENT or RESOURCE SOURCE IDENTIFIER
Experimental Models: Organisms/Strains
Zebrafish: Et(
26.5Hsa.WT1-
1gata2:EGFP)cn1 (epi:GFP)
Peralta et al., 2013 ZFIN ID: ZDB-ALT-170823-7
Zebrafish: iguanats294e Tay et al., 2010 ZFIN ID: ZDB-FISH-150901-16040
Zebrafish: ift88 tz288/oval Tsujikawa and Malicki, 2004 ZFIN ID: ZDB-ALT-980413-526
Zebrafish: elipsa tp49d Omori et al., 2008 ZFIN ID: ZDB-ALT-980413-466
Zebrafish: yap1 fu48 Agarwala et al., 2015 ZFIN ID: ZDB-ALT-160413-4
Zebrafish: amotl2a fu46 Agarwala et al., 2015 ZFIN ID: ZDB-ALT-160412-2
Zebrafish: 4xGTIIC:d2GFP Miesfeld and Link, 2014 ZFIN ID: ZDB-ALT-150928-6
Zebrafish: actb2:Mmu.Arl13b-GFP Borovina et al., 2010 ZFIN ID: ZDB-ALT-100721-1
Zebrafish: tcf21:NLS-EGFP Kikuchi et al., 2011 ZFIN ID: ZDB-ALT-110914-2
Mice: Ift20null/+ (C57BL background) Jonassen et al., 2008 MGI: 12565
Mice: PGK-Cre Lallemand et al., 1998 PMID: 9608738
Mice: Ift88null/+ (6JRj background) Haycraft et al., 2007 MGI:5505911
Oligonucleotides
siRNA ON-Target plus - Control pool Non-
targeting
Dharmacon D-001810-10-05
siRNA ON-Target plus - SMART pool
human IFT88
Dharmacon L-012281-01
Recombinant DNA
Flag-Amotl1 Pei et al., 2010 N/A
Ift20-GFP Follit et al., 2006 N/A
YAP1-Myc Boin et al., 2014 N/A
pEGFP-C1 Ragni et al., 2017 N/A
HA-Amotl1 Ragni et al., 2017 N/A
IFT88-GFP He et al., 2014 N/A
Software and Algorithms
FIJI Schindelin et al., 2012 RRID:SCR_002285
Imaris Bitplane RRID: SCR_007370
MATLAB MathWorks RRID: SCR_001622
Microsoft Excel Microsoft RRID: SCR_016137
GraphPad Prism 7 GraphPad Software RRID: SCR_002798
Other
GFP-Trap ChromoTek Cat. # gta-10
anti-FLAG M2 Affinity gel Sigma Cat. # A2220
Report
ll
OPEN ACCESSRESOURCE AVAILABILITY
Lead Contact
Further information and requests for reagents may be directed to and will be fulfilled by the Lead Contact, Julien Vermot (jvermot@
imperial.ac.uk).
Materials Availability
This study did not generate new unique reagents.
Data and Code Availability
This study did not generate any unique datasets or code.e2 Cell Reports 32, 107932, July 21, 2020
Report
ll
OPEN ACCESSEXPERIMENTAL MODEL AND SUBJECT DETAILS
Zebrafish (ZF) Models
Animal experiments were approved by the Animal Experimentation Committee of the Institutional Review Board of the IGBMC.
ZF lines used in the study were Et(
26.5Hsa.WT1-1gata2:EGFP)cn1 transgenic line (referred to as epi:GFP) (Peralta et al., 2013),
amotl2a fu46 (Agarwala et al., 2015), elipsa tp49d (Omori et al., 2008), ift88 tz288/oval (Tsujikawa and Malicki, 2004), iguana ts294e (Tay
et al., 2010), yap1 fu48 (Agarwala et al., 2015), 4xGTIIC:d2GFP (Miesfeld and Link, 2014), actb2:Mmu.Arl13b-GFP (Borovina et al.,
2010) and tcf21:NLS-EGFP (Kikuchi et al., 2011). Male and female samples were mixed. All animals were incubated at 28.5C for
24h before treatment with 1-phenyl-2-thiourea (PTU) (Sigma Aldrich) to prevent pigment formation.
Mouse Models
Ift20flox/+ (Jonassen et al., 2008) were crossed to PGK-Cremice (Lallemand et al., 1998) to generate Ift20null/+maintained on aC57BL/
6JRj genetic background. Ift88null/+ (Haycraft et al., 2007) mice were maintained on a B6D2 genetic background. Animal procedures
were approved by the ethical committee of the Institut Pasteur and the French Ministry of Research. E9.5 embryos were isolated in
200ng/ml cold heparin, incubated in cold 250mM KCl and fixed in 4% paraformaldehyde in PBS inside a rotative oven at 37C over-
night to remove excess of blood. Male and female samples were mixed.
METHOD DETAILS
In Vivo Imaging
ZF embryos were staged, anaesthetized with 0.02% tricaine solution and mounted in 0.7% low melting-point agarose (Sigma Al-
drich). Confocal imaging was performed on a Leica SP8 confocal microscope. Images were acquired bidirectionally with a low-
magnification water immersion objective (Leica HCX IRAPO L, 25X, N.A. 0.95). For time lapse, z stacks were acquired each 15 or
30 min, depending on the experiment. The optical plane was moved 15 mm between z sections.
Bright field experiments were performed on a Leica DMIRBE inverted microscope using a Photron SA3 high speed CMOS camera
(Photron, San Diego, CA) and water immersion objective (Leica 20X, NA 0.7). Image sequences were acquired at a frame rate of 150
frames per second.
ZF Treatments
Verteporfin (5 mM) (Sigma Aldrich) and XAV939 (10 mM) were diluted in fish tank water with 0.0033% PTU, in which larvae were incu-
bated in darkness at 28.5C for the required time.
ZF Immunofluorescence
Embryos were fixed at the desired stages in 4% paraformaldehyde (PFA) overnight at 4C. After washing in 0.1% PBS Tween 20,
embryoswere permeabilized in 0.5%PBS Triton X-100 for 20min at room temperature (RT). Samples werewashed and then blocked
(3% albumin from bovine serum (BSA), 5% goat serum, 20 mMMgCl2, 0.3% Tween 20 in PBS) during 2h at RT. Primary antibodies
were added in the blocking solution and incubated overnight at 4C. Secondary antibodies were added in 0.1% PBS Tween20 after
thorough washing and incubated overnight at 4C. Embryos were washed and incubated with DAPI (Invitrogen), 1:1000, for 15min at
RT. After being thoroughly washed, samples were mounted for imaging on a Leica SP8 confocal with a dipping immersion objective
(Leica HCX IRAPO L, 25X, N.A. 0.95). Z stacks were taken every 10 mm. 3D images were reconstructed using IMARIS software (Bit-
plane Scientific Software). The ventral pericardium was digitally removed to provide a clearer view of the heart.
Antibodies used were as follows: anti-myosin heavy chain (MF20, DSHB) 1:20, anti-GFP (AVES) 1:500, anti-phospho-Smad 1/5
(Ser463/465) (Cell signaling) 1:50, anti-Islet1 (Genetex) 1:100, anti-Yap1 (Lecaudey laboratory) 1:200. Secondary antibodies: goat
anti-chicken Alexa Fluor 488 IgY (H+L) (In vitrogen), goat anti-mouse IgG Cy3 conjugate (H+L) (Life technologies) and goat anti-rabbit
Alexa Fluor 647 (ThermoFisher) were used at 1:500.
To test the effects of Verteporfin treatment, embryos were rinsed in fish tank water before being fixed and processed as described
above.
In Situ Hybridization (ISH)
ISH was performed in whole embryos according to Thisse and Thisse, 2008, with minor modifications. Antisense mRNA probe used
was against full coding sequence of bmp4. To test the effects of Verteporfin and XAV939 treatments, embryos were rinsed in fish tank
water before being fixed and processed as described above.
Whole Mount Immunofluorescence in the Mouse
Embryos were fixed in paraformaldehyde. The cardiac region was dissected, permeabilized in 0.75% Triton. Aldehydes were
quenched with 2.6mg/ml NH4Cl. Immunostaining was performed in 10% inactivated horse serum, 0.5% Triton with a primary anti-
body against Wt1 (Santa Cruz sc-192, 1:50), and with Alexa Fluor conjugated secondary antibodies (1:300) and counterstained with
Hoechst (1:400). 80% glycerol was used to make the samples transparent.Cell Reports 32, 107932, July 21, 2020 e3
Report
ll
OPEN ACCESSImmunofluorescence on Cryosections in the Mouse
Embryos were embedded in 7% gelatin, 15% sucrose, frozen in cold isopentane and sectioned on a cryostat (10 mm). Immunostain-
ing was performed on cryosections as described above, with permeabilization in 0.5% Triton, and with an additional incubation in
0.2 mg/mL goat anti-mouse IgG Fab fragment to reduce non-specific reactivity of antibodies raised in themouse. Primary antibodies
against Tbx18 (Santa Cruz sc-17869, 1:100), Wt1 (Santa Cruz sc-192, 1:50), Yap1 (Santa Cruz sc-101199, 1:100), Amotl1 (Sigma
HPA001196, 1:50) and p-Smad1/5/9 (Cell signaling 13820, 1:250) were used, with Alexa Fluor conjugated secondary antibodies
(1:500) and Hoechst nuclear counterstaining (1:1000). Samples were imaged in DAKO mounting medium on a LSM700 (Zeiss)
confocal microscope with a 40X/1.3 objective. Z stacks were taken every 0.9 mm.
Cell Culture, siRNAs and Transfection
Cells were cultured in appropriate conditions: MDCK (MEM Eagle - Earle’s BSS, 10% FCS, AANE 0.1 mM, Sodium Pyruvate 1mM,
Gentamicin 40 mg/ml), HeLa (DMEM 4.5 g/l glucose, 10% FCS, Penicillin 100 UI/ml, Streptomycin 100 mg/ml), HEK293 cells (DMEM
1g/L glucose, FCS 10%, Penicillin 100UI/ml, Streptomycin 100 mg/ml) andDLD-1 (DMEM4.5 g/l glucose, 10%FCS, Penicillin 100UI/
ml, Streptomycin 100 mg/ml). siRNA (Dharmacon) ON-Target plus - Control pool Non-targeting (D-001810-10-05) and SMART pool
human IFT88 (L-012281-01) were used at 50 nM working concentration. Cells were transfected 16h after splitting using Opti-MEM
medium and Oligofectamine reagent.
Generation of DLD-1 IFT88-AID Targeted Cells
DLD-1 IFT88-AID cells were generated by adding an AID tag followed by a YFP tag at the 30 end of the last exon on the IFT88 genomic
locus. In detail, a clonal population of DLD-1 cells stably expressing TIR1-9xMyc protein was used for targeting (Holland et al., 2012).
sgRNA targeting two regions adjacent to the 30 end of IFT88 gene were introduced under the control of U6 transcription promoter into
two separate vectors encoding for the expression of the Cas9 nickase (D10A) (Cong et al., 2013) (addgene 42335). A donor construct
containingz600 bp recombination arms surrounding the 30 end of IFT88 locus, in frame with a sequence encoding for an AID-YFP-
Stop sequence, was generated. All three vectors were transfected into DLD-1 TIR1 cells using Xtreme Gene 9 DNA transfection re-
agent (Roche). Cells were sorted based on their YFP fluorescence and single clones were isolated. Homozygous targeted clones
were identified by PCR. Targeting of IFT88 and degradation of IFT88-AID-YFP was confirmed by immunoblot following addition
of Auxin (Sigma-Aldrich) at 500 mM in the culture medium for the indicated times.
Immunofluorescence on Cells
Cells were fixed in 100%MeOH for 6 min at
20C (Ift88 and g-tubulin antibodies), in 4% PFA for 7 min at RT (DLD-1 cells. Yap1 anti-
body) or in PFA 4% for 17 min at RT (MDCK and HeLa cells. Yap/Taz antibody). After washing in 0.1% PBS Tween20, cells were per-
meabilized in 0.5%PBS-NP40 and blocked in 5%BSA 1h at RT. Primary antibodies were added in the blocking solution and incubated
overnight at 4C. Secondary antibodies were added in 0.1% PBS Tween20 after washing and incubated for 2h at RT. Then cells were
incubatedwithDAPI (Invitrogen), 1:1000, for 15min atRT. After being thoroughlywashed, samplesweremounted for imaging ona Leica
SP5 (siRNA experiments) or SP8 (DLD-1 experiments and experiments to assess subcellular localization) confocal microscope with an
oil immersion objective (Leica HCX PL APO lambda blue, 63X, N.A. 1.4). Z stacks were taken every 1 mm.
Antibodies used were as follows: anti-g-tubulin (Santa Cruz) 1:500, anti-a-tubulin (Sigma, 1:2000), anti-IFT88 (Euromedex) 1:50,
anti-Yap1 (4912) (Cell signaling) 1:50 (DLD-1 experiments), and anti-Yap/Taz (D24E4) (Cell signaling) 1:50 (MDCK and HeLa siRNA
experiments). Secondary antibodies: goat anti-mouse IgG Cy3 conjugate (H+L) (Life technologies) and goat anti-rabbit Alexa Fluor
647 (In vitrogen) were used at 1:500.
Lysates and Immunoblotting
DLD-1 cell extracts were obtained after lysis with Laemmli sample buffer of an equal number of cells for each sample. Proteins were
resolved by SDS-PAGE, transferred to nitrocellulose membranes and revealed by immunoblot using Western Lightning Plus-ECL kit
(PerkinElmer).
IMMUNOPRECIPITATION (IP) ASSAYS
HEK293 Cells
HEK293 cells (Q-BIOgene AES0503) were co-transfected with plasmids Flag-Amotl1 (Pei et al., 2010) and Ift20-GFP (Follit et al.,
2006), or a Flag-control plasmid using Lipofectamine 2000 Transfection Reagent (ThermoFisher SCIENTIFIC) and cultured for
48h. Proteins were extracted in a lysis buffer (10mMTris-Cl pH 7.5, 5mM EDTA, 150mM NaCl, 10% glycerol and 5% CHAPS) in
the presence of protease inhibitors (cOmpleteTM Protease Inhibitor Cocktail, Roche). Immunoprecipitation of protein extracts
was performed using a monoclonal anti-Flag antibody covalently attached to agarose (Anti-FLAG M2 Affinity gel, Sigma). Proteins
were eluted in 2xNuPAGE LDS Sample Buffer (ThermoFisher). Proteins were separated on SDS–polyacrylamide gel electrophoresis
and transferred to a nitrocellulose membrane. Proteins were detected with the primary antibodies against Flag (1:1000, Sigmae4 Cell Reports 32, 107932, July 21, 2020
Report
ll
OPEN ACCESSF7425), GFP (1:1000, ThermoFisher CAB421), Ift20 (1:500, Proteintech, 13615-1-AP), Yap1 (1:1000, Cell Signaling 4912S) and
Amotl1 (1:1000, Sigma HPA001196), followed by HRP-conjugated secondary antibodies (1:5000, Jackson ImmunoResearch) and
the ECL detection reagent.
HeLa Cells
We performed IPs using GFP-Trap (ChromoTek) agarose beads in two conditions: Control IP (YAP1-Myc (Boin et al., 2014), pEGFP-
C1 and HA-Amotl1 (Ragni et al., 2017)) and IFT88 IP (YAP1-Myc, IFT88-GFP (He et al., 2014) and HA-Amotl1). HeLa cells (2x 10cm
dish/ condition) were transfected with Lipofectamine 2000. Experiments were performed using the following setup: Cells were
seeded at high density into 10cmdishes and transfected 16 hr after seeding at 95%confluency. Twenty-four hours post-transfection,
cells were seeded into 15cm dishes in order to achieve culture of isolated cells (10x 15cm dish/ condition). Proteins were extracted
60 hr post-transfection in a lysis buffer (10mM TrisHCl, pH7.5; 150mM NaCl; 0.5mM EDTA; 0.5% NP-40; protease inhibitors Com-
plete). GFP beads were washed once in lysis buffer and incubated with 16 mg of the protein lysate for 16 hr at 4C. Beads were
washed four times in buffer without detergent and proteins were eluted by boiling for 10 mins. The input (1%) and IP were analyzed
using immunoblot and the membranes were probed with anti-GFP (Abcam), anti-Yap/TAZ (D24E4, Cell Signaling) and anti-HA
(Sigma Aldrich) antibodies.
ZF avcPE Cell Quantification
Weperformedwhole-mount immunofluorescent staining on control andmutant embryos in the epi:GFP reporter line background and
imaged the heart using a confocal microscope with a z-step of 10 mm. The different tissues were labeled using anti-myosin heavy
chain (MHC) (myocardium) and anti-GFP (epi:GFP) antibodies, andDAPI dye to stain for nuclei. We thenmanually quantified the num-
ber of avcPE cells per z slice. We identified the avcPE clusters anatomically: avcPE clusters form in the dorsal pericardium, close to
the atrio-ventricular canal (avc). PE cells were identified by their rounded morphology in conjunction with their expression of GFP
(although some pericardial cells are also GFP-positive, they can be excluded due to their flat morphology). In order to count each
cell only once in the z stack, we only counted a cell when its nucleus was visible.
ZF Myocardial Cell Quantification
We performed whole-mount immunofluorescent staining on control and mutant embryos using anti-myosin heavy chain (MHC)
(myocardium) antibody and DAPI dye (nuclei). We imaged the heart using a confocal microscope with a z-step of 10 mm. We then
manually quantified the number of myocardial cells per z slice (nuclei surrounded by MHC signal).
Mice PE Volume Analysis
Wholemount embryos stained withWt1 antibody (Santa Cruz) were scanned on TCS SP8 DLS (Digital Light Sheet) Leica with a water
immersion objective (HC APO L, 10X, 0.3). Z stacks were taken every 2 mm. Both, coronal and sagittal views were acquired, if
possible, for a more precise analysis (Figure S2F). Using MATLAB software, the contour of the PE was manually drawn (Wt1 signal)
for each z plane and the area (A) was calculated. Volume (V) was estimated as:
V = Szn Ai x dz ; i= 1 ; zn= number of planes; dz= 2 mm
We performed a double-blind quantitative analysis. 3D images were reconstructed using IMARIS software (Bitplane Scientific
Software).
Mice Yap1- and Amotl1-Positive Cell Quantifications
Nuclei positions on the slides were defined using IMARIS (Bitplane Scientific Software) Spots detection function. The results were
manually corrected if needed. Nuclei positions were exported from Imaris and imported toMATLAB. Each cell was assigned a unique
index. The intensity of Tbx18 signal was evaluated in correspondence with the nuclei positions. Cells where the intensity was found
higher than a threshold were considered positive. The threshold was established according to the background noise intensity. Re-
sults were manually corrected if needed and Tbx18-positive cells were automatically counted. Yap1 signal was visualized in fire LUT
to facilitate perception of signal intensity. Nuclear Yap1-positive cells (higher signal in the nucleus than in the cytoplasm) were manu-
ally defined through index identification of cells and counted automatically. The same procedure was followed for Amotl1 signal. The
outline of the outer PE region wasmanually drawn and areas of the two regions were calculated automatically. MATLAB provided the
total positive cell number for each signal and area, including signal co-localization. We performed a double-blind quantitative
analysis.
Nuclear Yap Quantifications on Cells
Analyses were performed using Fiji (Schindelin et al., 2012). Nuclei areas were selectedmanually using DAPI signal as reference on z-
projection images (sum slices for DLD-1 experiment and maximum intensity projection for siRNA experiments). Yap average nuclear
signal intensity was measured for the selected areas. In experiments in DLD-1 cells the values were measured for each individual
nucleus, while in the case of siRNA experiments, all the nuclei in a slice were measured together. Values were normalized to their
controls in order to merge data from different experiments.Cell Reports 32, 107932, July 21, 2020 e5
Report
ll
OPEN ACCESSNuclear/cytoplasmic Yap1 Ratio
Analyses were performed using Fiji. Nuclei areas were selectedmanually using DAPI signal as reference on z-projection images (sum
slices). Cytoplasmic areas were selected using a-tubulin signal as reference. Yap average nuclear signal intensity was measured for
the nuclear ROI. Yap average cytoplasmic signal intensity was measured after subtracting nuclear ROI from the cytoplasmic ROI.
QUANTIFICATION AND STATISTICAL ANALYSIS
We applied D’Agostino & Pearson and Shapiro-Wilk normality tests to assess whether the samples fit a normal distribution and F test
to compare variances. For normal distributed and homoscedastic samples, we used t test or ANOVA. For non-parametric samples,
we applied Man-Whitney or Kruskal-Wallis. The pertinent statistical analyses for each experiment were performed using GraphPad
Prism 7 software. For the analysis of nuclear p-smad 1/5/9 and YAP1 signal in mice we used the non-parametric Chi-square test of
homogeneity to test whether the observed frequency of positive nuclei was equally distributed across the wild-type and mutant em-
bryos. In each figure legend is stated the number of embryos (n), as well as the meaning of error bars and p values.e6 Cell Reports 32, 107932, July 21, 2020