CDC42 is required for structural patterning of the lung during development
Huajing Wana,b,n, Caijun Liua, Susan E. Wertb, Wei Xua, Yong Liaoa, Yi Zhengc, Jeffrey A. Whitsettb
aKey Laboratory of Obstetric, Gynecologic and Pediatric Diseases and Birth Defects of the Ministry of Education, West China Institute of Women and Children’s Health,
and Department of Pediatrics, Huaxi Second University Hospital, Sichuan University, No. 17, Section 3, South Renming Rd., Chengdu, Sichuan 610041, PR China
bDivision of Pulmonary Biology, Cincinnati Children’s Hospital Medical Center and the University of Cincinnati College of Medicine, 3333 Burnet Avenue, Cincinnati,
OH 45229-3039, USA
cDivision of Experimental Hematology, Cincinnati Children’s Hospital Medical Center and the University of Cincinnati College of Medicine, 3333 Burnet Avenue, Cincinnati,
OH 45229-3039, USA
a r t i c l e i n f o
Received 23 June 2012
Received in revised form
16 November 2012
Accepted 17 November 2012
Available online 5 December 2012
a b s t r a c t
The formation of highly branched epithelial structures is critical for the development of many essential
organs, including lung, liver, pancreas, kidney and mammary glands. Elongation and branching of these
structures require precise control of complex morphogenetic processes that are dependent upon
coordinate regulation of cell shape, apical–basal polarity, proliferation, migration, and interactions
among multiple cell types. Herein, we demonstrate that temporal–spatial regulation of epithelial cell
polarity by the small GTPase, CDC42, is essential for branching morphogenesis of the developing lung.
Epithelial cell-specific deletion of CDC42 in fetal mice disrupted epithelial cell polarity, the actin
cytoskeleton, intercellular contacts, directional trafficking of proteins, proliferation and mitotic spindle
orientation, impairing the organization and patterning of the developing respiratory epithelium and
adjacent mesenchyme. Transition from a pseudostratified to a simple columnar epithelium was
impaired, consistent with coordinate dysregulation of epithelial cell polarity, mitotic spindle orienta-
tion, and repositioning of mitotic cells within the epithelium during cell cycle progression. Expression
of sonic hedgehog and its receptor, patched-1, was decreased, while fibroblast growth factor 10
expression in the mesenchyme was expanded, resulting in disruption of branching morphogenesis and
bronchiolar smooth muscle formation in this model. CDC42 is required for spatial positioning of
proliferating epithelial cells, as well as signaling interactions between the epithelium and mesenchyme
and is, therefore, essential for formation and maintenance of the respiratory tract during morphogen-
esis of the fetal lung.
& 2012 Elsevier Inc. All rights reserved.
Many organs, such as the lung, kidney, pancreas, liver, repro-
ductive tract, breast and other glands, are composed of highly
branched epithelial tubes through which gases and liquids are
transported to sustain life. Although tubular organs vary in size,
shape and specific functions, their tubules are lined by highly
polarized, interconnected epithelial cells with apical surfaces
facing the lumen, basal surfaces adjoining the basement mem-
brane, and lateral surfaces that are in contact with neighboring
cells via intercellular junctions and cell adhesion molecules.
Formation, modulation, and maintenance of epithelial cell polar-
ity are critical for tubule formation in Drosophila melanogaster
(D. melanogaster) and Caenorhabditis elegans (C. elegans), as well as
in cultured epithelial cells (Chung and Andrew, 2008). Because of
the structural and functional complexity of mammalian organs,
regulation of tubulogenesis is highly complex. In vertebrates,
epithelial tubes are derived from either non-polarized cells, as
occurs during formation of the kidney, pancreas, mammary
glands, and parts of the vascular system, or from budding of
previously polarized tubules, as occurs in the lung and liver
(Chung and Andrew, 2008). Recent studies demonstrated that
the acquisition of apical cell polarity is critical for epithelial tube
formation from non-polarized cell condensations in the pancreas
(Kesavan et al., 2009). The role of polarity in the formation of
tubular organs derived from previously established, highly polar-
ized, epithelial tubes, such as the lung, is less clear.
The respiratory tract consists of a series of branched tubes that
are derived from a previously polarized endodermal tube. Lung
morphogenesis in the mouse embryo begins at E9 to E9.5 with the
ventral evagination of the anterior foregut endoderm into the
splanchnic mesenchyme to form the trachea and the first two
primordial lung tubules. Thereafter, from E9.5 to E16.5, the
Contents lists available at SciVerse ScienceDirect
journal homepage: www.elsevier.com/locate/developmentalbiology
0012-1606/$-see front matter & 2012 Elsevier Inc. All rights reserved.
nCorresponding author at: The West China Second University Hospital of
Sichuan University, No. 17, Section 3, South Renming Rd., Chengdu, Sichuan
610041, China. Fax: þ86 28 85501863.
E-mail address: email@example.com (H. Wan).
Developmental Biology 374 (2013) 46–57
respiratory tubules undergo stereotypical branching to form the
conducting airways and the peripheral alveolar sacs required for
gas exchange after birth (Metzger et al., 2008). At the end of
branching morphogenesis, respiratory tubules are lined by a
simple columnar (proximal) or cuboidal (peripheral) epithelium,
consisting of a single layer of highly polarized epithelial cells.
During lung development, interactions between epithelial and
mesenchymal cells, the latter forming pulmonary vessels, stroma
and bronchial smooth muscle, are dependent upon apical–basal
cell polarity, cell adhesion, and the bi-directional secretion of
signaling molecules and extracellular matrix (ECM) proteins.
Complex autocrine, paracrine, and juxtacrine interactions among
various epithelial and mesenchymal cells mediate morphogenesis
of the lung, as well as differentiation of progenitor cells into the
diverse cell types that form the mature lung. These interactions
are modulated by multiple signaling pathways, including FGF,
SHH, WNT, BMP, TGFb, VEGF, PDGF, and NOTCH, which are
required for normal branching morphogenesis of the lung (Affolter
et al., 2009; Cardoso and Lu, 2006; Morrisey and Hogan, 2010).
Less well understood are the molecular and cellular mechanisms
that are dependent on epithelial cell polarity during formation of
the respiratory tubules in the developing lung.
CDC42 is a member of the Rho family of small GTPases.
Previous studies in yeast, C. elegans, D. melanogaster, and epithe-
lial cell culture have provided insights into the important role
of CDC42 in establishing polarity and tubulogenesis (Melendez
et al., 2011). Extracellular signals, cell–matrix and cell–cell inter-
actions, as well as intrinsic signals generated during the cell cycle,
influence the active GTP-bound and inactive GDP-bound forms of
CDC42, which, in turn, modulate interactions with various effec-
tor proteins, including PAR, PAK, WASP, and SEC, to regulate
diverse cellular functions that influence intercellular junctions,
directional protein secretion, spindle orientation and cytoskeletal
dynamics (Bryant and Mostov, 2008; Jaffe et al., 2008; Rojas et al.,
2001; Symons et al., 1996; Zhang et al., 2001). Recent tissue-
specific gene targeting studies demonstrated that CDC42 is
required for development of the pancreas, central nervous sys-
tem, liver, eye, skin, bone, blood and immune system (Cappello
et al., 2006; Guo et al., 2010; Maillet et al., 2009; Melendez et al.,
2011; van Hengel et al., 2008; Wang et al., 2006; Wu et al., 2006).
For example, CDC42 is required for de novo tube formation in the
pancreas, which, in turn, regulates pancreatic cell differentiation
(Kesavan et al., 2009). The potential role of CDC42 in formation
and maintenance of tubular organs derived from previously
polarized tubes, however, is currently unknown. In the present
study, mouse Cdc42 was conditionally deleted from respiratory
epithelial cells in the mouse embryo, disrupting formation of the
fetal lung. We found that CDC42 is required for branching
morphogenesis, mediating the regulation of spatial positioning
of developing respiratory epithelial cells, mitotic spindle orienta-
tion during cell proliferation, and directional signaling interac-
tions between the epithelium and mesenchyme.
Materials and methods
Animal protocols were approved by the Institutional Animal Care
and Use Committee at West China Second University Hospital
and Cincinnati Children’s Hospital Medical Center in accordance
with established guidelines. Cdc42loxP/loxPmice were generated as
previously described (Yang et al., 2007). Homologous recombination
was accomplished utilizing a (TetO)7CMV-Cre transgenic mouse line
(Sauer, 1998), kindly provided by Dr. Corrinne Lobe, University of
Toronto. For lung-specific, doxycycline-induced recombination, the
FVB. Cg-Tg(SFTPC-rtTA)5Jaw/J transgenic line was used (The Jackson
Laboratory, Bar Harbor, ME, USA). Triple transgenic mice (TetO)7-
Crewt/tg/SFTPC-rtTAwt/tg/Cdc42loxP/loxP, termed Cdc42D/Dafter expo-
sure to doxycycline, were generated in a mixed genetic background by
crossing (TetO)7-Crewt/tg/Cdc42loxP/loxPand SFTPC-rtTAwt/tg/Cdc42loxP/loxP
mice. Dams bearing double- and triple-transgenic fetuses were main-
tained on doxycycline-containing food (625 mg/kg, Suzhou Shuangshi
Laboratory Animal Feed Science CO., Suzhou, Jiangshu, PR China) from
E6.5 to E14.5. Doxycycline-treated SFTPC-rtTAwt/tg/(TetO)7-Crewt/tg
mice, (TetO)7-Crewt/tg/Cdc42loxP/loxPmice, SFTPC-rtTAwt/tg/Cdc42loxP/loxP
mice, or untreated SFTPC-rtTAwt/tg/(TetO)7-Crewt/tg/Cdc42loxP/loxPmice,
served as controls. Transgenic mice were identified by PCR, using
genomic DNA from the tails of fetal and postnatal mice as previously
described (Perl et al., 2002; Sauer, 1998; Yang et al., 2007). Cdc42D/D
and control mice were present at ratios consistent with Mendelian
inheritance at all time points examined. All experiments shown are
representative of findings from at least 2 independent dams, generat-
ing at least 4 triple transgenic offspring that were compared with
Histology, immunohistochemistry, in situ hybridization, and electron
Fetal lung tissue from E11.5-E18.5 was immersion-fixed in 4%
paraformaldehyde, embedded in paraffin or frozen in OCT, sec-
tioned, and immunostained, as previously described (Dave et al.,
2008; Tompkins et al., 2011; Wan et al., 2008). Primary antibodies
are listed in Table S1. Biotinylated (Vector Laboratories, Burlin-
game, CA) and fluorochrome-conjugated (Alexa Fluor 488 or 594;
Invitrogen, Grand Island, NY, USA) secondary antibodies were
used for detection of the primary antibodies. Substitution of the
primary antibody with non-immune sera or specific mouse IgG
isotypes was used to assess non-specific binding of primary
antibodies to the tissue. In situ hybridization for Fgf10 mRNA
was performed on paraffin sections of lung tissue harvested from
E14.5 Cdc42D/Dand littermate controls, using
riboprobes as described previously (Clark et al., 2001; Wert et al.,
1993). Fetal lung tissue from E14.5 Cdc42D/Dand littermate
controls was fixed in 3% glutaradehyde, post-fixed in 1% osmium
tetroxide (OsO4), and embedded in EMbed 812 (Electron Micro-
scopy Sciences, Hatfield, PA, USA) for ultrastructural analysis.
Ultrathin sections were stained with lead citrate and uranyl
acetate and then examined with a Hitachi H-7600 transmission
electron microscope (Hitachi High-Technologies America, Inc.,
Pleasanton, CA, USA).
RNA and protein analysis
RNA was isolated from whole lung and reverse transcribed
according to the manufacturer’s protocol (EP0441, Fermentas
China, Shenzhen City, Guangdong, PR China), prior to RT-PCR
analysis. Densitometric quantification of the PCR products was
carried out using Quality One software (Bio-Rad Laboratories,
Philadelphia, PA, USA). The relative concentrations of mRNAs
were normalized to b-actin mRNA. Primer sequences for Cdc42,
Shh, Ptch, b-actin, Aqp5, and T-1a are listed in Table S2. Total lung
proteins were harvested and Western blotting was performed
with antibodies to pERK and GAPDH (Table S1).
Cell proliferation, cell cycle progression, and nuclear/cell positioning
Pregnant dams were injected i.p. with 50 mg/kg of body weight
of 5-Bromo-20-Deoxyuridine (BrdU) (Thermo Fisher Scientific,
Waltham, MA, USA) 10 or 30 min before harvesting the pups.
Fetal lung tissues were immersion-fixed, embedded in OCT,
frozen, sectioned, and immunostained for BrdU or phosphohistone
H. Wan et al. / Developmental Biology 374 (2013) 46–57
H3 (pHH3), and counter-stained with DAPI. In order to quantify
cell proliferation, BrdU-, pHH3-, and DAPI-positive epithelial cells
were counted manually, and the ratios of positive epithelial cells
to total number of epithelial cells examined were expressed as
percentages. At least 5 sections from each of 4 control and
4 experimental mice were used for the analysis; and 1500
epithelial cells were counted to obtain the BrdU- and pHH3-
labeled cell counts.
Quantification of mitotic spindle orientation
Lung sections from E14.5 Cdc42D/Dand littermate control mice
were stained with DAPI and photographed using fluorescent
optics. Single or paired dividing cells at anaphase and telophase
were identified by their characteristic morphological patterns
(Goto et al., 2002). When paired dividing nuclei were identified
on the micrographs, orientation of the mitotic spindle pole (MSP)
was determined by a line linking the midpoints of the paired
dividing nuclei. When only one of the paired nuclei was identi-
fied, orientation of the MSP was determined by the axis perpen-
dicular to the longitudinal axis of the dividing nucleus. y, defined
as the projection angle of the MSP onto the plane of the basement
membrane, was measured as previously described (El-Hashash
et al., 2011; Lechler and Fuchs, 2005). Parallel cell division was
defined as 0ryr30 and perpendicular cell division was defined
as 60ryr90. Micrographs were acquired from at least 3 sections
for each of 5 control and 5 Cdc42D/Dmice. The ratio of parallel cell
division to perpendicular cell division was calculated for each
group, and the means7s.e.m. were calculated and compared
between controls and Cdc42D/Dmice.
The Student’s t test was used to determine differences
between groups, and the P value was set at po0.05 for signifi-
cance (n). Values for all measurements were expressed as the
CDC42 is required for normal structural development of the lung
In order to address the role of CDC42 during lung develop-
ment, Cdc42loxP/loxPmice were bred into (TetO)7-Crewt/tgand
surfactant protein C (SFTPC)-rtTAwt/tgtransgenic mice to produce
triple transgenic mice, termed Cdc42D/D, in which Cdc42 was
selectively deleted in respiratory epithelial cells upon adminis-
tration of doxycycline to pregnant dams from E6.5 to E14.5.
All newborn Cdc42D/Dmice died from respiratory failure at birth,
while control mice survived (n¼4 litters, including 10 Cdc42D/D
mice). Efficiency of Cre-mediated Cdc42 deletion was assessed by
RT-PCR at E12.5 and E14.5. Cdc42 mRNA was decreased significantly
at E14.5, but not at E12.5 (Fig. 1). No significant changes in lung-
to-body weight ratios were observed at E14.5 in the Cdc42D/Dwhen
compared to littermate controls (0.018570.0036 vs. 0.01867
0.0034, respectively, p40.05). At E18.5, however, lung weights
(0.015770.003 mg, Cdc42D/Dmice vs. 0.042770.001 mg, controls,
po0.05) and lung-to-body weight ratios (0.011270.0018, Cdc42D/D
mice vs. 0.030570.0017,controls,
decreased in the Cdc42D/Dmice. Epithelial cell proliferation in the
Cdc42D/Dmice, as assessed by BrdU incorporation, was decreased
significantly by E14.5 compared to controls (14.479.6% vs.
36.9717.5, respectively, po0.05). Likewise, the mitotic index, as
assessed by staining for pHH3, was decreased at E14.5 in the Cdc42D/D
mice compared to controls (4.372.8% vs. 6.5474%, respectively,
po0.05), a result consistent with the BrdU data. These data indicate
that loss of CDC42 decreased cell proliferation, impairing growth of
the fetal lung. In contrast, no changes in cleaved caspase-3 staining
were detected at E14.5 (Fig. S1), suggesting that CDC42 is not
involved in the regulation of cell survival.
Histologic abnormalities found after deletion of CDC42 were
confined to the lung. Although abnormalities in lung morpho-
genesis were not detected in isolated Cdc42D/Dmice at E12.5
(Fig. 2A, B), enlarged tubules were easily seen by E14.5 in whole
mounts of lungs isolated from Cdc42D/Dmice (Fig. 2C, D). At E12.5,
tubular epithelial cells were organized as pseudostratified epithe-
lium in both control (Fig. 2E) and Cdc42D/D(Fig. 2F) mice,
consistent with the lack of changes in Cdc42 mRNA at this time
point (Fig. 1). By E14.5, however, branching morphogenesis was
severely disrupted in the Cdc42D/Dmice. Dilated tubules with
highly disorganized epithelial cells, as well as disorganized
peripheral buds, were observed in the Cdc42D/Dmice (Fig. 2H,
J, L, N), consistent with the decrease in Cdc42 mRNA observed at
this time point (Fig. 1). In the controls, most of the respiratory
tubules were composed of a single layer of columnar (proximal)
or cuboidal (peripheral) epithelial cells with nuclei located near
the basal pole of the cells (Fig. 2I, K, M). In the Cdc42D/Dmice, the
respiratory tubules were disorganized (Fig. 2H, J, arrow) and/or
dilated (Fig. 2L, N), and epithelial cell nuclei were positioned at
various levels throughout the epithelium, similar to that observed
in the normal pseudostratified epithelium at E12.5. By E18.5,
peripheral regions of the lung lacked normal saccular structures,
and large cysts with detached epithelial cells were observed in
the lumen (Fig. 2P). The normally well-organized boundaries
Fig. 1. Conditional deletion of Cdc42 in fetal mouse lungs at E14.5. In triple
expressed in epithelial cells under the control of the human SFTPC promoter (A).
In the presence of doxycycline, rtTA binds to the (tetO)7CMV promoter and
activates the expression of Cre-recombinase, causing recombination and deletion
of exon 2 in Cdc42loxP/loxPmice, producing Cdc42D/Dmice (A). Arrows indicate the
primers, P1 and P2, used for RT-PCR analysis. mRNAs from Cdc42D/Dmice and
littermate controls were isolated at E12.5 (B) and E14.5 (C). RT-PCR was performed
for Cdc42 and b-actin mRNA. Densitometric quantification of the PCR products,
normalized to b-actin, demonstrated significantly decreased levels of Cdc42 mRNA
at E14.5 (C), but not at E12.5 (B). Relative mRNA levels are shown as the
H. Wan et al. / Developmental Biology 374 (2013) 46–57
Fig. 2. Morphological analysis of lung development after CDC42 deletion. Triple
transgenic mice (SP-C-rtTAwt/tg/(tetO)7CMV-Crewt/tg/Cdc42loxP/loxP) and littermate con-
trols were maintained on doxycycline from E6.5 to E12.5 or E14.5. At E12.5, no gross
differences in overall lung development were seen in whole mounts of isolated control
(A) and Cdc42D/Dlungs (B). By E14.5, however, enlarged, dilated tubular structures
were seen in the peripheral lungs of the Cdc42D/Dmice (D) compared to controls (C). At
E12.5, respiratory tubules were lined with pseudostratified epithelia in both the control
(A, E) and Cdc42D/Dmice (B, F). At E14.5, disorganized respiratory tubules were ob-
served in the lungs of Cdc42D/Dmice (H, J, L, N) compared to the littermate controls (G,
I, K, M). In controls, the peripheral epithelial buds (I, arrow) were lined by a single layer
of cuboidal cells (K), while the proximal epithelial tubes were lined by a single layer of
columnar cells (M), both with nuclei positioned along the basal aspect of the tubule. In
Cdc42D/Dmice, both peripheral (L) and proximal tubules (N) were dilated and com-
posed of epithelial cells of differing shapes and sizes, positioned at various levels within
the epithelium as demonstrated by the irregularly arranged nuclei. At E18.5, the lungs
of Cdc42D/Dmice were markedly perturbed (P) compared with controls (O). Peripheral
regions of the Cdc42D/Dlung lacked acinar structures, and the boundaries between
epithelial and mesenchymal cell compartments were lost (P, arrow). Enlarged cysts
with detached epithelial cells inside their lumens were also observed (P, arrowhead).
Fig. 3. Ultrastructural analysis of the tubular epithelium after CDC42 deletion.
Lung tissue from E14.5 Cdc42D/Dmice (B, D, F, H) and littermate controls (A, C,
E, G) was fixed and processed for electron microscopy. In the controls (A), the
tubular epithelium was organized into a single layer of columnar cells with short
microvilli at the apical cell surface and rounded nuclei located towards the basal
pole of the cell. Electron dense TJs (arrowheads) were observed on the apical–
lateral membranes of adjacent cells (A, C). In the Cdc42D/Dmice (B), the tubular
epithelium was loosely organized with irregular or staggered, elliptical or
pyramidal shaped cells and nuclei located towards the center of the cell.
Intercellular spaces were greatly expanded (B, D), and the lateral cell membranes
exhibited multiple, pseudopodia-like, cellular extensions of varying lengths,
projecting into the intercellular space (arrows) (B). Adherens junctions (AJs) were
absent, and less extensive, electron dense, TJs (arrowheads) were observed at the
apical cell junctions (B, D). In the controls (E), the more peripheral tubular
epithelium was organized into a tightly packed layer of epithelial cells with
microvilli at the apical surface and nuclei at the base of each cell. Electron dense
TJs (arrowheads) were observed at the apical cell junctions (E, G). In the Cdc42D/D
mice (F, H), the peripheral tubular epithelium was loosely organized with
irregularly shaped cells and nuclei located at different levels throughout the
epithelium. Electron dense TJs were rarely found (H, arrowhead) and AJs were
absent. mvb, multivesicular body.
H. Wan et al. / Developmental Biology 374 (2013) 46–57
between epithelial and mesenchymal cell compartments were
lost (Fig. 2P). These results demonstrate that CDC42 plays a
critical role in establishing fetal lung structures.
In order to assess the ultrastructural features of the respiratory
epithelium, lung tissue from E14.5 Cdc42D/Dmice and littermate
controls was examined by electron microscopy. In control mice,
the tubular epithelium was organized into a single layer of well
polarized columnar cells with short microvilli confined to the
apical cell surface and rounded nuclei located towards the basal
pole of each cell (Fig. 3A, E, G). Tight junctions (TJs), adherens
junctions (AJs), and lateral cell borders were clearly defined
(Fig. 3A, C, E, G; Fig. S2). In the Cdc42D/Dmice, the tubular
epithelium was loosely organized with staggered or irregularly
spaced cells, containing nuclei located towards the center of the
cell (Fig. 3B), an arrangement resembling that of the pseudos-
tratified epithelium at E12.5. Cellular contacts were disrupted, as
demonstrated by shortening or loss of apical–lateral TJs, loss
of the lateral AJs, and loss of cell-cell adhesion, resulting in
abnormally expanded intercellular spaces (Fig. 3B, D; Fig. S2).
The epithelium of the more peripheral tubules was loosely
organized, and electron dense TJs were rarely found (Fig. 3F, H).
No apparent differences in the Golgi apparatus, smooth or rough
endoplasmic reticulum, mitochondria, or the number and location
of clathrin-coated pits were observed between the controls and
the Cdc42D/Dmice. These data demonstrate that deletion of
CDC42 disrupted polarized nuclear positioning and cell junctions,
as well as organization of the developing respiratory tubules.
CDC42 is required for temporal–spatial regulation of epithelial cell
polarity during lung branching morphogenesis
Cell polarity associated proteins play important roles in the
organization of epithelial structures (Tepass et al., 2001). In order
to determine if loss of CDC42 altered epithelial polarity, staining
for discs large homolog 1 (DLG1), partitioning-defective homolog
3 (PAR3) and epithelial cell surface domain-associated proteins,
including zonula occludens 1 (ZO-1) and E-cadherin, was assessed
at E14.5. In controls, DLG1, a basal–lateral membrane marker, was
detected primarily at the basal–lateral surfaces of established
epithelial tubules (Fig. 4A, C, D), with more intense staining at the
basal membrane (Fig. 4A, C, arrows). Diffuse cytoplasmic staining
for DLG1 was observed in epithelial cells at the growing tips of the
peripheral tubules (Fig. 4A, D arrowhead), suggesting that the
epithelial cells located in the budding tips were more weakly
polarized than those in the more proximal tubules. After deletion
of CDC42, DLG1 was absent or abnormally distributed in the
epithelium (Fig. 4B, E, F). In E14.5 control lungs, PAR3, an apical
polarity marker required for junction formation (Zen et al., 2009),
was detected at apical epithelial cell junctions (Fig. 4G). After
deletion of CDC42, staining for PAR3 was lost or decreased in the
epithelium (Fig. 4H). Likewise, staining for the tight junction-
associated protein, ZO1, was decreased in the Cdc42D/Dmice
(Fig. 4J) compared to controls (Fig. 4I). E-cadherin staining was
restricted to the lateral membranes of epithelial cells in E14.5
control lungs (Fig. 4K). After deletion of CDC42, ectopic staining
for E-cadherin (E-cad) was observed at the apical cell surface
(Fig. 4L, arrow), as well as in its normal location at the lateral cell
membrane. Consistent with these findings, staining with phalloi-
din demonstrated that the polarized distribution of F-actin,
normally found at the apical and basal surfaces of the respiratory
epithelium, was significantly disrupted in the Cdc42D/Dmice at
E14.5, compared with controls (Fig. S3). These results indicate
that CDC42 is required for proper regionalization of epithelial cell
membrane domains, as well as the actin cytoskeleton. RAC1 is
another Rho GTPase that is important for organization of the actin
cytoskeleton (Rathinam et al., 2011). Interestingly, no changes in
Fig. 4. Analysis of epithelial cell polarity-associated proteins after CDC42 deletion.
Lung sections from E14.5 Cdc42D/Dmice (B, E, F, H, J, L) and littermate controls
(A, C, D, G, I, K) were immunostained for DLG1 (A–F), PAR3 (G, H), ZO-1 (I, J) and
E-cad (K, L). Nuclei were counterstained with DAPI (C–F). In controls, DLG1 was
detected at the basal–lateral surfaces of the tubular epithelial cells (A, C, arrow)
and in the cytoplasm of epithelial cells at the budding tips (A, D arrowhead). In
Cdc42D/Dmouse lungs, ectopic staining for DLG1 was detected in the epithelial
cells of both established tubules (E, F, arrow) and budding tips (B, arrowhead). In
controls, PAR3 (G, arrow) and ZO-1 (I, arrow) were detected at the apical junctions
of the tubular epithelial cells, while loss of staining for PAR3 (H) and ZO-1 (J) was
observed in the tubular epithelial cells of the Cdc42D/Dmice. In both control
(K) and of Cdc42D/Dmice (L), E-cadherin was localized to the lateral cell
membranes. Ectopic E-cadherin staining was detected at the apical cell surface
in the Cdc42D/Dmice (L). Insets are higher magnification illustrations of cells
indicated by the arrows. AP indicates apical side of the epithelium.
H. Wan et al. / Developmental Biology 374 (2013) 46–57
RAC1 protein levels were detected at E14.5 in the CDC42 deficient
lung (Fig. S4).
Precise temporal–spatial expression and polarized secretion of
both ECM and functional proteins by respiratory epithelial cells
are required for formation and function of the lung (McGowan,
1992; Schuger et al., 1990; Weaver et al., 2003). In controls, the
basement membrane-associated proteins, fibronectin 1 (FN1) and
laminin 5 (LAMA5), were detected at the basal surface of the
respiratory tubules, while ectopic localization of FN1 and LAMA5
was observed in Cdc42D/Dmice at E14.5 (Fig. 5A-D), indicating
that CDC42 was required for polarized deposition of ECM pro-
teins. The propeptides for both surfactant protein C (proSP-C) and
surfactant protein B (proSP-B) are normally routed to lamellar bodies
at the apical pole of peripheral epithelial cells, where they are
proteolytically processed to the mature peptides, mSP-C and mSP-B,
and then secreted into the alveolar lumen along with surfactant
phospholipids during late gestation and after birth. At E14.5, proSP-C
was localized to the apical regions of epithelial cells lining the
peripheral respiratory tubules of control mice (Fig. 5E). In contrast,
proSP-C was distributed diffusely throughout the cytoplasm of
the epithelial cells of Cdc42D/Dmice at E14.5 (Fig. 5F). Functional
maturation of the lung during late gestation is associated with
increased expression, processing, and secretion of mSP-B. At E18.5,
mSP-B was detected primarily at the apical surface of peripheral
epithelial cells and in the alveolar spaces of control mice (Fig. 5G).
Although mSP-B was readily detected in peripheral epithelial cells
after deletion of CDC42, it was distributed diffusely throughout the
cytoplasm of the cells and was not detected in the alveolar spaces,
indicating that directional trafficking and secretion of mSP-B by
alveolar type II cells was disrupted by deletion of CDC42 (Fig. 5H).
These results demonstrate that CDC42 is required for normal basal
and apical routing of secreted proteins, a process critical for lung
development and function.
CDC42 influences epithelial cell stratification during cell proliferation
During organ formation, epithelial cell polarity is coordinated
with diverse cellular processes, including proliferation, cell shape
changes, and migration (Bryant and Mostov, 2008). In previously
published studies, Interkinetic Nuclear Migration (INM), defined
as periodic nuclear migration along the apical–basal axis of the
cell that is in phase with the cell cycle, was associated with
changes in cell polarity and proliferation and was required for
organ development and cell fate specification during neuronal,
hepatic, and intestinal development (Ayala et al., 2007; Bort et al.,
2006; Sauer, 1936; Solecki et al., 2006). In this study, we found
that proliferation in the respiratory tubules was associated with a
process similar to INM, involving nuclear and/or cell migration
between the basal and apical axis of the tubular epithelium
during progression of the cell cycle. At E11.5 and E12.5, the
respiratory tubules of control mice were highly proliferative and
lined by a pseudostratified epithelium containing cells of various
shapes with nuclei located at different positions within the
epithelium (Fig. 6I; Fig. S5). Nuclei in different phases of the cell
cycle were identified by their nuclear morphology after staining
for BrdU, pHH3, and/or DAPI (Goto et al., 2002). S-phase nuclei,
identified after pulse labeling and staining for BrdU, were located
in the basal regions of the epithelium (Fig. 6I-A). Nuclei in late G2
exhibited focal nuclear staining for pHH3 and were located in
cells found in central regions of the epithelium (Fig. 6I-B). pHH3
positive nuclei in various stages of M phase, including prometa-
phase, metaphase, and telophase, were found in cells located in
more apical regions of the epithelium (Fig. 6I-C, D, E). A gradual
decline in nuclear staining for pHH3 was noted at telophase in
cells located near the luminal surface (Fig. 6I-E, F). These data
indicated that temporal–spatial positioning of nuclei and/or cells
is precisely regulated during cell division. At E14.5 and E16.5,
epithelial cell proliferation decreased, and the respiratory tubules
were lined by a single layer of highly polarized columnar
(proximal) and cuboidal (peripheral) cells, wherein most of the
nuclei were located in the basal region of the epithelium (Fig. S5).
As observed at E11.5–E12.5, S-phase nuclei were detected in basal
regions of epithelium, and metaphase nuclei were detected in
apical regions of the epithelium at the luminal surface (Fig. S5).
Although this pattern of interkinetic nuclear and/or cell migration
was characteristic of epithelial cells in highly proliferative regions
of pseudostratified epithelia lining the respiratory tubules at
E11.5 and E12.5, similar repositioning of S phase and metaphase
nuclei and/or cells in the respiratory tubules was observed during
later stages of lung development (Fig. S5).
Fig. 5. Analysis of secretory proteins required for lung maturation and function
after CDC42 deletion. In E14.5 control lungs (A, C), immunostaining for the
basement membrane proteins, FN1 (A) and LAMA5 (C), was localized primarily
to the basal lamina of the tubular epithelium (arrows). In E14.5 Cdc42D/Dmouse
lungs, FN1 (B) and LAMA5 (D) staining was markedly decreased in the basal
lamina of epithelial cells. In addition, ectopic accumulation of FN1 was detected
inside many of the epithelial cells (B, arrows), while FN1 staining was also
decreased in the mesenchyme (B, arrowhead). Immunostaining for proSP-C,
normally detected in apical regions of the epithelial cells lining the peripheral
respiratory tubules at E14.5 (E), was diffusely distributed throughout the cyto-
plasm of the epithelial cells in Cdc42D/Dmice (F). At E18.5, immunostaining for
mSP-B was detected in the lumen (G, arrowhead) and at the apical cell surface (G,
arrow) in control lungs, while diffuse intracellular staining for mSP-B was
observed in Cdc42D/Dlungs (H, arrow).
H. Wan et al. / Developmental Biology 374 (2013) 46–57
In order to determine if the polarized distribution of specific
membrane domain-associated proteins was coordinately modulated
during the cell cycle, staining for E-cadherin and DLG1 was
performed at different stages of branching morphogenesis (E11.5–
E14.5), and the nuclei were counterstained with DAPI. In polarized
columnar epithelial cells, staining for E-cadherin was restricted to
the lateral cell surfaces (Fig. 6I–G). As the cell cycle progressed, the
cells became more spherical and staining for E-cadherin gradually
extended around the entire circumference of the dividing cells
(Fig. 6I–H, I, J, K). Likewise, the distribution of DLG1 was altered
during the cell cycle. Although staining for DLG1 was restricted to
the basal–lateral surface of polarized columnar epithelial cells
(Fig. 6I–L, M, N, O), weak, diffuse staining for DLG1 was detected
in restricted basal–lateral regions of mitotic cells (Fig. 6I–M, N, O).
These data support the concept that epithelial polarity, as indicated
by concurrent changes in nuclear and/or cell positioning and
epithelial cell membrane domain specification, was integrated with
cell cycle progression during branching morphogenesis of the lung.
In order to assess the mechanisms underlying the structural
changes caused by CDC42 deletion, the temporal–spatial coupling
Fig. 6. Dynamic changes in epithelial cell polarity during cell cycle progression after CDC42 deletion. (I) Modulation of epithelial cell polarity during the cell cycle in
normal fetal lung. (A–F) Lung sections from E11.5 wild type C57/B6 mice were immunostained for BrdU, (A) or pHH3 (B–F), and counterstained with DAPI. Nuclei in
different phases of the cell cycle were identified by their nuclear morphology after staining for BrdU (yellow), pHH3 (green) and DAPI (blue). Proliferation-associated
changes in nuclear position were observed throughout the cell cycle. S-phase nuclei stained with BrdU were located in basal regions of the epithelial cells (A). Nuclei in late
G2 exhibited focal nuclear staining with pHH3 and were located in central regions of the cells (B). Nuclei in prometaphase and metaphase exhibited intense pHH3 staining
and were located near the apical pole of the cells (C, D). A gradual decline in pHH3 staining was noted at telophase (E) and during cytokinesis (F) when compared to
staining in the S/G2 phases of the cell cycle, during which nuclei of dividing cells were located in the basal regions of the cells (A, B). (G–K) Lung sections from E11.5 wild
type mice were stained for E-cad (green) and counterstained with DAPI (blue). Cell cycle phases were identified by nuclear staining patterns using DAPI. In columnar
epithelial cells, in which nuclei were located in the basal region, E-cadherin staining was restricted to the lateral surface (G). From prometaphase through cytokinesis
(H–K), the more spherical, mitotic, epithelial cells were outlined by E-cad staining, which gradually extended around the entire circumference of the cell (I–K). (L–O) Lung
sections from E14.5 wild type mice were stained for DLG1 (green) and DAPI (blue). Intense staining for DLG1 was detected at basal membranes of columnar epithelial cells
whose nuclei were located at or near the base of the cell (L, arrow). Diffuse staining for DLG1 (arrow) was observed in basal–lateral regions of proliferating epithelial cells
at metaphase (M), telophase (N), and during cytokinesis (O). Dotted lines delineate cell boundaries, drawn from background images acquired in the Texas-Red fluorescence
channel. AP indicates apical side of the epithelium. (II) Deletion of CDC42 disrupts epithelial polarity during the cell cycle. Lungs from Cdc42D/Dmice and littermate
controls were pulse labeled with BrdU for 30 min before sacrificing at E14.5. Embryonic lungs were dissected, washed, and embedded. Lung sections were stained for BrdU
(A, B), pHH3 (C, D) or DLG (E, F, G), and the nuclei were counterstained with DAPI. In controls, BrdUþnuclei (yellow), representing S phase, were located in basal regions of
the epithelial cells (A). In contrast, apical localization of BrdUþnuclei was observed in epithelial tubes of Cdc42D/Dmice (B, arrow). In controls, metaphase nuclei stained
intensely for pHH3 (green) and were located in the apical regions of the epithelium (C). In Cdc42D/Dmice, metaphase nuclei were often observed in a basal location
(D, arrow). In controls, weak basal–lateral staining for DLG1 was barely detected in apically located cells (E, arrow). In Cdc42D/Dmice, strong cytoplasmic (F, arrow) or
apical staining (G, arrow) for DLG1 was detected with abnormally intense apical staining observed in epithelial cells with M phase nuclei (G, arrowhead). Dotted lines
delineate apical surfaces (A, B) or basal surfaces of tubular epithelium (E, F, G).
H. Wan et al. / Developmental Biology 374 (2013) 46–57
of epithelial polarity with cell cycle progression was analyzed in
lungs of Cdc42D/Dmice and littermate controls at E14.5. Consis-
tent with the findings in wild type mice, pulse-labeled, BrdU-
stained nuclei in the control mice were located in basal regions of
the tubular epithelium at E14.5 (Fig. 6II-A), while all nuclei with
intense pHH3 staining, typical of metaphase nuclei, were located
in apical regions (Fig. 6II-C). In contrast, significant numbers of
irregularly arranged BrdU-stained nuclei were found in non-basal
regions of the tubular epithelium in Cdc42D/Dmice (Fig. 6II-B,
arrow), indicating that these epithelial cells entered S phase
without repositioning of the nuclei and/or cells to the basal
region of the epithelium, as normally occurs after completion of
cytokinesis. Basally located metaphase nuclei were also identified
in Cdc42D/Dmice (Fig. 6II-D) in comparison to the controls, in
which metaphase nuclei were found only in the apical regions of
the epithelium (Fig. 6II-C). Although staining for DLG1 was barely
detected on basal–lateral membranes of apically located non-
mitotic cells in the controls (Fig. 6II-E, arrow), intense cytoplasmic
staining for DLG1 was detected in apically located cells after
deletion of CDC42 (Fig. 6II-F, G). These data demonstrate that
both cell membrane specification and nuclear/cell positioning
associated with the cell cycle are regulated by CDC42 during
formation of the lung.
CDC42 regulates directional cell division during normal lung
To test whether CDC42 regulated mitotic spindle orientation,
parallel versus perpendicular cell division was quantified in the
lungs of control and Cdc42D/Dmice (Fig. 7A–E). At E 14.5, the ratio
of parallel to perpendicular cell division was markedly decreased
in Cdc42D/Dmice, compared with controls (Fig. 7F). Thus, differ-
ences in mitotic spindle orientation during cell division are likely
to contribute to the abnormalities in branching morphogenesis
observed after deletion of CDC42.
CDC42 is not required for proximal-distal specification
or cytodifferentiation of the respiratory epithelium
No differences in immunostaining for SOX2 or SOX9 were
observed between the controls and the Cdc42D/Dmice at E14.5,
indicating that proximal-distal specification of the respiratory
tubules was not perturbed despite disruption of branching mor-
phogenesis (Fig. S6). To address the role CDC42 in cytodifferentia-
tion of respiratory epithelial cells, epithelial cell markers and/or
functional proteins that distinguish ciliated and non-ciliated
(Clara cells) bronchiolar cells, as well as alveolar type I and type
II cells, were assessed at E18.5 (Fig. 8). Although lung structure
was dramatically disrupted in the Cdc42D/Dmice, thyroid tran-
scription factor 1 (TTF1), a marker for the respiratory epithelium,
and mSP-B, a marker for alveolar type II cells, were detected in the
peripheral respiratory epithelium (Fig. 8A–D). Staining for Clara
cell secretory protein (CCSP) and the transcription factor, FOXJ1, a
ciliated cell marker, was also detected in the conducting airways
of the Cdc42D/Dmice (Fig. 8E–H). While alveolar saccules were
markedly disrupted and squamous alveolar type I cells were
absent (Fig. 8J), mRNA expression for the type I alveolar epithelial
cell markers, aquaporin 5 (Aqp5) and T1-alpha (T1-a), were read-
ily detected in the lungs of the Cdc42D/Dmice, with a slight but
significant decrease in T1-a mRNA (Fig. 8K, IL). Taken together, the
expression of various epithelial cell differentiation markers in
both conducting airways and peripheral regions of the lung was
generally maintained after deletion of CDC42.
CDC42 regulates paracrine signaling required for normal lung
To test whether differentiation and patterning of mesenchy-
mal cells were regulated by CDC42, fetal lung tissue was stained
for a-smooth muscle actin (a-SMA). Smooth muscle cells were
detected surrounding the stalks of the respiratory epithelial
tubules, as indicated by the circumferential staining of a-SMA
observed at E14.5 in control mice (Fig. 9A-a). In contrast, a-SMA
staining was markedly decreased and diffusely localized in the
lung mesenchyme of Cdc42D/Dmice at E14.5 (Fig. 9A-b). Sonic
hedgehog (SHH), expressed in respiratory epithelial cells of the
fetal lung, provides paracrine signals required for branching
morphogenesis and differentiation of airway smooth muscle cells
(Pepicelli et al., 1998; Weaver et al., 2003). Consistent with the
abnormal distribution of airway smooth muscle cells observed
after deletion of CDC42, mRNAs for Shh and its receptor Ptch1
were decreased at E14.5 (Fig. 9C–E). In contrast, no abnormalities
in staining for CD34, a marker for the capillary endothelial cells,
were detected after deletion of CDC42 (Fig. S7).
Fibroblast growth factor 10 (FGF10) is normally expressed in a
precise pattern in mesenchymal cells adjacent to epithelial cells
at the tips of respiratory tubules during early lung development.
FGF10 expression in fetal lung mesenchyme activates its receptor,
FGFR2IIIB, in the epithelium, serving as a chemotactic factor
required for lung branching morphogenesis. In turn, FGF10
influences SHH signaling required for normal lung development
(Morrisey and Hogan, 2010). FGF10 expression was assessed by
in situ hybridization at E14.5 in the Cdc42D/Dand control mice
(Fig. 9). The distribution of Fgf10 mRNA expression in the
mesenchyme of the Cdc42D/Dmice was more extensive than in
the controls, expanding to surround the abnormally dilated
peripheral lung tubules (Fig. 9A-c, d). FGF10 signaling activates
ERK phosphorylation to regulate cell migration and proliferation
required for branching morphogenesis. Consistent with the
expanded FGF10 expression observed above, phosphorylation of
ERK was significantly increased in lungs from Cdc42D/Dmice at
E14.5 (Fig. 9B). These results indicate that deletion of CDC42
Fig. 7. Deletion of CDC42 disrupts mitotic spindle orientation. Lung sections from
E14.5 Cdc42D/Dand littermate mice were stained with DAPI. Single or paired
condensed anaphase and telophase nuclei (CN) were visualized. (A) The diagram
illustrates how mitotic spindle poles (MSP) and mitotic angles (y) were measured
with respect to the plane of the basement membrane. MSP was determined from
the line linking the midpoints of the paired CN or by an axis perpendicular to the
longitudinal axis of the CN. y is the projection angle of the MSP onto the plane of
basement membrane. Parallel division was defined as 0ryr30 and perpendi-
cular division as 60ryr90. (B–E) Nuclei undergoing parallel division and
perpendicular division (arrow) are shown. The basement membranes are high-
lighted with dotted lines. (F) A bar graph showing the ratio of parallel to
perpendicular division in epithelial cells from lungs of control and CDC42D/Dmice
are shown as the mean7s.e.m.,npo0.05.
H. Wan et al. / Developmental Biology 374 (2013) 46–57
altered SHH activity, which inhibits FGF10 activity during normal
lung development (Morrisey and Hogan, 2010). Disruption of
these reciprocal, signaling pathways is, therefore, another factor
that is likely contribute to the abnormalities in branching mor-
phogenesis observed in the Cdc42D/Dmice.
While branched structures are formed in some organs by the
process of tubulogenesis from non-polarized cells, as occurs in the
pancreas and mammary gland, tubular structures in other organs,
including the lung, are derived from existing, highly polarized,
epithelial tubes (Chung and Andrew, 2008). The respiratory tract
is derived from the foregut endoderm, a polarized epithelial tube.
Once the lung bud forms (E9.0–E9.5), it undergoes proximal to
distal elongation with subsequent expansion and bifurcation
of the growing buds to form the respiratory tubules of the fetal
lung (Cardoso and Lu, 2006). Reiteration of this process during
fetal lung development gives rise to the highly branched, tubular
network required for gas exchange at birth (Metzger et al., 2008).
Several different geometrical modes of proximal-to-distal branch-
ing of the respiratory tubules are used during branching morpho-
genesis of the lung. These include domain branching, which
forms the proximal respiratory tubules, followed by many rounds
of planar and orthogonal bifurcation, which form the more distal
tubules (Metzger et al., 2008). During this process, the respiratory
epithelium evolves from a pseudostratified epithelium to a simple,
single cell-layered, columnar epithelium in the proximal conduct-
ing airways and, finally, to a simple squamous epithelium in the
peripheral airspaces (Ten Have-Opbroek, 1991). The complex
morphogenetic processes involved in these transitions are depen-
dent on the combined behaviors of epithelial cells. The mech-
anisms by which individual epithelial cells retain positional
information during branching morphogenesis, i.e., by modulating
both cell–cell contacts and apical–basal polarity during cell
proliferation, cell shape changes, cell–cell and cell–matrix inter-
actions, and cell migration, are poorly understood (Warburton
et al., 2010). In the present study, deletion of CDC42 demon-
strated its critical role in the maintenance of spatial information
and positioning of developing epithelial cells, and, thereby, in
structural patterning of the lung. CDC42 was shown to be critical
Fig. 8. Respiratory epithelial cell differentiation after CDC42 deletion. Lungs from Cdc42D/Dmice and littermate controls were harvested at E18.5. Lung sections were
stained with cell type-specific markers, including TTF1 (A, B), mature SP-B (C, D), CCSP (E, F), FOXJ1 (G, H) and pan-cytokeratin (I, J). Cell type specific markers were readily
detected in bronchiolar (CCSP, FOXJ1) or alveolar (TTF1, mature SP-B) epithelial cells of E18.5 Cdc42D/Dmice (B, D, F, H, J) and littermate controls (A, C, E, G, I). In the
controls, dilated peripheral saccules were lined by squamous type I cells (I, arrow) and cuboidal type II cells (I, arrowhead). In Cdc42D/Dmice, the peripheral saccules were
absent (J). Pan-cytokeratin-stained epithelial cells were cuboidal (arrowhead), while squamous epithelial cells were rarely detected. No differences in Aqp5 mRNA were
detected in Cdc42D/Dmouse lungs at E18.5 (K). T1-a mRNA, however, was slightly, but significantly, decreased in the lungs of E18.5 Cdc42D/Dmice (L). Relative mRNA levels
are shown as the mean7s.e.m.,npo0.05.
H. Wan et al. / Developmental Biology 374 (2013) 46–57
for coordinating proliferation with epithelial cell polarity, cell–
cell and cell–matrix interactions, and paracrine signaling upon
which lung formation depends.
CDC42 is required for maintenance of epithelial cell polarity in the
Cell polarity is an intrinsic feature of epithelial structures,
determining cell position, shape, and directional secretion of
proteins, which in turn, define apical, basal and lateral cell
surfaces and membrane structures required for proper cell–cell
and cell–matrix interactions (Bryant and Mostov, 2008). Epithelial
cell polarity depends upon the directional trafficking of proteins
to specific cellular sites, thereby establishing distinct apical and
basal–lateral domains in the cell (Zhang et al., 2001). Immuno-
detection of polarity-associated proteins, such as E-cadherin,
PAR3, ZO-1, and DLG1, in control mice indicated that the tubular
epithelium was highly polarized at E14.5. In the Cdc42D/Dmice,
decreased expression for CDC42 was correlated with loss and/or
ectopic localization of these markers, indicating that CDC42 was
required for the maintenance of epithelial cell polarity. Ultra-
structural alterations in TJs and loss of AJs at the apical–lateral
cell surface in the Cdc42D/Dmice further supports the essential
role of CDC42 in maintaining polarized epithelial cell surface
domains and related membrane structures. These findings were
consistent with those found in vitro demonstrating that CDC42 is
a major determinant of apical localization of the PAR6–PAR3–
aPKC complex, which is required for establishing and maintaining
apical polarity and cell junctions (Baum and Georgiou, 2011;
Joberty et al., 2000; Rojas et al., 2001; Simmons, 1992; Welchman
et al., 2007).
Epithelial cell polarity, directional trafficking of proteins to
specific cellular sites and membrane domains, and protein secre-
tion into the extracellular space, are dependent upon organization
of the actin cytoskeleton (Ayscough et al., 1997; Novick and
Botstein, 1985; Pruyne et al., 1998). In the present study, routing
of specific basement membrane proteins was impaired, a result
that was consistent with disruption of the actin cytoskeleton and
epithelial cell polarity after deletion of CDC42, a finding also
observed in cultured MDCK cells (Kroschewski et al., 1999).
Deletion of CDC42 disrupted the polarized secretion of LAMA5
and FN1, which are ECM proteins required for formation of the
basement membrane, for circumferential alignment of airway
smooth muscle cells in the conducting airways, and for stabiliza-
tion of the more proximal respiratory tubules (McGowan, 1992;
Tran et al., 2006). Thus, the loss of polarized secretion of these
ECM proteins is likely to contribute to the disruption of lung
morphogenesis in the CDC42 deficient mice. These findings
indicate that CDC42 is required for the directional trafficking of
proteins critical for the organization of lung architecture and
function of the lung and are consistent with previous studies
demonstrating that CDC42 is required for remodeling of other
epithelial structures (Jaffe et al., 2008; Kesavan et al., 2009; Rojas
et al., 2001).
CDC42 is required for transition from a pseudostratified to a simple,
single cell-layered epithelium during lung morphogenesis
During early lung branching morphogenesis, the tubular
epithelium is highly polarized, with a basement membrane
adjacent to the basal surface and intercellular junctions in the
apical regions of lateral cell contact. Tubular epithelial cell
polarity changed dynamically with cell proliferation throughout
branching morphogenesis, as indicated by the temporal–spatial
coupling of nuclear and/or cell positioning with the distribution of
polarity-associated proteins and specific phases of the cell cycle.
In agreement with previous studies, demonstrating that Inter-
kinetic Nuclear Migration associated with proliferation is an
intrinsic feature of developing pseudostratified epithelia (Del
Bene, 2011; Fish et al., 2008), migration of nuclei and/or cells to
apical regions of the tubular epithelium during mitosis was
observed in the pseudostratified epithelium lining the respiratory
tubules of control mice during early branching morphogenesis.
Likewise, polarized E-cadherin and DLG1 staining patterns were
altered as proliferating cells moved from basal to apical regions of
the respiratory tubule epithelium. These data indicate that
nuclear and/or cell positioning and membrane domain specifica-
tion changed coordinately during cell proliferation, the precise
regulation of which may be required for formation and main-
tenance of tubular structures in the developing lung.
The embryonic lung (E11.5–E12.5) is lined by a pseudostrati-
fied epithelium composed of highly proliferative epithelial cells.
Fig. 9. Deletion of CDC42 disrupts paracrine signaling during branching morpho-
genesis. (A) Overall, staining for a-SMA was decreased in the mesenchyme
adjacent to the respiratory tubules of E14.5 Cdc42D/Dmice (b) when compared
to controls (a). The well-organized a-SMA staining pattern surrounding the
respiratory tubules of the control mice (a) was also disrupted in the Cdc42D/D
mice, exhibiting diffuse focal regions of mesenchymal staining between the
tubules (b). In control mice, Fgf10 mRNA was expressed in the mesenchyme at
the periphery of the lung, appearing in a discontinuous arch around the tips
of the respiratory tubules (c). In the Cdc42D/Dmice, Fgf10 mRNA expression was
localized to broad regions of mesenchyme surrounding the abnormal respiratory
tubules (d). (B) Immunoblots with antibodies for CDC42, pERK, and GAPDH were
performed on lung homogenates from E14.5 Cdc42D/Dmice and littermate
controls. Significantly increased phosphorylated ERK was detected in the lungs
of Cdc42D/Dmice. (C–E) RT-PCR, normalized to b-actin, was used to quantify Shh
and Ptch1 mRNAs in lungs from Cdc42D/Dmice and littermate controls at E14.5.
Shh and Ptch1 mRNAs were decreased significantly in the lungs of Cdc42D/Dmice
(D–E). Relative mRNA levels are shown as the mean7s.e.m.,npo0.05.
H. Wan et al. / Developmental Biology 374 (2013) 46–57
By E14.5, epithelial cell proliferation decreases and most of the
respiratory tubules are lined by a single layer of highly polarized
epithelial cells with nuclei located at the basal pole of the cells.
Although epithelial proliferation was significantly decreased,
deletion of CDC42 in epithelial progenitor cells disrupted coordi-
nate regulation of both spatial positioning of nuclei and/or cells
and distribution of polarity proteins during the cell cycle. As a
result, transition from a pseudostratified epithelium to a simple
columnar epithelium, a process that normally occurs between
E12.5 and E14.5 (Ten Have-Opbroek, 1991), was absent in the
Cdc42D/Dmice. Thus, CDC42 is a critical regulator of cell polarity
in the developing lung and is required for spatial coordination of
nuclear and/or cell positioning during cell proliferation.
Recently loss of the EYA1 protein disrupted epithelial cell
polarity and mitotic spindle orientation, resulting in abnormal
lung development (El-Hashash et al., 2011). Significantly, changes
in mitotic spindle orientation/angle were also demonstrated in
the CDC42 deficient lung, suggesting that loss of CDC42 alters
directional cell division, resulting in abnormal branching of the
developing lung. Similarly, disruption of the planar cell polarity
signaling pathway, which may function upstream of the Rho
GTPases, also impaired branching morphogenesis and cellular
organization of the peripheral lung (Yates et al., 2010).
CDC42 is not required for proximal-distal specification and epithelial
Although normal respiratory tubules were formed at E12.5 in
the Cdc42D/Dmice, organization of the respiratory tubules was
demonstrating that CDC42 is required for formation and main-
tenance of tubular structures. The conditional system used to
delete CDC42 depends upon the activity of the human SFTPC
promoter element, which is active during the early embryonic
stage of lung development, i.e., before branching morphogenesis
of the lung is initiated (Perl et al., 2002). Loss of CDC42 expres-
sion, however, was not detected at E12.5, consistent with the lack
of histological changes in the developing lungs of the Cdc42D/D
mice at this time. By E14.5, both CDC42 mRNA and protein levels
were decreased significantly and were correlated with significant
changes in the histological organization of the epithelium and in
expression patterns of various polarity-associated proteins. Thus,
whether CDC42 is required for initiation or maintenance of early
lung bud formation between E9.5 and E12.5 remains unclear.
Despite the marked disruption of lung architecture observed after
deletion of CDC42, both the proximal-distal axis of the lung and
expression of epithelial cell type specific markers were main-
tained. The latter finding is distinct from that reported for the
pancreas, skin, and developing mouse brain, wherein progenitor
cell fate specification was disrupted by loss of CDC42 (Cappello
et al., 2006; Kesavan et al., 2009; Wu et al., 2006).
Paracrine signaling required for branching morphogenesis
is dependent upon CDC42
Lung branching morphogenesis requires precise regulation of
reciprocal signaling interactions between the respiratory epithe-
lium and the mesenchyme. Spatially restricted expression of SHH
in the epithelium and its extra-cellular gradient provides infor-
mation that directs differentiation and patterning of vascular
smooth muscle, pulmonary blood vessels and other stromal
elements along the developing bronchiolar structures of the lung
(Morrisey and Hogan, 2010). Previous studies indicated that SHH
could be functionally linked to CDC42 signaling through regula-
tion of the actin cytoskeleton (Bijlsma et al., 2007; Sasaki et al.,
2010; Xiao et al., 2010). In the present study, Shh and Ptch1
mRNAs were significantly decreased, and the expression domain
of FGF10 in the mesenchyme was expanded. These data are
consistent with previous studies demonstrating that FGF10
directs epithelial cell migration associated with tubular elonga-
tion (Bellusci et al., 1997). These findings support the concept
that loss of CDC42 altered lung morphogenesis, at least in part,
by disrupting normal epithelial–mesenchymal interactions that
are dependent on reciprocal FGF/SHH signaling (Morrisey and
An important link between FGF signaling, epithelial cell
polarity, and branching morphogenesis was recently provided
by mathematical modeling, which demonstrated that FGFR-ERK1/
2 signaling in the developing respiratory epithelium is critical for
normal stereotypic budding and branching of the developing lung
(Tang et al., 2011). In our model, phosphorylation of ERK was
significantly increased in lungs from Cdc42D/Dmice at E14.5
(Fig. 8B), suggesting that CDC42 regulates FGFR-ERK1/2 activity,
which, in turn, modulates signaling between the SHH and FGF
pathways during normal lung development.
Currently, there is considerable interest in the process of
cell-based therapies for various lung disorders. Recent studies
have demonstrated the feasibility of introducing lung progenitor
cells into appropriate cellular microenvironments to enhance
regeneration of functional tissue for repair of lung parenchyma
(Kajstura et al., 2011). The present study demonstrates the critical
role of CDC42 in the regulation of complex cellular processes,
which must be coordinated during remodeling of the fetal lung
and may be highly relevant to lung repair and remodeling after
injury. Understanding the role of CDC42 in cell polarity, prolifera-
tion, signaling, migration, and tissue organization may provide
important insights useful in regenerative medicine for pulmonary
The authors gratefully acknowledge support from the Program
for Changjiang Scholars and Innovative Research Team in Uni-
versity (PCSIRT, No. 0935), Liqian Zhang and Paula Blair for
technical support. This research was supported by internal funds
from the West China Second University Hospital, Grants from the
National Natural Science Foundation of China (NSFC30700339,
H.W.), Fundamental Research Funds for the Central University
(2010SCU23009, H.W.), the Key Clinical Project of the Chinese
Ministry of Health (No. 2010439, H.W.), and the National Insti-
tutes of Health, (HL090156, and HL110964, J.A.W., S.E.W.).
Appendix A. Supplementary information
Supplementary data associated with this article can be found in
the online version at http://dx.doi.org/10.1016/j.ydbio.2012.11.030.
Affolter, M., Zeller, R., Caussinus, E., 2009. Tissue remodelling through branching
morphogenesis. Nat. Rev. Mol. Cell Biol. 10, 831–842.
Ayala, R., Shu, T., Tsai, L.H., 2007. Trekking across the brain: the journey of
neuronal migration. Cell 128, 29–43.
Ayscough, K.R., Stryker, J., Pokala, N., Sanders, M., Crews, P., Drubin, D.G., 1997.
High rates of actin filament turnover in budding yeast and roles for actin in
establishment and maintenance of cell polarity revealed using the actin
inhibitor latrunculin-A. J. Cell Biol. 137, 399–416.
H. Wan et al. / Developmental Biology 374 (2013) 46–57
Bellusci, S., Grindley, J., Emoto, H., Itoh, N., Hogan, B.L., 1997. Fibroblast growth
factor 10 (FGF10) and branching morphogenesis in the embryonic mouse lung.
Development 124, 4867–4878.
Baum, B., Georgiou, M., 2011. Dynamics of adherens junctions in epithelial
establishment, maintenance, and remodeling. J. Cell Biol. 192, 907–917.
Bijlsma, M.F., Borensztajn, K.S., Roelink, H., Peppelenbosch, M.P., Spek, C.A., 2007.
Sonic hedgehog induces transcription-independent cytoskeletal rearrange-
ment and migration regulated by arachidonate metabolites. Cell Signalling
Bort, R., Signore, M., Tremblay, K., Martinez Barbera, J.P., Zaret, K.S., 2006. Hex
homeobox gene controls the transition of the endoderm to a pseudostratified,
cell emergent epithelium for liver bud development. Dev. Biol. 290, 44–56.
Bryant, D.M., Mostov, K.E., 2008. From cells to organs: building polarized tissue.
Nat. Rev. Mol. Cell Biol. 9, 887–901.
Cappello, S., Attardo, A., Wu, X., Iwasato, T., Itohara, S., Wilsch-Brauninger, M.,
Eilken, H.M., Rieger, M.A., Schroeder, T.T., Huttner, W.B., Brakebusch, C., Gotz,
M., 2006. The Rho-GTPase cdc42 regulates neural progenitor fate at the apical
surface. Nat. Neurosci. 9, 1099–1107.
Cardoso, W.V., Lu, J., 2006. Regulation of early lung morphogenesis: questions,
facts and controversies. Development 133, 1611–1624.
Chung, S., Andrew, D.J., 2008. The formation of epithelial tubes. J. Cell Sci. 121,
Clark, J.C., Tichelaar, J.W., Wert, S.E., Itoh, N., Perl, A.K., Stahlman, M.T., Whitsett,
J.A., 2001. FGF-10 disrupts lung morphogenesis and causes pulmonary
adenomas in vivo. Am. J. Physiol. Lung Cell. Mol. Physiol. 280, L705–L715.
Dave, V., Wert, S.E., Tanner, T., Thitoff, A.R., Loudy, D.E., Whitsett, J.A., 2008.
Conditional deletion of Pten causes bronchiolar hyperplasia. Am. J. Respir. Cell.
Mol. Biol. 38, 337–345.
Del Bene, F., 2011. Interkinetic nuclear migration: cell cycle on the move. EMBO J.
El-Hashash, A.H., Turcatel, G., Al Alam, D., Buckley, S., Tokumitsu, H., Bellusci, S.,
Warburton, D., 2011. Eya1 controls cell polarity, spindle orientation, cell fate
and Notch signaling in distal embryonic lung epithelium. Development 138,
Fish, J.L., Dehay, C., Kennedy, H., Huttner, W.B., 2008. Making bigger brains-the
evolution of neural-progenitor-cell division. J. Cell Sci. 121, 2783–2793.
Goto, H., Yasui, Y., Nigg, E.A., Inagaki, M., 2002. Aurora-B phosphorylates Histone
H3 at serine28 with regard to the mitotic chromosome condensation. Genes
Cells 7, 11–17.
Guo, F., Hildeman, D., Tripathi, P., Velu, C.S., Grimes, H.L., Zheng, Y., 2010.
Coordination of IL-7 receptor and T-cell receptor signaling by cell-division
cycle 42 in T-cell homeostasis. Proc. Natl. Acad. Sci. USA 107, 18505–18510.
Jaffe, A.B., Kaji, N., Durgan, J., Hall, A., 2008. Cdc42 controls spindle orientation to
position the apical surface during epithelial morphogenesis. J. Cell Biol. 183,
Joberty, G., Petersen, C., Gao, L., Macara, I.G., 2000. The cell-polarity protein Par6
links Par3 and atypical protein kinase C to Cdc42. Nat. Cell Biol. 2, 531–539.
Kajstura, J., Rota, M., Hall, S.R., Hosoda, T., D’Amario, D., Sanada, F., Zheng, H.,
Ogorek, B., Rondon-Clavo, C., Ferreira-Martins, J., Matsuda, A., Arranto, C.,
Goichberg, P., Giordano, G., Haley, K.J., Bardelli, S., Rayatzadeh, H., Liu, X.,
Quaini, F., Liao, R., Leri, A., Perrella, M.A., Loscalzo, J., Anversa, P., 2011.
Evidence for human lung stem cells. N. Engl. J. Med. 364, 1795–1806.
Kesavan, G., Sand, F.W., Greiner, T.U., Johansson, J.K., Kobberup, S., Wu, X.,
Brakebusch, C., Semb, H., 2009. Cdc42-mediated tubulogenesis controls cell
specification. Cell 139, 791–801.
Kroschewski, R., Hall, A., Mellman, I., 1999. Cdc42 controls secretory and endocytic
transport to the basolateral plasma membrane of MDCK cells. Nat. Cell Biol. 1,
Lechler, T., Fuchs, E., 2005. Asymmetric cell divisions promote stratification and
differentiation of mammalian skin. Nature 437, 275–280.
Maillet, M., Lynch, J.M., Sanna, B., York, A.J., Zheng, Y., Molkentin, J.D., 2009. Cdc42
is an antihypertrophic molecular switch in the mouse heart. J. Clin. Invest. 119,
McGowan, S.E., 1992. Extracellular matrix and the regulation of lung development
and repair. FASEB J. 6, 2895–2904.
Melendez, J., Grogg, M., Zheng, Y., 2011. Signaling role of Cdc42 in regulating
mammalian physiology. J. Biol. Chem. 286, 2375–2381.
Metzger, R.J., Klein, O.D., Martin, G.R., Krasnow, M.A., 2008. The branching
programme of mouse lung development. Nature 453, 745–750.
Morrisey, E.E., Hogan, B.L., 2010. Preparing for the first breath: genetic and cellular
mechanisms in lung development. Dev. Cell 18, 8–23.
Novick, P., Botstein, D., 1985. Phenotypic analysis of temperature-sensitive yeast
actin mutants. Cell 40, 405–416.
Pepicelli, C.V., Lewis, P.M., McMahon, A.P., 1998. Sonic hedgehog regulates
branching morphogenesis in the mammalian lung. Curr. Biol. 8, 1083–1086.
Perl, A.K., Wert, S.E., Nagy, A., Lobe, C.G., Whitsett, J.A., 2002. Early restriction of
peripheral and proximal cell lineages during formation of the lung. Proc. Natl.
Acad. Sci. USA 99, 10482–10487.
Pruyne, D.W., Schott, D.H., Bretscher, A., 1998. Tropomyosin-containing actin
cables direct the Myo2p-dependent polarized delivery of secretory vesicles
in budding yeast. J. Cell Biol. 143, 1931–1945.
Rathinam, R., Berrier, A., Alahari, S.K., 2011. Role of Rho GTPases and their
regulators in cancer progression. Front. Biosci. 16, 2561–2571.
Rojas, R., Ruiz, W.G., Leung, S.M., Jou, T.S., Apodaca, G., 2001. Cdc42-dependent
modulation of tight junctions and membrane protein traffic in polarized
Madin-Darby canine kidney cells. Mol. Biol. Cell 12, 2257–2274.
Sasaki, N., Kurisu, J., Kengaku, M., 2010. Sonic hedgehog signaling regulates actin
cytoskeleton via Tiam1-Rac1 cascade during spine formation. Mol. Cell
Neurosci. 45, 335–344.
Sauer, B., 1998. Inducible gene targeting in mice using the Cre/lox system.
Methods 14, 381–392.
Sauer, F.C., 1936. The interkinetic migration of embryonic epithelial nuclei.
J. Morphol. 60, 1–11.
Schuger, L., O’Shea, S., Rheinheimer, J., Varani, J., 1990. Laminin in lung develop-
ment: effects of anti-laminin antibody in murine lung morphogenesis. Dev.
Biol. 137, 26–32.
Simmons, C., 1992. The development of cellular polarity in transport epithelia.
Semin. Perinatol. 16, 78–89.
Solecki, D.J., Govek, E.E., Tomoda, T., Hatten, M.E., 2006. Neuronal polarity in CNS
development. Genes Dev. 20, 2639–2647.
Symons, M., Derry, J.M., Karlak, B., Jiang, S., Lemahieu, V., McCormick, F., Francke,
U., Abo, A., 1996. Wiskott–Aldrich syndrome protein, a novel effector for the
GTPase CDC42Hs, is implicated in actin polymerization. Cell 84, 723–734.
Tang, N., Marshall, W.F., McMahon, M., Metzger, R.J., Martin, G.R., 2011. Control of
mitotic spindle angle by the RAS-regulated ERK1/2 pathway determines lung
tube shape. Science 333, 342–345.
Ten Have-Opbroek, A.A., 1991. Lung development in the mouse embryo. Exp. Lung
Res. 17, 111–130.
Tepass, U., Tanentzapf, G., Ward, R., Fehon, R., 2001. Epithelial cell polarity and cell
junctions in Drosophila. Annu. Rev. Genet. 35, 747–784.
Tompkins, D.H., Besnard, V., Lange, A.W., Keiser, A.R., Wert, S.E., Bruno, M.D.,
Whitsett, J.A., 2011. Sox2 activates cell proliferation and differentiation in the
respiratory epithelium. Am. J. Respir. Cell Mol. Biol. 45, 101–110.
Tran, T., McNeill, K.D., Gerthoffer, W.T., Unruh, H., Halayko, A.J., 2006. Endogenous
laminin is required for human airway smooth muscle cell maturation. Respir.
Res. 7, 117.
van Hengel, J., D’Hooge, P., Hooghe, B., Wu, X., Libbrecht, L., De Vos, R.,
Quondamatteo, F., Klempt, M., Brakebusch, C., van Roy, F., 2008. Continuous
cell injury promotes hepatic tumorigenesis in cdc42-deficient mouse liver.
Gastroenterology 134, 781–792.
Wan, H., Luo, F., Wert, S.E., Zhang, L., Xu, Y., Ikegami, M., Maeda, Y., Bell, S.M.,
Whitsett, J.A., 2008. Kruppel-like factor 5 is required for perinatal lung
morphogenesis and function. Development 135, 2563–2572.
Wang, L., Yang, L., Filippi, M.D., Williams, D.A., Zheng, Y., 2006. Genetic deletion of
Cdc42GAP reveals a role of Cdc42 in erythropoiesis and hematopoietic stem/
progenitor cell survival, adhesion, and engraftment. Blood 107, 98–105.
Warburton, D., El-Hashash, A., Carraro, G., Tiozzo, C., Sala, F., Rogers, O., De Langhe,
S., Kemp, P.J., Riccardi, D., Torday, J., Bellusci, S., Shi, W., Lubkin, S.R., Jesudason,
E., 2010. Lung organogenesis. Curr. Top. Dev. Biol. 90, 73–158.
Weaver, M., Batts, L., Hogan, B.L., 2003. Tissue interactions pattern the mesench-
yme of the embryonic mouse lung. Dev. Biol. 258, 169–184.
Welchman, D.P., Mathies, L.D., Ahringer, J., 2007. Similar requirements for CDC-42
and the PAR-3/PAR-6/PKC-3 complex in diverse cell types. Dev. Biol. 305,
Wert, S.E., Glasser, S.W., Korfhagen, T.R., Whitsett, J.A., 1993. Transcriptional
elements from the human SP-C gene direct expression in the primordial
respiratory epithelium of transgenic mice. Dev. Biol. 156, 426–443.
Wu, X., Quondamatteo, F., Lefever, T., Czuchra, A., Meyer, H., Chrostek, A., Paus, R.,
Langbein, L., Brakebusch, C., 2006. Cdc42 controls progenitor cell differentia-
tion and beta-catenin turnover in skin. Genes Dev. 20, 571–585.
Xiao, C., Ogle, S.A., Schumacher, M.A., Schilling, N., Tokhunts, R.A., Orr-Asman,
M.A., Miller, M.L., Robbins, D.J., Hollande, F., Zavros, Y., 2010. Hedgehog
signaling regulates E-cadherin expression for the maintenance of the actin
cytoskeleton and tight junctions. Am. J. Physiol. Gastrointest. Liver Physiol.
Yang, L., Wang, L., Geiger, H., Cancelas, J.A., Mo, J., Zheng, Y., 2007. Rho GTPase
Cdc42 coordinates hematopoietic stem cell quiescence and niche interaction
in the bone marrow. Proc. Natl. Acad. Sci. USA 104, 5091–5096.
Yates, L.L., Schnatwinkel, C., Murdoch, J.N., Bogani, D., Formstone, C.J., Townsend,
S., Greenfield, A., Niswander, L.A., Dean, C.H., 2010. The PCP genes Celsr1 and
Vangl2 are required for normal lung branching morphogenesis. Hum. Mol.
Genet. 19, 2251–2267.
Zen, K., Yasui, K., Gen, Y., Dohi, O., Wakabayashi, N., Mitsufuji, S., Itoh, Y., Zen, Y.,
Nakanuma, Y., Taniwaki, M., Okanoue, T., Yoshikawa, T., 2009. Defective
expression of polarity protein PAR-3 gene (PARD3) in esophageal squamous
cell carcinoma. Oncogene 28, 2910–2918.
Zhang, X., Bi, E., Novick, P., Du, L., Kozminski, K.G., Lipschutz, J.H., Guo, W., 2001.
Cdc42 interacts with the exocyst and regulates polarized secretion. J. Biol.
Chem. 276, 46745–46750.
H. Wan et al. / Developmental Biology 374 (2013) 46–57