doi: 10.1242/10.1242/dev.00164
Dishevelled 2 is essential for cardiac outflow tract development, somite segmentation and neural tube closure
Natasha S. Hamblet1,2,3,
Nardos Lijam3,
Pilar Ruiz-Lozano2,
Jianbo Wang1,2,
Yasheng Yang4,
Zhenge Luo5,
Lin Mei5,
Kenneth R. Chien2,
Daniel J. Sussman4 and
Anthony Wynshaw-Boris1,2,3,*
1 Departments of Pediatrics and Medicine, UCSD Comprehensive Cancer Center,
University of California, San Diego, 9500 Gilman Drive, La Jolla, CA
92093-0627, USA
2 Institute of Molecular Medicine, University of California, San Diego, 9500
Gilman Drive, La Jolla, CA 92093-0641, USA
3 Genetic Disease Research Branch, National Human Genome Research Institute,
National Institutes of Health, Bethesda, MD 20814, USA
4 University of Maryland School of Pharmacy, Department of Pharmaceutical
Sciences, 20 N. Pine Street, Baltimore, MD 21201, USA
5 Departments of Neurobiology, Pathology, and Physical Medicine and
Rehabilitation, University of Alabama at Birmingham, 1530 Third Avenue South,
Birmingham, AL 35294-0021, USA
View larger version (31K):
[in a new window]
|
Fig. 1. Targeted inactivation of the Dvl2 gene. (A) Diagrammatic
representation of the wild-type Dvl2 allele (top), the construct used
for generating a null allele (middle) and the inactivated gene after
homologous recombination (bottom). A PGKneo cassette was inserted in the
opposite orientation relative to the start of Dvl2 transcription in
exon 2 and exons 3-6 were removed. (B) Southern blot analysis of genomic DNA
from targeted (lanes 2-6) and wild-type (lane 1) embryonic stem cell clones
digested with BAmHI/EcoRI and using the indicated 5'
flanking probe. The targeted band was approximately 3 kb larger because of the
loss of the EcoRI site in exon 5. The positions of the targeted loci
(KO) and the wild-type loci (+) are shown. (C) Southern blot analysis. Genomic
tail DNA was digested with BamHI and detected with the 5'
probe. The targeted allele was now smaller because a BamHI site is
included in the PGKneo cassette. (D) Immunoblot analysis. Brain lysates from
Dvl2+/+ and Dvl2-/- adults were used
for western blot analysis, using C-terminal antibodies to Dvl1 (polyclonal),
Dvl2 (monoclonal) and Dvl3 (monoclonal). No Dvl2 was detected in the
Dvl2-/- samples. Dvl1 and Dvl3 protein levels were similar
in both genotypes. Asterisk indicates Dvl1-specific band, identified by its
absence in Dvl1 mutant mice (data not shown). There is a higher
molecular weight contaminant band used to assess loading. (E) PCR genotyping
of genomic tail DNA. Upon amplification wild-type and knockout loci generate
391 bp and a 600 bp fragments, respectively. Lanes 1, 3, 5, 6, 8 and 9 are
wild type; lanes 2, 4 and 7 are Dvl2+/-; and lane 10 is a
Dvl2-/- mouse.
|
|
View larger version (113K):
[in a new window]
|
Fig. 2. Cardiovascular abnormalities in Dvl2-/- embryos.
Frontal views of Dvl2+/+ (A,B) and
Dvl2-/- (C-F) hearts at 18.5 dpc. (A,C,E) Light microscope
photograph; and (B,D,F) scanning electron microscopy (SEM) of identical hearts
shown in left panel. (A,B) Normal heart outflow tract septation in which the
pulmonary (PT) trunk arises from the right ventricle (RV) and the aorta (A)
from the left ventricle (LV). The pulmonary trunk was located posteriorly and
to the right rear of the aorta. The heart was rotated so that the right atrium
was in the foreground, making the right atrium appear larger than the left
atrium. The right atrium was visibly connected to the right ventricle. (C,D)
In the most common defect observed in the Dvl2-/- mutants,
both great vessels emerged from the right ventricle (DORV). The aorta was
located to the left of and juxtaposed next to the pulmonary trunk. (E,F) A
second defect observed in the Dvl2-/- hearts was the lack
of two distinct outflow tract vessels. Instead a singular conotruncus emerged
from the ventricle (PTA). No interventricular septum was visible. The
ventricular surface was smooth and the irregularly shaped chamber was
enlarged.
|
|
View larger version (57K):
[in a new window]
|
Fig. 3. Histological sections of Dvl2+/- and
Dvl2-/- 18.5 dpc embryos stained with Hematoxylin and
Eosin. (A) In the wild type, the right and left ventricle is divided by a
membranous septum. (B) DORV and (C) transposition of the great arteries (TGA)
in Dvl2-/- mutant. In B, the arrow indicates the pulmonary
artery. Ao, aorta; PA, pulmonary artery; RV, right ventricle; LV, left
ventricle.
|
|
View larger version (78K):
[in a new window]
|
Fig. 4. In situ analysis using Pitx2 and PlexinA2 as markers of
cardiac neural crest. (A-F) Pitx2 was used as a probe on wild-type
(A,C,E) and Dvl2-/- (B,D,F) embryos at 10.5 dpc. (A,B) The
left side views reveal signal in the branchial arches and migrating cardiac
neural crest of wild-type (CNC in A) but not Dvl2-/-
embryos (asterisk in B). (C,D) The right sided views reveal signal in the
outflow tract (outlined by two asterisks) of wild-type (C), but not
Dvl2-/- embryos (D). (E,F) Embryos in C and D were
dissected to reveal staining in the cardiac outflow tract (OT) of wild-type
(E) but not Dvl2-/- mutant (F) embryos. (G,H) Plexin A2
was used as a probe for migrating neural crest cells in the outflow tract (OT)
of hearts, which were present in wild-type (G) but not
Dvl2-/- hearts (H) at 10.5 dpc.
|
|
View larger version (80K):
[in a new window]
|
Fig. 5. Dvl2-/- mutant mice show variable rib defects. At 18.5
dpc, embryos were stained with Alizarin Red (bone) and Alcian Blue
(cartilage). (A) Thirteen distinct ribs are attached to a complementary
vertebral body, but there is a hemivertebrae at the 12th rib. Sternal ribs are
seen in the background, while vertebral ribs are in the foreground. Vertebral
ribs occasionally show an inappropriate fusion at the proximal rib site
(arrowheads, B) or at slightly more distal regions of the rib (arrowheads in
C). An abnormal ossification bridge links two neighboring vertebral ribs in
some mutants (arrowhead in D). (E) Vertebral rib from
Dvl2-/- E18.5 embryo showing a bowed perforation in the
proximal vertebral rib. (F) Dvl1-/-
Dvl2-/- double mutant showing irregular rib fusion and
splitting of ribs along the vertebral column (arrowheads). Abnormalities in
the cervical and thoracic vertebrae are evident. The sternum has been cut to
better visualize the rib defects. Frontal view of sternum from a wild-type (G)
and Dvl2-/- mutant (H) 18.5 dpc embryo. During development
the ribs move ventromedially and fuse inhibiting bone formation at the contact
sites and forming ossification centers around these sites termed sternbrae.
Defects detected in sternbrae 6 and the xiphoid process (XP) of
Dvl2-/- mutants include bifurcation (arrowhead, H) as well
as perforations.
|
|
View larger version (66K):
[in a new window]
|
Fig. 6. Expression of mesodermal markers in wild-type, Dvl2-/-
and Dvl1-/- Dvl2-/- embryos.
Whole-mount in situ hybridization was carried out on E8.5, E9.5 and E10.5
embryos. Labeling with myogenin, a myotomal marker, revealed abnormal
segmentation into somites (S) of Dvl2-/- embryos (B). The
regular metameric pattern seen in wild-type embryos (A) was maintained in
mutant embryos (B) except for a distinct band of expression connecting two
somites (B, arrowhead and inset). (C-F) Coordinated segmentation was normal in
Dvl2-/- mutants. Notch1 was properly expressed in
the presomitic mesoderm (pm, arrow) of the wild-type (C) and of the
Dvl2-/- mutant embryo (D). Two pairs of newly forming
somites were seen on either side of the neural tube (S and arrow in E,F) in
wild-type (E) and Dvl2-/- (F) embryos using
Notch2, a presomitic marker. Wild-type (+/+, G,I,K) and
Dvl1/Dvl2 double homozygous embryos (Dvl1-/-
Dvl2-/-, H,J,L). Expression of lunatic fringe was detected
in neural tube, tail mesoderm, rhombomeres and sclerotome of wild-type and
Dvl1-/- Dvl2-/- double knockouts but
expression in the sclerotome was decreased in the mutant embryo (inset H).
(I,J) The posterior sclerotomal marker Uncx4.1 reveals normal
expression in the somites of wild-type embryos but this pattern was disrupted
in caudal somites of the Dvl1-/-
Dvl2-/- embryo. In the mutant embryo (J) the somite signal
was not as distinct as that of the normal embryo (I) and two stripes of
Uncx4.1 expression were fused in the mutant (inset J). (K,L) Delta 1
was normally expressed in the caudal somite of wild-type (white arrowhead in
K) and mutant embryos but abnormal expression was detected in some somites of
the Dvl1-/- Dvl2-/- embryo (arrows).
An additional patch of expression was detected in a sagittal section of the
mutant somite (inset L).
|
|
View larger version (73K):
[in a new window]
|
Fig. 7. Neural tube defects in Dvl2-/- and
Dvl1-/-;Dvl2-/- mice. (A) Lateral view of
wild-type 14.5 dpc embryo. (B) Example of spinal bifida and exencephaly seen
at low penetrance (2-3%) in an 14.5 dpc Dvl2-/- embryo.
(C) Craniorachisis was evident in Dvl1/Dvl2 double mutant littermate
The tail was frequently tightly curled. All other major body structures were
normal, including the limbs and face. (D) The neural tube was completely
closed in wild-type embryos at 10.5 dpc. (E) SEM studies at 9.5 dpc revealed a
severe open neural tube that extends from the midbrain-hindbrain junction to
the tail. The posterior region of the Dvl1/Dvl2 double mutant just
rear of the hindlimb bud was truncated as there was the loss of a properly
formed tail and rostral somites. Sagittal sections through 14.5 dpc
Dvl1-/-;Dvl2-/- embryonic head (G) indicated
major brain regions were present but disorganized in the mutant embryos when
compared with wild type (F). (H,I) Transverse sections through the thoracic
cavity. The spinal cord of Dvl1/Dvl2 double mutant embryos (I) was
open but recognizable with a well differentiated floor plate (arrow), compared
with wild type (H).
|
|
© The Company of Biologists Ltd 2002
