|
Figure 1: Sequence alignment of the Xenopus CaD and other CaD isoforms. Amino acid sequences of CaD isoforms
from Xenopus CaD, chicken nonmuscle CaD (M59762), mouse CaD 1 (NM_145575), and human CaD (M83216) were
aligned using the Clustal W program in DNAstar. Major binding domains for myosin, tropomyosin, actin, and Ca2+-
calmodulin are highlighted in the green, yellow, red, and blue boxes, respectively. Xenopus CaD is highly conserved relative
to those from other species in the binding domains, with 82, 91, 93, and 100% identity at the amino acid level, respectively.
|
|
Figure 2: CaD is expressed in neural crest tissue during early
Xenopus development. Expression of CaD begins at late gastrula
stage (stage 13) at the borders of the forming neural plate, the site of
initial neural crest induction. Its expression increases in the CNC
region by neurula stages (stage 15) and continues in the migrating
neural crest cells as they populate the branchial arches (stage 20
onward). By tailbud stages, the expression decreases and only is
present in part of the ganglia and somite at low levels (stage 26). All
embryos are oriented with anterior to the left.
|
|
Figure 3: CaD is not required for neural crest induction. Embryos
were injected with control-MO or CaD-MO (10 ng) in one cell at the
two-cell stage, together with a lineage tracer (nβGal, stained with
Red-Gal, bottom). No change in expression of neural crest genes twist
and sox10 was observed by in situ hybridization at neurula stages,
suggesting that they were not affected by CaD-MO. All embryos
were in dorsal view, with anterior to the left.
|
|
Figure 4: CaD-MO disrupted the expression of neural crest genes
at tailbud stages. Embryos receiving control-MO or CaD-MO (10 ng)
in one cell at the two-cell stage were collected at tailbud stages and
analyzed for neural crest gene expression by in situ hybridization. The
injected side for each embryo is marked by an asterisk and the
control side of the same embryo is placed below serving as internal
control. Whereas control-MO did not affect the expression of general
neural crest markers twist and sox10 or the third branchial arch
marker EphA4, the expression of all three genes was disrupted by
CaD-MO (arrows). Marker analysis suggests that neural crest cells
failed to extend into the lateral portion of the branchial arches and to
separate into clear streams in the cases of twist and sox10.
Simultaneous injection of CaD (100–200 pg) rescued the expression
of neural crest genes. All embryos were oriented with anterior to the
left and dorsal up.
|
|
Figure 5: CaD-MO inhibited the formation of cranial cartilage.
Ventral view of cartilage from stage 45+ tadpoles stained with Alcian
blue. Embryo halves received 10 ng of control-MO or CaD-MO on the
left (marked by an asterisk). Cartilage on the CaD-MO–injected side
was often reduced in size, distorted, or missing completely, whereas
coexpression of CaD rescued the cartilage formation effectively. CB,
ceratobranchial cartilage (from third and fourth branchial arch
streams); CH, ceratohyal cartilage (from hyoid stream); M, Meckel’s
cartilage (from mandibular stream).
|
|
Figure 6: CaD is required for CNC explants to spread and
segregate on FN matrix. (A) CNC explants were dissected from early
neurula embryos injected with control-MO or CaD-MO (10 ng) and
plated on FN. By 10 h, control-MO–expressing cells spread and
migrated extensively and segregated into three streams. However,
cells from CaD-MO–expressing explants failed to spread efficiently or
separate into streams. Coinjection of 100–200 pg of CaD largely
rescued the spread and segregation of the explants. Arrows with
numbers marked the segregated lobes. High-magnification
differential interference contrast imaging of the explants (B, 40×) and
fluorescence imaging of cells with membrane labeling (C, 20×) show
that whereas control-MO–expressing cells extended multiple
membrane protrusions, CaD-MO–expressing cells remained
unpolarized and were often rounded. (D) The cells were tracked for
2 h and trajectories plotted. Although the speed of migration was
largely unaffected, the directionality of migration was significantly
impaired by CaD-MO. Black traces indicate that the ratio of the final
distance to the entire route covered is >0.3, and red indicates that it
was <0.3.
|
|
Figure 7: CaD-MO disrupted CNC migration in vivo. (A) GFP-labeled CNC grafts receiving
control-MO or CaD-MO (10 ng) were transplanted into unlabeled host embryos at neurula
stages and imaged at tailbud stages. Fluorescent and merged images are shown side by side
with anterior to the left. Whereas control-MO–receiving grafts migrated laterally into separate
branchial arches, CaD-MO–expressing grafts migrated less far and failed to segregate into
streams. Coexpression of a low-dose CaD restored the migration of CNC transplants. (B) The
embryos were binned into four categories on the basis of the migratory behaviors. Whereas
79% of control grafts migrated and segregated normally (n = 61), 18% of CaD-MO–expressing
grafts failed to migrate at all. Another 67% of the grafts migrated shorter distances, over half of
which did not segregate (n = 45). Coexpression of CaD (100–200 pg) rescued the segregation
(83%) and migration (44%, n = 18) remarkably.
|
|
Figure 8: CaD-MO disrupts the organization of actin filament. (A) CNC explants receiving
membrane-tethered EGFP together with control-MO or CaD-MO at 16- to 32-cell stages were
fixed on FN and stained with rhodamine-conjugated phalloidin. Whereas actin associated with
protrusions and stress fibers in control-MO–expressing cells, actin distributed in a relatively
ubiquitous manner in CaD-MO–expressing cells. Arrows indicate injected cells, and arrowheads
mark uninjected cells in the same explant. Right, actin staining alone in dissociated cells.
(B) Phalloidin staining in cells expressing 0.1 ng of EGFP-CaD revealed colocalization of CaD
with actin filaments in membrane protrusions and stress fibers (arrows). Images were taken at
40×. Scale bars, 20 μm.
|
|
Figure 9: Overexpression of various CaD constructs affects
actin-associated structures. Phalloidin staining was performed on
CNC explants receiving 0.2 ng of EGFP-CaD, CaD39-AB, CaD39-6F,
and CaD39-PAKA. To distinguish cells receiving the RNA, H2B-EGFP
was coinjected with the mutant constructs. Whereas overexpression
of EGFP-CaD produced abundant ectopic actin-rich protrusions,
CaD39-AB increased stress fiber formation. CaD39-6F and CaD39-
PAKA caused enrichment of short actin bundles around the cell
periphery to different extents. Arrows indicate actin-rich structures.
Images were taken using 40× objectives. Scale bars, 20 μm.
|
|
Supplementary Figure 1. CaD-MO efficiently inhibits CaD protein translation. Protein
extracted from uninjected and CaD-MO injected CNC tissue was resolved on SDS page
gel and blot against CaD antibody (Santa Cruz Biotechnology, sc-15374). No CaD
protein was detected from CaD-MO expressing tissue indicating CaD-MO efficiently
blocked CaD translation. Blotting against α-tubulin antibody served as loading control.
|
|
Supplementary Figure 2. Expression of CaD in preplacodal region. Anterior views of
stage 13-14 embryos show that CaD is expressed in anterior and lateral borders of the
rostral neural plate, where prospective placodes and cranial neural crest will arise. Dorsal
is oriented to the top.
|
|
Supplementary Figure 3. CaD-MO disrupts CNC cell adhesion. (A) Cell-cell adhesion
examined by dissociation/reaggregation experiments. While control-MO expressing CNC
cells reaggregated into one big cell clump at 6 hours, CaD-MO expressing cells only
grouped to medium sized aggregates. (B) Cell adhesion on FN matrix. Dissociated cells
were plated on FN-coated dish for an hour and unattached cells were washed away by
flipping over the dish. Much fewer cells remained on the FN plate in CaD-MO treated
cells.
|
|
cald1 (caldesmon 1) gene expression in Xenopus laevis embryo, assayed via in situ hybridization, NF stage 15, dorsal view, anterior left.
|
|
cald1 (caldesmon 1) gene expression in Xenopus laevis embryo, assayed via in situ hybridization, NF stage 21, dorsal view, anterior left.
|
|
cald1 (caldesmon 1) gene expression in Xenopus laevis embryo, assayed via in situ hybridization, NF stage 26, lateral view, dorsal up, anterior left.
|
|
FIGURE 1:. Sequence alignment of the Xenopus CaD and other CaD isoforms. Amino acid sequences of CaD isoforms from Xenopus CaD, chicken nonmuscle CaD (M59762), mouse CaD 1 (NM_145575), and human CaD (M83216) were aligned using the Clustal W program in DNAstar. Major binding domains for myosin, tropomyosin, actin, and Ca2+-calmodulin are highlighted in the green, yellow, red, and blue boxes, respectively. Xenopus CaD is highly conserved relative to those from other species in the binding domains, with 82, 91, 93, and 100% identity at the amino acid level, respectively.
|
|
FIGURE 2:. CaD is expressed in neural crest tissue during early Xenopus development. Expression of CaD begins at late gastrula stage (stage 13) at the borders of the forming neural plate, the site of initial neural crest induction. Its expression increases in the CNC region by neurula stages (stage 15) and continues in the migrating neural crest cells as they populate the branchial arches (stage 20 onward). By tailbud stages, the expression decreases and only is present in part of the ganglia and somite at low levels (stage 26). All embryos are oriented with anterior to the left.
|
|
FIGURE 3:. CaD is not required for neural crest induction. Embryos were injected with control-MO or CaD-MO (10 ng) in one cell at the two-cell stage, together with a lineage tracer (nβGal, stained with Red-Gal, bottom). No change in expression of neural crest genes twist and sox10 was observed by in situ hybridization at neurula stages, suggesting that they were not affected by CaD-MO. All embryos were in dorsal view, with anterior to the left.
|
|
FIGURE 4:. CaD-MO disrupted the expression of neural crest genes at tailbud stages. Embryos receiving control-MO or CaD-MO (10 ng) in one cell at the two-cell stage were collected at tailbud stages and analyzed for neural crest gene expression by in situ hybridization. The injected side for each embryo is marked by an asterisk and the control side of the same embryo is placed below serving as internal control. Whereas control-MO did not affect the expression of general neural crest markers twist and sox10 or the third branchial arch marker EphA4, the expression of all three genes was disrupted by CaD-MO (arrows). Marker analysis suggests that neural crest cells failed to extend into the lateral portion of the branchial arches and to separate into clear streams in the cases of twist and sox10. Simultaneous injection of CaD (100–200 pg) rescued the expression of neural crest genes. All embryos were oriented with anterior to the left and dorsal up.
|
|
FIGURE 5:. CaD-MO inhibited the formation of cranial cartilage. Ventral view of cartilage from stage 45+ tadpoles stained with Alcian blue. Embryo halves received 10 ng of control-MO or CaD-MO on the left (marked by an asterisk). Cartilage on the CaD-MO–injected side was often reduced in size, distorted, or missing completely, whereas coexpression of CaD rescued the cartilage formation effectively. CB, ceratobranchial cartilage (from third and fourth branchial arch streams); CH, ceratohyal cartilage (from hyoid stream); M, Meckel's cartilage (from mandibular stream).
|
|
FIGURE 6:. CaD is required for CNC explants to spread and segregate on FN matrix. (A) CNC explants were dissected from early neurula embryos injected with control-MO or CaD-MO (10 ng) and plated on FN. By 10 h, control-MO–expressing cells spread and migrated extensively and segregated into three streams. However, cells from CaD-MO–expressing explants failed to spread efficiently or separate into streams. Coinjection of 100–200 pg of CaD largely rescued the spread and segregation of the explants. Arrows with numbers marked the segregated lobes. High-magnification differential interference contrast imaging of the explants (B, 40×) and fluorescence imaging of cells with membrane labeling (C, 20×) show that whereas control-MO–expressing cells extended multiple membrane protrusions, CaD-MO–expressing cells remained unpolarized and were often rounded. (D) The cells were tracked for 2 h and trajectories plotted. Although the speed of migration was largely unaffected, the directionality of migration was significantly impaired by CaD-MO. Black traces indicate that the ratio of the final distance to the entire route covered is >0.3, and red indicates that it was <0.3.
|
|
FIGURE 7:. CaD-MO disrupted CNC migration in vivo. (A) GFP-labeled CNC grafts receiving control-MO or CaD-MO (10 ng) were transplanted into unlabeled host embryos at neurula stages and imaged at tailbud stages. Fluorescent and merged images are shown side by side with anterior to the left. Whereas control-MO–receiving grafts migrated laterally into separate branchial arches, CaD-MO–expressing grafts migrated less far and failed to segregate into streams. Coexpression of a low-dose CaD restored the migration of CNC transplants. (B) The embryos were binned into four categories on the basis of the migratory behaviors. Whereas 79% of control grafts migrated and segregated normally (n = 61), 18% of CaD-MO–expressing grafts failed to migrate at all. Another 67% of the grafts migrated shorter distances, over half of which did not segregate (n = 45). Coexpression of CaD (100–200 pg) rescued the segregation (83%) and migration (44%, n = 18) remarkably.
|
|
FIGURE 8:. CaD-MO disrupts the organization of actin filament. (A) CNC explants receiving membrane-tethered EGFP together with control-MO or CaD-MO at 16- to 32-cell stages were fixed on FN and stained with rhodamine-conjugated phalloidin. Whereas actin associated with protrusions and stress fibers in control-MO–expressing cells, actin distributed in a relatively ubiquitous manner in CaD-MO–expressing cells. Arrows indicate injected cells, and arrowheads mark uninjected cells in the same explant. Right, actin staining alone in dissociated cells. (B) Phalloidin staining in cells expressing 0.1 ng of EGFP-CaD revealed colocalization of CaD with actin filaments in membrane protrusions and stress fibers (arrows). Images were taken at 40×. Scale bars, 20 μm.
|
|
FIGURE 9:. Overexpression of various CaD constructs affects actin-associated structures. Phalloidin staining was performed on CNC explants receiving 0.2 ng of EGFP-CaD, CaD39-AB, CaD39-6F, and CaD39-PAKA. To distinguish cells receiving the RNA, H2B-EGFP was coinjected with the mutant constructs. Whereas overexpression of EGFP-CaD produced abundant ectopic actin-rich protrusions, CaD39-AB increased stress fiber formation. CaD39-6F and CaD39-PAKA caused enrichment of short actin bundles around the cell periphery to different extents. Arrows indicate actin-rich structures. Images were taken using 40× objectives. Scale bars, 20 μm.
|
