|
Figure 1. Morpholinos Targeting Two Distinct chordin mRNAs Reduce Chordin Expression and Promote Ventralization (A) Sequence of the morpholinos targeting the two Chordin mRNAs. (B) Western blot analysis of secreted Chordin protein. Two-cell embryos were injected four times with a total of 8 ng of Chd-MO-1 (lane 2) or Chd-MO-2 (lane 3) or with a combination of the two (4 ng each; lane 4). Dorsal lips were isolated at gastrula, and explants were cultured in Ca2 -Mg2 -free medium (CMFM) for 12 hr. The supernatant was then analyzed in western blots with an immunopurified anti-Chordin antibody; Coomassie blue staining of the cell pellets served as loading control. (C) Embryos microinjected four times at the two-cell stage with a total of 8 ng of each Chordin morpholino or a mixture of the two. The ventralized phenotype observed by coinjection of the two morpholinos generated a highly penetrant ventralized phenotype ( 75%) in many independent experiments. (F) The effect of Chd-MO(1 2) was rescued by microinjection of 10 pg chd mRNA lacking the 5 UTR.
|
|
Figure 2. Inhibition of Endogenous Chordin Results in Reduction of Neuroectodermal and Expansion of Ventral Mesodermal Markers A mixture of the two Chordin morpholinos (4 ng of each) was injected four times, radially, at the two-cell stage. (A and B) Expression of the organizer marker genes chordin and noggin was unchanged in wild-type (wt, left) and Chd-MO(1 2)-injected (right) embryos at early gastrula. (C) The neural plate was reduced in Chd-MO(1 2)-injected embryos, as shown by the neural marker Sox2 (stage 12). (D, E, and F) At stage 12 the expression of chordin, noggin, and follistatin was not significantly affected by Chd-MO(1 2) injection. The expression domains are shorter, indicating a mild inhibition of convergence-extension gastrulation movements. (G) RT-PCR analysis of sibling embryos. The mRNA levels of chordin, noggin, and follistatin were not significantly changed in Chd-MO(1 2)- injected embryos at stage 10.5 or stage 12. (H) At tadpole stages the A-P pattern of neural organization was relatively normal, as shown by expression of Otx2 (forebrain), Krox-20 (hindbrain), and HoxB9 (spinal cord). (I) Ventral mesoderm, marked by the expression of sizzled, is expanded in Chd-MO(1 2)-injected embryos. The upper panel shows a lateral view, and the lower panel shows a ventral view, of the same embryos. (J and K) At late tadpole stages (stage 42) Chd-MO(1 2)-injected embryos develop with smaller head structures. Histological sections (J , J′, K , and K′) revealed a reduction of midbrain (mb), pharynx (ph), eye (ey), and spinal cord (sc) and an expansion of ventral blood islands (bl) in Ch-MO(1 2)-injected embryos. No obvious changes in somites (so) or notochord (no) were observed.
|
|
Figure 3. Reduction of Chordin and Overexpression of xTsg Synergize, Causing Ventralization of Xenopus Embryos (A) Uninjected wild-type embryo at stage 40, stained with the notochord-specific antibody MZ15 (A ) or the muscle-specific antibody 12-101 (A′). (B) Embryos microinjected with Chd-MO(1 2) at the two-cell stage have normal notochord staining (B ), but the muscles have adopted a U-shaped phenotype (B′). (C) Embryos injected once ventrally with 1 ng of xTsg mRNA at the four- to eight-cell stage. Note that the notochord is reduced and absent from the posterior in strongly affected embryos (C′). Muscle development is disturbed (C′), and muscles adopt a U-shaped phenotype. (D) Embryos coinjected with xTsg and Chd-MO(1 2) are strongly ventralized. (E) N-tubulin expression in stage 25 wild-type embryos, embryos injected at the two-cell stage with 8 ng Chd-MO(1 2), 1 ng xTsg mRNA, or both. (F) RT-PCR analysis of sibling embryos; Chd-MO(1 2) and xTsg mRNA cause the repression of anterior neural markers Rx2a (eye), Six-3 (forebrain), and engrailed (midbrain). The pan-neural marker NCAM and the dorsal mesodermal marker actin are reduced. The ventral mesodermal marker sizzled is upregulated by Chd-MO(1 2) injections, but repressed by xTsg mRNA injections (Chang et al., 2001).
|
|
Figure 4. Chordin Is Required for the Dorsalization of Xenopus Embryos by LiCl Treatment(A) Stage 45 untreated tadpole.(B) Sibling embryo treated with 120 mM LiCl at the 32-cell stage for 30 min.(C) Stage 45 embryo microinjected into the animal pole with 200 pg chordin mRNA at the four-cell stage.(D) Chordin production is inhibited by Chd-MO(1+2), even after LiCl treatment (compare lanes 2 and 3). Whole embryos were dechorionated and dissociated at stage 9 and cultured for 14 hr at 19°C in CMFM to allow detection of Chordin protein in the supernatant.(E and F) LiCl-treated embryos display a radial cement gland at tadpole stages and do not develop trunk-tail structures (dorso anterior index [DAI] = 8, n = 73)(G) Embryos at the two-cell stage with 8 ng Chd-MO(1+2) before LiCl treatment are resistant to LiCl (G; DAI = 6.4, n = 38).
|
|
Figure 5. Chordin Is Required for the Induction of Dorsal Mesoderm by Activin(A–F) Two-cell embryos were injected four times into the animal pole with a total of 8 ng of Chd-MO-1, Chd-MO-2, Chd-MO(1+2), or with Chd-MO(1+2) and 10 pg of chordin mRNA lacking the 5′ UTR of the chordin cDNA. Animal explants were isolated at stage 8, treated with 2 ng/ml Activin, and cultured until stage 25. Three independent experiments were performed with similar results.(G) RT-PCR analysis of samples shown in (A–F). Activin treatment induced the expression of the somite marker genes Myf5 and Actin and the pan-neural marker NCAM. Chd-MO(1+2) inhibited the expression of dorsal markers but upregulated the expression of ventral marker genes (xHox3 and α-Globin).(H) A subset of the animal caps were cultured overnight in CMFM. The supernatant was analyzed in Western blots with an immunopurified antibody specific for the interrepeat region of Chordin (α-I-Chd; Larraı́n et al., 2001), and the cell pellet was analyzed with an antibody specific for phospho-Smad-1.(I) RT-PCR analysis of Activin-treated animal caps at stage 10.5 injected with Chd-MO(1+2). The morpholinos did not affect mesoderm induction (Xbra) or the expression of other BMP antagonists (cerberus, chordin, noggin, and follistatin).(J) RT-PCR analysis of Activin-treated animal caps at stage 12. The microinjection of Chd-MO(1+2) changed noggin and follistatin mRNA levels only slightly, inhibited chordin expression, and upregulated the ventral mesoderm marker xHox3.
|
|
Figure 6. Chordin Is Required for Spemann Organizer Activity (A) Experimental design of the isotopic and isochronic transplantation experiment. Embryos were injected twice into dorsal blastomeres with BDA (biotin dextran amine lineage tracer) alone or with Chd-MO(1 2). Dorsal lip explants were isolated from these donor embryos and transplanted at gastrula into unlabeled host embryos. After the embryos were sectioned in paraffin wax, BDA was detected with streptavidin conjugated alkaline phosphatase and BM purple. (B) Transplantation of an organizer injected with lineage tracer alone allowed normal development in most cases, but some embryos developed shortened axes (12%, n 33). Histological sections showed that BDA staining was detected in the notochord (no) and floor plate (fp) throughout the embryo and in anterior dorsal endoderm (de). Seven embryos were serially sectioned, stained with Streptavidin-AP, and developed with BMP purple. (C) Embryos that received Chd-MO(1 2)-injected organizer transplants developed with shortened body axes (100%, n 15) and small eyes (47%, n 15) or no eyes (20%, n 15). Lineage label was detected in the notochord, anterior dorsal endoderm, and medial somite (so). Surprisingly, transplanted cells stained the epidermis of the host embryo (epi) in the posterior and were excluded from neural tissue. Six embryos were serially sectioned. (D) Dorsal lip explants transplanted into the ventral side of a host embryo recapitulate the classical Spemann-Mangold experiment. (E) The transplantation of a wild-type organizer induced the formation of secondary axes (100%, n 17), of which most (82%) were complete axes containing heads and eyes. In the 11 embryos serially sectioned, BDA could be detected in the notochord, floor plate, and dorsal endoderm of the secondary axes. (F) Transplantation of a Chd-MO(1 2)-injected organizer blocked the axis-inducing activity of the explant. A few embryos formed protrusions (31%, n 16), the strongest of which is shown here. In the seven embryos serially sectioned, BDA staining was detected in the ventral mesoderm (vm), ventral endoderm (ve), and epidermis (epi). Note that Chd-MO(1 2)-injected grafts did not induce the CNS, somites, or notochord.
|
|
Figure 7. The Mesodermal Specification of Dorsal Lip Explants Is Not Changed by Chd-MO(1 2) Microinjection (A) Explanted dorsal lips (n 25) developed into anterior neural tissue (eye, ne), notochord (no), endoderm (en), and cement gland (cg) at stage 38. (B) Chd-MO(1 2)-injected dorsal lip explants (n 27) developed into notochord (no), muscle (mu), posterior neural tissue (ne), otic vesicle (ov), and epidermis (epi). (C and D) Uninjected and Chd-MO(1 2)-injected dorsal lip explants were positive for the notochord-specific antibody MZ15 (n 5). (E and F) Uninjected and Chd-MO(1 2)-injected dorsal lip explants were stained by the muscle-specific antibody 12-101. (G) RT-PCR analysis of dorsal lip explants at stage 10.5 showing that organizer gene expression (chd, noggin, follistatin, and cerberus) was not affected by Chd-MO(1 2) at the time of the transplantations. (H) RT-PCR analysis of Chd-MO(1 2)-injected dorsal lip explants at stage 26. In Chd-MO(1 2)-microinjected embryos the anterior neural markers Six-3 and Rx2a were inhibited, the pan-neural marker NCAM and the cement gland marker XAG were reduced, and the notochord marker Xnot and the somite marker gene actin were not affected.
|