XB-ART-2158Development 2005 Apr 01;1328:1807-18. doi: 10.1242/dev.01712.
Show Gene links Show Anatomy links
The pro-apoptotic activity of a vertebrate Bar-like homeobox gene plays a key role in patterning the Xenopus neural plate by limiting the number of chordin- and shh-expressing cells.
Targeted disruption of effectors molecules of the apoptotic pathway have demonstrated the occurrence and magnitude of early programmed cell death (EPCD), a form of apoptosis that affects proliferating and newly differentiated cells in vertebrates, and most dramatically cells of the central nervous system (CNS). Little is known about the molecular pathways controlling apoptosis at these early developmental stages, as the roles of EPCD during patterning of the developing nervous system. We describe a new function, in Xenopus neurodevelopment, for a highly conserved homeodomain protein Barhl2. Barhl2 promotes apoptosis in the Xenopus neuroectoderm and mesoderm, acting as a transcriptional repressor, through a mechanism that cannot be attributed to an unspecific cellular stress response. We show that the pro-apoptotic activity of Barhl2 is essential during normal neural plate formation as it limits the number of chordin- and Xshh-expressing cells in the prospective notochord and floorplate, which act as organizing centers. Our findings show that Barhl2 is part of a pathway regulating EPCD. They also provide evidence that apoptosis plays an important role in regulating the size of organizing centers.
PubMed ID: 15772136
Article link: Development
Species referenced: Xenopus
Genes referenced: barhl1 barhl2 bcl2 chrd.1 eng gli1 gli3 gsc hesx1 krt12.4 myc ncam1 nppa pax6 shh sox3 tbx2 ventx2.2
Morpholinos: barhl2 MO1 barhl2 MO2
Article Images: [+] show captions
|Fig. 1. Phylogenetic analysis and early expression pattern of XBarhl2. (A) Phylogenetic tree based on identity in 60 amino acid homeobox region between the Barhl proteins in human, mouse, rat, Xenopus, C. elegans and Drosophila and the Barx proteins of human and chicken (Barlow and Francis-West, 1997; Edelman et al., 2000; Hjalt and Murray, 1999). Mouse Barhl2 (Mo et al., 2004), mouse Barhl1 (Bulfone et al., 2000; Li et al., 2002) and XBarhl2 (Patterson et al., 2000; Poggi et al., 2004) sequences were determined by us. Human Barhl2 (GenBank AL355867) and human Barhl1 (GenBank AL354735) sequences were predicted from their genomic DNA sequences; rat Barhl2 (MBH1) (Saba et al., 2003; Saito et al., 1998). The tree was determined using MegAlign from the Lasergene Navigator, licensed by DNAStar. d, Drosophila melanogaster; m, Mus musculus; r, Rattus norvegicus; x, Xenopus laevis; d-rer, Dario rerio; ce, Caenorhabditis elegans; h, Homo sapiens. (B) Expression of Xbarhl2 as determined by RT-PCR analysis on embryos at different embryonic stages and dissected parts of embryos. (a) Xbarhl2 starts to be expressed at stage 10.5. XEF-1α was analyzed as a control. (b) At stage 10.5 Xbarhl2 is expressed in the dorsal part of the embryo. (C,D) Whole-mount in situ hybridization of embryos, using Xbarhl2 as a probe. (C) Dorsal view of stage 14 embryos showing expression of Xbarhl2 in the dorsal diencephalon. (D) Same embryo as C rotated showing expression along the midline (arrow). A, anterior; P, posterior. (E) RT-PCR analysis. (a) Drawing of a stage 14 embryo showing dissected parts (Nieuwkoop and Faber, 1967). (b) Xbarhl2 is expressed in the posterior neural plate, posterior dorsal mesoderm and in the midline of stage 14 embryo. Chordin and Ncam were analyzed as mesodermal and neuroectodermal markers, respectively.|
|Fig. 2. Changes in Barhl2 levels alter neural plate formation. (A) Sequence comparison between the Eh1 motifs of Gsc, Eng, Nkx, Anf-1, Hesx1 and Msx, and the FILs domains of Barhl proteins. Black shading indicates identical residues, gray shading indicates similarity. Most of the sequences, except for Barhl members, come from Smith and Jaynes (Smith and Jaynes, 1996). (B) Map of the RNA constructs. The two blue boxes indicate the FIL domains; the green box indicates the Myc epitope; the yellow box indicates the Engrailed repression domain. (C) Sequences of the Xbarhl2 antisense morpholinos. The five mispairs in Xbarhl2ASIII are in red and the ATG is in bold. For comparison, the N-terminal sequences of mouse Barhl2, Xbarhl1 and Xbarhl2 are shown. (D) In situ hybridization of embryos injected with mouse Barhl2, Barhl2δFIL, Barhl2δFIL-enR, Xbarhl2ASI, Xbarhl2ASII, Xbarhl2ASIII and analyzed at stage 15. Dorsal view, anterior end upwards. The broken line indicates the midline and `inj' indicates the injected side. Embryos labeled with Xsox3 as probe: (a) control; injected with mouse Barhl2 at (b) 10 pg and (c) 50 pg; (e) injected with mouse Barhl2δFIL (100 pg); (f) injected with mouse Barhl2δFIL-enR (50 pg); (g) injected with Xbarhl2ASI; (h) injected with Xbarhl2ASII; injected with Xbarhl2ASII at (i) 20 ng, (j) 50 ng and (k) 100 ng. Embryos labeled with XK81 as a probe: injected with mouse Barhl2 at 10 pg (d); injected with Xbarhl2ASII (l). (E) Two-cell embryos were injected in one blastomere and analyzed at stage 9. Representative embryos are shown under fluorescent light. (a) Embryos injected with Xbarhl2-myc (100 pg); (b) injected with Xbarhl2-myc (100 pg) together with Xbarhl2ASII (100 ng); (c) injected with Xbarhl2-myc (100 pg) together with Xbarhl2ASIII (100 ng). (F) Western analysis on protein extracts from stage 9 embryos injected with Xbarhl2-myc (100 pg) together with: Xbarhl2ASIII (100 ng) (a); Xbarhl2ASII (50 ng) (b) or Xbarhl2ASII (100 ng) (b′) using anti-Myc antibody. (G) Embryos labeled with Xsox3 as probe. The percentage of embryos exhibiting the staining patterns: (a) injected with mouse Barhl2 (10 pg), 78%; (b) injected with Xbarhl2ASII (50 ng), 88%; (c) injected with mouse Barhl2 (10 pg) together with Xbarhl2ASII (50 ng), 60%.|
|Fig. 4. Barhl2 pro-apoptotic activity is normally involved in regulating endogenous apoptosis. (A) TUNEL staining of stage 12.5 (a), stage 15 (b) and stage 18 (c,d) embryos. (d) Magnified view of the boxed region in c. (e,f) Double in situ hybridization of stage 18 embryos using digoxigenin- and fluorescein-labeled RNA probes: Xbarhl2 (dark blue) and Xshh (light blue) (e); Xbarhl2 (dark blue) and Xpax6 (light blue) (f). (g) In situ hybridization of Xbarhl2AS-injected embryo using Xpax6 as a probe. (a,b) Dorsal views, anterior upwards. (c-f) Anterior views, dorsal upwards. (B) Comparison of the percentage of TUNEL-positive embryos (defined as exhibiting more then 30 apoptotic nuclei in the axial midline, i.e. where endogenous apoptosis normally occurs) between wild-type embryos and embryos injected in the two dorsal blastomeres at the two-cell stage with Xbarhl2AS. Each experimental batch (n=60) was assessed independently. Three different experiments were performed and the results are shown as mean±s.e.m. (C) Kinetics of Xbarhl2 induced apoptosis between stage 8 and stage 13. Embryos were injected in one dorsal blastomere at the two-cell stage with GFP (dark blue), Xbarhl2 (red), Xbarhl2 together with Xbarhl2AS (green) or Xpax6 (light blue) and collected at indicated developmental stages. Cell death was measured by cell death detection ELISA. The apoptotic EF is calculated using GFP-injected embryos as a control. We carried out three different experiments and the results are shown as mean±s.e.m. (D) Western blot analysis on 3 μg of protein extracted from stage 12 embryos injected with Xbarhl2 (100 pg) together with Xbarhl2ASIII or Xbarhl2 (100 pg) together with Xbarhl2ASII, Xpax6 (100 pg), as indicated. The arrow indicates the large (19 kDa) cleaved caspase 3 fragment. The broken arrow shows a nonspecific band used as an internal control.|
|Fig. 5. Manipulation of Barhl2 levels changes the number of chordin- and Xshh-expressing cells. In situ hybridization of embryos injected in either two (A-E) or one (F) dorsal blastomere with mouse Barhl2, Xbarhl2AS or human BCL2. Embryos are shown at (A) stage 10, (B,C) stage 15, (D) stage 18, (E) stage 12, (F) stage 15. (A-D) In situ hybridization using chordin as a probe. (A,B) control (a), mouse Barhl2 (b), Xbarhl2AS (c), human BCL2 (d). (C) Serial sections. (a) Scheme of a stage 14 embryo showing position of cuts. (b-i) Sections of Xbarhl2AS-injected embryo. α level: (b) control; (c) Xbarhl2AS. β level: (d) control; (e) Xbarhl2AS. γ level: (f) control; (g) Xbarhl2AS. Thinβ level sections in the prospective notochord area: (h) control; (i) Xbarhl2AS. (D) β level sections of stage 18 embryos: (a) control; (b) mouse Barhl2; (c) Xbarhl2AS. (E) In situ hybridization using Xvent2 as a probe: (a) control; (b) mouse Barhl2; (c) Xbarhl2AS. (F) In situ hybridization using Xshh as a probe (a-i). (a) Control; (b) mouse Barhl2; (c) Xbarhl2AS; (d) human BCL2. (e-h) Magnified views of a-d, respectively. (g) β level section of Xbarhl2AS-injected embryo. (h,l) In situ hybridization using gli1 as a probe. (i) In situ hybridization using gli3 as a probe. The line in e-h is shown as a scale indicator.|
|Fig. 3. Barhl2 induces apoptosis in Xenopus neuroectoderm. (A) Stage 15 embryos. (a) Embryo injected with Xbarhl2AS and immunostained using anti-phospho-histone H3 antibody. (b,c) In situ hybridization using Xsox3 as a probe on embryos injected with: (b) Xbarhl2AS and treated with HU; (c) human BCL2 (500 pg). (B) TUNEL staining of stage 15 embryos representative of those injected with mouse Barhl2 (20 pg) (a), Xbarhl2 (20 pg) (b) and mouse Barhl2 (10 pg) (c). (a,b) Anterior views, dorsal upwards; (c) dorsal view, anterior upwards. (C) Number of TUNEL+ cells were counted on the injected side and compared to the number on the control side at stage 15. The results were assessed by Student's t-test. Results are shown as mean.e.m. Injection of mouse Barhl2 increased apoptosis 2.6 times (10 pg to 25 pg) (n=70, P=3.10), as does Xbarhl2 (20 pg, n=20, P=9.10). Injection of human BCL2 (500 pg) reduced apoptosis (n=35, P=0.002), and the effect of Barhl2 (10 pg) was reversed by co-injection of human BCL2 (500 pg) (n=40, P>0.01). Injection of mouse Barhl2δFIL had no effect (n=28, P>0.01). (D) In situ hybridization using Xsox3 as a probe of stage 14 embryos injected with: (a) mouse Barhl2 (10 pg), 82%; (b) human BCL2 (500 pg) 89%; and (c) mouse Barhl2 (10 pg) and human BCL2 (500 pg), 60%.|