QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online January 13, 2009
doi: 10.1242/10.1242/dev.024901

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in Development Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Strate, I. Articles by Pera, E. M. Search for Related Content PubMed Articles by Strate, I. Articles by Pera, E. M. Social Bookmarking                         What's this?

Retinol dehydrogenase 10 is a feedback regulator of retinoic acid signalling during axis formation and patterning of the central nervous system

¹ Stem Cell Center, Lund University, 22184 Lund, Sweden.
² Department of Developmental Biochemistry, Institute of Biochemistry and Cell Biology, Georg August University Göttingen, 37077 Göttingen, Germany.

View larger version (54K): [in this window] [in a new window]   Fig. 1. Xenopus retinol dehydrogenase 10. (A) SDS-PAGE of conditioned medium from HEK 293T cells labelled with 35S-methionine and 35S-cysteine and mock-transfected (control) or transfected with XRDH10 cDNA. The diffuse band at 42 kDa corresponds to the full-length XRDH10 protein. (B) Protein structure of XRDH10. The invariant sequences TGxxxGxG (co-factor binding; x indicates any amino acid residue), NNAG and YxxxK (active site) are characteristic for members of the short-chain dehydrogenase/reductase family (Persson et al., 2003). SP, signal peptide. (C) Sequence alignment of Xenopus, human (H), mouse (M), chick (C) and zebrafish (Z) RDH10 proteins. Two X. laevis RDH10 alleles (XRDH10a and XRDH10b) are shown. The arrowhead indicates the predicted signal peptide cleavage site, and dots underline conserved sequences. (D) Evolutionary relationship of RDH10 sequences.

View larger version (105K): [in this window] [in a new window]   Fig. 2. Expression of Xenopus RDH10. (A,D) RT-PCR analysis of whole embryos (A) and embryonic explants (D). Histone H4 was used as an RNA loading control. (B,C,E-Z) Whole-mount in situ hybridization with an antisense RNA probe for XRDH10 (B,C,E,F,I-K,N,Q-X), XRALDH2 (G,L,O,Y) and XCYP26A1 (H,M,P,Z). Embryos are shown in lateral (B,C,Q,R,Y,Z), vegetal (E,G,H), dorsal (I,J,L,M) and anterior (N-P) views. Specimens are hemi-sectioned (F) and transversally sectioned (K,S-X). The asterisk in R indicates the midbrain-hindbrain boundary. alp, anterior lateral plate; cc, cardiac crescent; dac, dorsal animal cap; dbl, dorsal blastopore lip; ea, ear; ey, eye; hp, head process; mb, midbrain; n, notochord; nc, neural crest; ol, olfactory system; pba, posterior branchial arch; pm, presomitic mesoderm; pn, pronephros; pr, proctodeum; sc, spinal cord; te, telencephalon; vbl, ventral blastopore lip.

View larger version (78K): [in this window] [in a new window]   Fig. 3. Retinoic acid downregulates XRDH10 expression. (A-L) Whole-mount in situ hybridization analysis of XRDH10 transcription at neurula (A,B,E-L) and tailbud (C,D) stage. Embryos are shown in anterior (A,B,E,F,K,L), lateral (C,D, insets) and dorsal (G-L) views. (A-D) Embryos were treated from stage 11 (A,B) or stage 16 (C,D) onwards with 0.05% DMSO as a control or with 5 µM retinoic acid (RA). Note that RA induces a significant reduction in XRDH10 expression. (E,F) Embryos were microinjected into the animal pole at the four-cell stage with 2 ng XRALDH2 mRNA and treated from stage 11 onwards with 0.05% ethanol as a control (E) or 5 µM retinal (F). (G-J) Treatment from stage 11 onwards with the RA inhibitors disulfiram (10 µM) or citral (20 µM) causes an elevation of XRDH10 expression. (K,L) Embryos were animally injected into a single blastomere at the four-cell stage with 300 pg nlacZ mRNA as lineage tracer (red nuclei) alone (K) or together with 2 ng XCYP26A1 mRNA (L). Note that XCYP26A1 induces an upregulation of XRDH10 expression on the injected side (arrowhead). The indicated gene expression patterns were obtained in: A, 55/55; B, 22/31; C, 30/30; D, 37/37; E, 11/11; F, 16/17; G, 29/29; H, 49/49; I, 31/31; J, 54/57; K, 36/36; L, 48/53 embryos. (M) Negative-feedback regulation of RA biosynthesis.

View larger version (50K): [in this window] [in a new window]   Fig. 4. XRDH10 induces retinoic acid signalling and differentially affects organizer gene expression. (A) Uninjected tadpole-stage embryo. (B) Animal injection of 4 ng XRDH10 mRNA at the four-cell stage induces a slight reduction of head structures and a shortening of the tail. (C) Treatment with 0.1 µM RA between stages 9 and 12 induces microcephaly and tail shortening. (D) Injection of 0.5 ng XCYP26A1 mRNA reverts the effect of XRDH10 mRNA and restores normal head and tail development. (E) Treatment with 4 µM citral at stages 9-12 abrogates the activity of XRDH10 mRNA. (F) RT-PCR analysis of animal caps explanted from stage 8 embryos and cultured until stage 12.5. Embryos were injected with 4 ng XRDH10 mRNA (lane 3) and animal caps treated with 5 µM RA (lane 4). Note that XRDH10 stimulates the transcription of all the RA target genes tested. (G-V) Whole-mount in situ hybridization of gastrula embryos in vegetal view. Insets depict lateral views. Embryos were injected in the margin of each blastomere at the four-cell stage with 1 ng XRDH10 mRNA (H,L,P,T) or treated from stage 8 onwards with DMSO as a control (I,M,Q,U) or 5 µM RA (J,N,R,V). Note that XRDH10 mRNA and RA expand the expression of Xlim1 and Chordin, but reduce the expression of Goosecoid and ADMP in the dorsal blastopore lip. Frequency of embryos with the indicated phenotypes was: B, 30/39; C, 25/25; D, 30/40; E, 29/39; G, 45/45; H, 19/39; I, 31/31; J, 41/41; K, 38/38; L, 29/43; M, 30/36; N, 16/29; O, 6/8; P, 4/6; Q, 22/28; R, 15/24; S, 14/14; T, 9/13; U, 64/69; V, 38/50.

View larger version (66K): [in this window] [in a new window]   Fig. 5. Overexpression of XRDH10 and XRALDH2 results in an anteriorward shift of neural markers, whereas XCYP26A1 has the opposite effect. Whole-mount in situ hybridization of embryos after microinjection of mRNA into the animal pole of one dorsal blastomere at the four-cell stage. The lineage tracer nlacZ (red nuclei) labels the injected right-hand side. (A-E) Late gastrula embryos in dorsal view (anterior to the top). HoxD1 demarcates the ectoderm and mesoderm in the trunk with an anterior expression boundary at the level of rhombomere 4 (horizontal line). (F-J) Early neurula embryos in dorsal view, showing Xlim1 expression in two lines of neural cells (arrow). (K-O) Early tailbud embryos in anterior view (posterior to the top) and schematic overviews demarcating Rx2A expression in the eyes and Krox20 expression in rhombomeres 3 and 5 of the hindbrain. (P-T) FoxG1 labels the telencephalon, and En2 the midbrain-hindbrain boundary. (U) Synergistic effects of XRDH10 and XRALDH2 on hindbrain patterning. The anteriorward shift of Krox20 expression is shown in response to mRNA injections at the indicated doses. Note that XRDH10 has little effect on its own, but strongly enhances the posteriorizing effect of XRALDH2. nlacZ mRNA was injected as a control. Injected RNA amounts were (where not otherwise noted): nlacZ (300 pg), XRDH10 (1 ng), XRALDH2 (1 ng) and XCYP26A1 (0.5 ng). ey, eye; rh, rhombomere; R2, XRALDH2; R10, XRDH10. The indicated changes in gene expression were observed in: B, 35/78; C, 43/59; D, 18/29; E, 9/9; G, 24/96; H, 45/95; I, 30/51; J, 13/13; L, 7/36; M, 22/33; N, 22/33; O, 15/15; Q, 6/56 (En2); R, 7/19 (En2); S, 8/20 (En2); T, 25/25 (En2) embryos.

View larger version (57K): [in this window] [in a new window]   Fig. 6. XRDH10 co-operates with retinol during head development. (A-D) Embryos were injected into the animal pole at the four-cell stage with the indicated mRNAs and treated with DMSO or retinol at stages 9-12. (A) DMSO-treated control embryo at tadpole stage. (B) Retinol (50 µM) induces microcephaly at the tadpole stage. (C) Injection of XRDH10 mRNA (1 ng into four blastomeres) and subsequent retinol treatment causes anencephaly. (D) XCYP26A1 mRNA (2.5 ng) partially restores eye and head structures in retinol and XRDH10-treated embryos. (E) Eye deficiencies induced by retinol and XRDH10, and dose-dependent rescue by XCYP26A1 mRNA in stage 40 embryos. (F) Control embryo at the tail bud stage after single injection of nlacZ mRNA. (G) Retinol (25 µM) leads to a slight reduction of the Rx2A-positive eye field (arrowheads). (H,I) In the retinol-treated embryos, XRDH10 mRNA (1 ng in one dorsal blastomere) causes a unilateral collapse of Rx2A expression (arrowhead in H), which is rescued by the co-injection of 2.5 ng XCYP26A1 mRNA (arrowhead in I). The indicated phenotypes were observed in: A, 24/24; B, 25/27; C, 28/45; D, 51/59; G, 13/15; H, 20/35; I, 10/13 embryos.

View larger version (47K): [in this window] [in a new window]   Fig. 7. Knockdown of XRDH10 and XRALDH2 induces ventralization and influences mesodermal gene expression. Antisense morpholino oligonucleotides (MOs) were injected marginally at the two-cell stage (5.2 pmol per blastomere), followed by injection of non-targeted mRNA constructs (XRDH10* and mRALDH2) at the four-cell stage (1 ng per blastomere). (A,B) MOs target the translation initiation sites of two pseudoalleles of Xenopus laevis RDH10 and RALDH2. (C,D) Protein synthesis of XRDH10 and XRALDH2 is specifically inhibited by XRDH10-MO and XRALDH2-MO, but not by control-MO of random sequence. (E-J) Microinjection of XRDH10-MO and XRALDH2-MO leads to microcephaly and enlarged ventroposterior structures in tailbud embryos (E-G), and to reduced eye structures and shortened tails in tadpoles (H-J). Normal development is restored by XRDH10* and mRALDH2 mRNAs, respectively (insets). (K-S) Gastrula embryos in dorsal view. XRDH10- and XRALDH2-morphants have reduced Chordin expression (K-M) and expanded expression domains of Goosecoid (N-P) and ADMP (Q-S). (T-V) Neurula embryos in dorsal view (anterior to the top) after a single injection of MOs with the lineage tracer nlacZ mRNA (red nuclei). XRDH10-MO and XRALDH2-MO reduce Xlim1 expression in the pronephros (arrowhead). The indicated phenotypes were observed in: E, 101/117; F, 62/84 (inset, 73/98); G, 44/67 (inset, 80/84); H, 74/79; I, 39/50 (inset, 56/68); J, 18/48 (inset, 58/64); K, 33/33; L, 21/36 (inset, 19/27); M, 28/36 (inset, 20/28); N, 51/56; O, 28/38 (inset, 46/53); P, 37/49 (inset, 34/44); Q, 52/52; R, 14/26 (inset, 51/62); S, 51/62 (inset, 23/32); T, 11/15; U, 14/20 (inset, 28/32); V, 16/24 (inset, 50/51) embryos.

View larger version (74K): [in this window] [in a new window]   Fig. 8. XRDH10 contributes to CNS patterning and the posteriorizing effect of retinol. Morpholino oligonucleotides (MOs; each 2.6 pmol per embryo) were injected into the margin of one blastomere at the two-cell stage. The non-targeted mRNA constructs XRDH10* and mRALDH2 (each 1 ng) and the lineage tracer nlacZ mRNA were co-injected. Embryos are shown in dorsal view (anterior to the top). (A-F) Late gastrula embryos. XRDH10-MO, XRALDH2-MO, or a combination of both morpholinos, cause a reduction and posteriorward retraction of HoxD1expression, which is reverted by XRDH10* and mRALDH2 mRNA. (G-L) Neurula embryos showing expression of En2 (midbrain-hindbrain boundary), HoxB3 (hindbrain rhombomeres 5 and 6) and HoxC6 (anterior spinal cord). (M-R) Tailbud embryos depicting expression of Rx2A (eyes) and Krox20 (rhombomeres 3 and 5). (S) Effects of XRDH10 and XRALDH2 knockdown on hindbrain patterning. The posteriorward shift of Krox20 expression is shown in response to MO injections at the indicated doses. (T-W) Treatment with 100 µM retinol at stages 9-12 induces a robust anterior expansion of HoxD1 expression in late gastrula embryos (V). XRDH10-MO reverts the effect of retinol on the injected right-hand side (W). Frequency of embryos with the indicated phenotype was: A, 77/88; B, 55/105; C, 30/68; D, 54/88; E, 19/21; F, 23/25; G, 34/35; H, 17/31 (En2); H, 30/31 (HoxB3); H, 27/31 (HoxC6); I, 15/33 (En2); I, 31/33 (HoxB3); I, 32/33 (HoxC6); J, 6/9 (En2); J, 8/9 (HoxB3); J, 5/9 (HoxC6); K, 20/23; L, 39/39; M, 10/10; N, 35/73; O, 37/69; P, 38/60; Q, 10/10; R, 13/14; T, 9/9; U, 7/13; V, 9/9; W, 20/33 embryos.

View larger version (41K): [in this window] [in a new window]   Fig. 9. Model for the establishment of retinoic acid morphogen gradients in the early embryo. The nested gene expression and combinatorial action of RDH10 and RALDH2 causes a posteriorward flow of retinal that generates an initial RA gradient in the trunk mesoderm with a peak at the level of the hindbrain-spinal cord boundary. Subsequent diffusion of RA creates two gradients across the hindbrain and spinal cord, which acquire their final shape through CYP26A1-mediated RA degradation at the anterior (ant.) and posterior (post.) ends of the neural plate. FB, forebrain; MB, midbrain; HB, hindbrain; SC, spinal cord.

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter