XB-ART-44583Curr Biol January 10, 2012; 22 (1): 33-9.
Serotonin signaling is required for Wnt-dependent GRP specification and leftward flow in Xenopus.
In vertebrates, most inner organs are asymmetrically arranged with respect to the main body axis . Symmetry breakage in fish, amphibian, and mammalian embryos depends on cilia-driven leftward flow of extracellular fluid during neurulation [2-5]. Flow induces the asymmetric nodal cascade that governs asymmetric organ morphogenesis and placement [1, 6, 7]. In the frog Xenopus, an alternative laterality-generating mechanism involving asymmetric localization of serotonin at the 32-cell stage has been proposed . However, no functional linkage between this early localization and flow at neurula stage has emerged. Here, we report that serotonin signaling is required for specification of the superficial mesoderm (SM), which gives rise to the ciliated gastrocoel roof plate (GRP) where flow occurs [5, 9]. Flow and asymmetry were lost in embryos in which serotonin signaling was downregulated. Serotonin, which we found uniformly distributed along the main body axes in the early embryo, was required for Wnt signaling, which provides the instructive signal to specify the GRP. Importantly, serotonin was required for Wnt-induced double-axis formation as well. Our data confirm flow as primary mechanism of symmetry breakage and suggest a general role of serotonin as competence factor for Wnt signaling during axis formation in Xenopus.
PubMed ID: 22177902
Article link: Curr Biol
Genes referenced: cat.2 ctnnb1 dvl2 foxj1 foxj1.2 gal.2 nodal nodal1 nodal3.1 nodal3.2 pitx2 rpsa sia1 tekt2 wnt8a
Morpholinos: htr3a MO1 htr3a MO2
Article Images: [+] show captions
|Figure 1. Serotonin Signaling Is Required for the Development of Leftward Flow (A and B) Altered Pitx2c mRNA expression patterns and quantification of results in stage (st) 28-33 xHtr3 morphants and LBD-injected specimens. The following abbreviations are used: wt, left-sided; bi, bilateral. (C) Leftward flow in dorsal explants of stage 17 embryos. Quantification of flow quality (rho) is shown, as calculated from time-lapse movies. (D) GRP analysis. (D) SEM photographs of GRP areas in wild-type embryo (D), MO1 morphant (E), MO1 + xFL mRNA (F), and xLBD-injected specimen (G). Coloration indicates nonciliated cells (red), cells with normal posteriorly polarized cilia (green), and cells with cilia in other locations (blue). Insets show cilia in higher magnification (arrowheads). Scale bar represents 25 mm. (H) Quantification of ciliation rate (H), relative cilia length (I), and cilia polarization (J) of GRP cilia in defined areas, indicated by white squares in (D)G). x in box plots represents mean values. The following abbreviations are used: a, anterior; l, left; p, posterior; r, right; *, significant; **, highly significant; ***, very highly significant; n.s., not significant. Numbers in parentheses indicate number of embryos or dorsal explants analyzed. Statistical significance was calculated using Pearson chi-square test (B) or Student t test (C and H; ). See also Figure S1.|
|Figure 2. Serotonin Signaling Is Required for SM Specification Expression of SM marker genes Foxj1 (A and G) and Xnr3 (D and H) was downregulated in MO1 morphants (B, E, G, and H) and xLBD- or hLBD- injected specimens (F). Coinjection of xFL rescued Foxj1 mRNA levels (C and G). (A) Vegetal view, dorsal side up (A); dorsal view, animal side up (F). Quantification and statistical analysis of results (G and H). Numbers in parentheses represent number of assessed embryos. Statistical significance was calculated using Pearson chi-square test . The following abbreviation is used: ***, very highly significant. See also Figure S2.|
|Figure 3. Interplay of Serotonin and Canonical Wnt Signaling (A and B) Induction of complete or partial secondary (2 ) axes by ventral injection of XWnt8 mRNA at the four-cell stage was inhibited by parallel injection of MO1, MO2, xLBD, or hLBD mRNA. Twinning was rescued upon coinjection of xFL and MO1 or upon coinjection of xLBD or hLBD and serotonin. (C and D) No inhibition of dvl2 (C) or XSia (D) induced 2 axis formation in serotonin signaling- impaired embryos. (E) Reduction of Foxj1 expression in morphants was rescued by coinjection of dvl2 (E) or b-cat (F) mRNA. Gastrula embryos at stage 10-11.5 in vegetal view (dorsal side up) are shown. (G) Quantification of results. Numbers in parentheses represent number of embryos analyzed. Statistical significance was calculated using Pearson chi-square test . The following abbreviations are used: **, highly significant; ***, very highly significant; n.s., not significant. See also Figure S3.|
|Figure 4. Symmetric Serotonin Localization from Early Cleavage Stages through Gastrulation Suggests a Role as Competence Factor for Canonical Wnt Signaling (A) Experimental setup. Embryos injected at the four-cell stage with 20 ng serotonin into single ventral or dorsal blastomeres (B) or uninjected embryos (F) were fixed at the stages indicated, bisected equatorially into animal and vegetal halves (B) or sagittally (I), immu- nostained for serotonin, and mounted and viewed under a confocal laser-scanning microscope as shown. Z stacks of optical sections were sampled at intervals as described in the Supplemental Experimental Proce- dures, beginning at the first section beyond the uneven surface of the cut. Images in (B)F) and (I) are maximum projections of z stacks. (B) Serotonin-injected embryos at the 32-cell stage (B and D) and 256-cell stage (C and E). Note that sero- tonin localization was confined to the injected cell lineage throughout the culture period, i.e., did not spread or accumulate elsewhere (inset in D). (F) Endogenous distribution of serotonin at the 32-cell stage. In the equatorial section cut just below the blasto- coel floor (F), two general localizations of serotonin are seen in each blastomere: a thick cortical layer (G) and a perinuclear cloud (H). The inset in (G) shows a higher magnification to demonstrate the punctate nature of cytoplasmic serotonin staining. (I) Endogenous distribution of serotonin at the onset of gastrulation (stage 10). Although serotonin is found in every cell as at earlier stages, it becomes enriched in the superficial layer of the animal cap (I and J). The contrast in serotonin content between the future floor and roof of the archenteron is evident by the abrupt transition seen at the dorsal lip (K). Serotonin staining in red; green color represents autofluorescence (yolk). The following abbreviations are used: an, animal; bc, blastocoel; d, dorsal; dc, deep cells; v, ventral; veg, vegetal. Asterisk highlights nucleus. (L) Role of serotonin signaling in LR axis specification: a model. Schematic representation of serotonin localiza- tion at blastula stage. Serotonin has accumulated in the superficial animal epithelium of the blastula embryo and mediates competence for canonical Wnt signaling to specify the precursor of the GRP in the dorsal superfi- cial layer, i.e., the SM. Differences in serotonin concen- trations indicated by color code. Wnt signaling is depicted by the nuclear b-cat gradient as black dots of differing intensity . Lack of Wnt signaling on the ventral side precludes SM specification. The following abbreviations are used: an, animal; d, dorsal; st. 9, stage 9; v, ventral; veg, vegetal. For details see text. See also Figure S4.|
|Figure S1, Serotonin Receptors: Expression, Structure and Phenotypes, Related to Figure 1 (A) Expression of xHtr3 (left) and xHtr4 (right) by RT-PCR analysis. Absence of maternal transcripts and onset of zygotic xHtr3 transcription post mid-blastula transition. No detectable xHtr4 mRNA transcription before tadpole stages. (B) 5'-UTR and translational start sequence of xHTR3. Binding sites for MOs are indicated. (C) Serotonin receptor type 3. Schematic representation of domain structure of full- length frog (xFL) and human (hFL) receptor subunits (x5-Ht3, h5-HT3A), consisting of signal peptide (blue), ligand binding domain (LBD; orange) and C-terminal region (green). Six protein loops in the LBD involved in serotonin binding (A-F; red), and four transmembrane domains (TM, dark green) are characteristic features of h5- HT3A . A human splice variant (hsv) lacks essential loops F and C and contains additional amino acids (black) due to read-through into intronic sequences (5-HT3AT in ). The truncated constructs encoding xLBD and hLBD lack the C-terminus. The domain structure of xFL was deduced from amino acid alignments with the hFL and a hydrophobicity plot. (D, E) Serotonin signaling on the dorsal side is required for asymmetric marker gene expression. (D) No significant alteration of Pitx2c expression upon ventral injection of co-MO, MO1, and xLBD or hLBD. (E) Highly significant reduction of wildtype left- sided expression of Xnr1 in morphants following dorsal MO1 injection. (F) Embryonic malformations in embryos injected with higher MO or LBD doses. Embryos were injected at the 4-cell stage into the prospective DMZ of dorsal blastomeres and analyzed for Pitx2c expression at stage 28-33. Knockdown of serotonin signaling resulted in absence of Pitx2c expression, reduced head size, dorsal bending and neural tube closure defects (NTD). (G-J) Expression of cilia and GRP marker gene Tekt2 (G, J) was reduced in MO1 morphants (H, J) and rescued upon xFL co-injection (I, J). Quantification of expression levels (J). Dorsal explants are displayed in ventral view (anterior up). bi, bilateral; abs, absent; inv, right-sided; wt, left-sided. Numbers in brackets represent number of embryos. Statistical significances were calculated using Pearson's chi-square test .|
|Figure S2. Loss of Serotonin Signaling Compromises SM Marker Gene Expression without Alterating GRP Cell Lineage, Related to Figure 2 (A-C) FoxJ1 mRNA expression marks the superficial mesoderm (SM) prior to gastrulation. Before (A) and during gastrulation (B, C) Foxj1 mRNA is detected in the superficial layer (sagittal section in A') of the dorsal marginal zone, i.e. the superficial mesoderm (arrow head, A'). Note that Foxj1 signal intensity is found at comparable levels throughout these stages. (D, E) Down-regulation of Foxj1 transcription upon injection of LBD-constructs. (A-E) Vegetal views, dorsal side up. (F, G) Unaltered cell migration of SM cells in MO1 morphants. SM cells were labelled unilaterally by injection of gal mRNA into the left dorsal marginal zone region of the 4-cell embryo. Analysis prior to (F) and post involution (G) demonstrates labeling of left SM cells (F) and left GRP cells (G), unaltered from wildtype embryos. (F') Para- sagittal section of embryo displayed in (F). (F) external dorsal view. (F') para-sagittal section (plane indicated by dashed line in F). (G) ventral view of dorsal explant exposing the GRP (dotted lines). Solid lines represent dorsal midline. a, anterior; an, animal; d, dorsal; l, left; p, posterior; r, right; v, ventral; ve, vegetal.|
|Figure S3. Serotonin Signaling Is Required for Canonical Wnt Signaling, Related to Figure 3 Induction of twinned axes by XWnt8 (A) but not by XSia (B) is compromised upon MO1 (A', B') or xLBD (A'', B'') mediated loss of serotonin signaling.|
|Figure S4. Uniform Serotonin Distribution Relative to Incipient Dorso-Ventral Axis in Cleavage-Stage Embryos, Related to Figure 4. Whole-mount serotonin staining in equatorial planes of animal and vegetal half embryos. Color images are merged from the adjacent serotonin (red) and yolk (autofluorescence; green) panels (maximum Z-stack projections). (A, C, E) animal hemispheres; (B, D, F) vegetal hemispheres. (A, B) no-primary antibody control, 32-cell embryo; (C, D) primary antibody against serotonin, 32-cell embryo; (E, F) primary antibody against serotonin, 256-cell embryo. (G) Whole-mount staining in equatorial plane of a vegetal half-embryo at the 32-cell stage. (H) Higher-magnification inspection at various sites reveal nearly identical cortical serotonin concentrations around the entire equatorial region. (I, J) Distribution of serotonin among blastomeres at the 256-cell stage. (I1-I4) Representative frames from movie S2, which plays through a Z-stack of a serotonin antibody-stained 256-cell embryo (red). One blastomere (arrowheads) gives the appearance of more intense staining than its neighbors, particularly when the image stack is projected (arrowhead in J). Note that this apparent difference results from the fact that the section planes happen to parallel an incompletely closed cleavage furrow (as indicated by the arrowheads in I1, I2), and that the grazing sections therefore oversample signal associated with a blastomere's surface, giving the erroneous impression of selective staining.|