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

First published online August 25, 2005
doi: 10.1242/10.1242/dev.01938

This Article Summary Full Text Full Text (PDF) Supplementary Material An erratum has been published Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Dawes-Hoang, R. E. Articles by Wieschaus, E. F. Search for Related Content PubMed PubMed Citation Articles by Dawes-Hoang, R. E. Articles by Wieschaus, E. F.

folded gastrulation, cell shape change and the control of myosin localization

¹ Department of Molecular Biology, Howard Hughes Medical Institute, Princeton University, NJ 08544, USA
² Harvard Medical School, 25 Shattuck Street, Boston, MA 02115, USA
³ Department of Biological Sciences, Stanford University, CA 94305
⁴ Wellcome Trust/Cancer Research UK Gurdon Institute and Department of Anatomy, University of Cambridge, Cambridge CB2 1QN, UK

View larger version (61K): [in a new window]   Fig. 1. folded gastrulation and myosin localization. (A-C) Antibody staining for fog protein (green) and cell outlines (Nrt, red) in cross-section of ventral furrow (A), sagittal section of posterior midgut (B) and apical surface of ventral cells (C). There is punctate fog staining and localization towards the apical half of the cells. (D-F) Antibody staining for fog protein in the region of the posterior midgut in control shibirets embryos kept at the permissive temperature (D), in shibirets embryos shifted to the non-permissive temperature at gastrulation (E), and in embryos from Rho-kinase germline clones (F). Punctate staining and apical concentration of fog protein is reduced when endocytosis is blocked using the non-permissive shibirets but is maintained in Rho-kinase embryos despite the failure of the latter to form a posterior midgut invagination. (G,I) Frames from time-lapse movies of sqhGFP-expressing embryos 13 minutes after cellularization. In the control (G), myosin is seen along the entire apical surface of the posterior midgut (between arrowheads) and at the junctions of cells beyond this region. In fog mutants (I), apical myosin is severely disrupted occurring in only a few isolated cells underlying the pole cells (arrows). (H,J) Cross-sections showing anti-myosin II antibody staining along the apical surface of the ventral furrow (between arrowheads) in control OreR embryos (H) and in fog mutants (J), in which it is present in some cells (+) but not others (–).

View larger version (82K): [in a new window]   Fig. 2. Myosin localization in embryos misexpressing fog. (A-F) Staining for Fog protein in control embryos expressing just the mat67;mat15 driver line (A-C) and this line driving UASfog6 expression (D-F). At the onset of cellularization (A,D), little fog expression is seen, but early in cellularization, before nuclear elongation (B,E), there is high fog expression throughout UASfog embryos. At the onset of gastrulation, there is low fog expression in the ventral furrow of control embryos (asterisk in C) and high expression throughout the UASfog embryos (F). (G-L) Control (G-I) and UASfog-expressing (J-L) embryos stained for myosin II. Myosin localizes normally to the basal cellularization front and decreases normally in ventral cells, which show the usual increase in cell depth (G,J). At the onset of gastrulation, myosin continues to be localized basally and now also localizes to the apical side of ventral cells in both control (H) and UASfog (K) embryos. Apical myosin is more intense in UASfog embryos and is no longer restricted to the ventral furrow, occurring in dorsal and lateral cells (arrow) too. This continues through later stages of gastrulation (comparing arrow in I and L). (M-O) Quantification of myosin intensity in driver-line control (blue) and UASfog-expressing (pink) embryos. Measurements were taken from the apical and basal sides of ventral and lateral cells at three different stages: (M) end of cellularization (see G,J), (N) onset of gastrulation (see H,K) and (O) later gastrulation (see I,L). No difference is seen between control and UASfog embryos at the end of cellularization (M) but during gastrulation (N,O) myosin increases on the apical side of both ventral and lateral cells in UASfog embryos with no corresponding decrease from the basal side. Error bars indicate s.e.m.

View larger version (90K): [in a new window]   Fig. 3. Localization of non-actin binding YFP-Myosin IIDN. The schematic of myosin II structure shows the hexamer of two heavy chains (blue), two essential light chains (red) and two regulatory light chains (yellow). The schematic of mYFP-myosin IIDN shows the actin-binding head domain of the heavy chain replaced with the YFP moiety (green), resulting in two forms of YFP-containing myosin (heterodimer and homodimer) when expressed alongside endogenous myosin II, with both forms compromised in their ability to bind actin. (A,G) Cells undergoing cytokinesis are identified by antibody staining for phosphohistone H3 (red), nuclei are stained with Hoechst (blue) and myosin II (A) is stained with anti-myosin II antibody (green), whereas mYFP-myosin IIDN (G) is stained with anti-GFP antibody (green). Both myosin II and mYFP-myosinIIDN are localized to the cytokinetic furrow of dividing cells (arrow). (B-F,H-L) Anti-NRT staining (red), anti-myosin II staining (green in B-F) and anti-GFP staining (green in H-L). Despite the tendency of mYFP-myosin IIDN to form aggregates (asterisk in H), both the endogenous myosin II and the mYFP-myosinIIDN localize to the basal cellularization front (arrow in B,C,H,I), where levels later become greatly reduced in the ventral cells (arrow in F,L). This reduction is, however, delayed (E,K) and patchy (F,L) for mYFP-myosin IIDN. mYFP-myosin IIDN fails to localize to the apical side of ventral cells during gastrulation (compare asterisks in D,E with those in J,K).

View larger version (88K): [in a new window]   Fig. 4. Myosin localization in RhoGEF2 mutant embryos. (A-F) Localization of myosin (white) to the cellularization front of Ore and RhoGEF2 mutant embryos (arrow in A,D). The cellularization front of RhoGEF2 mutant embryos is irregular at mid cellularization (E). (G-L) Localization of myosin during gastrulation. For precise staging, embryos were viewed under oil and individually fixed at the onset of gastrulation. Within this range of ages, 41% (n=39) of Ore embryos accumulate myosin on the apical side of ventral furrow cells (asterisk in H), but in RhoGEF2 mutant embryos collected in the same way (n=30), myosin never accumulates apically above background levels (asterisk in K), whereas basal loss of myosin proceeds normally (arrow).

View larger version (89K): [in a new window]   Fig. 5. Myosin localization in Rho-kinase mutant embryos. Staining for Dorsal protein (red) marks ventral cells, nuclei are Hoechst stained (blue) and anti-myosin II antibody staining is in green. (A-F) Myosin localization in control embryos (A-C) and equivalently staged Rho-kinase (ROK) mutant embryos (below, D-F). ROK embryos have irregular incorporation of myosin to the cellularization front (+, myosin present; –, reduced myosin incorporation), irregular nuclear morphology and a failure to localize myosin to the apical side of ventral cells at the onset of gastrulation (compare asterisk in C,F). (G-K) Control embryos (G,H) and ROK mutant embryos (J,K) at later stages of development. Cells in later ROK mutant embryos are irregular (K). e, epidermal cell layer; f, folds; m, mesodermal cell layer. (I,L) nullo mutant embryos show both basal loss (arrow) and apical localization of myosin (asterisk) in ventral cells at gastrulation (L), despite earlier cellularization defects (I).

View larger version (91K): [in a new window]   Fig. 6. The effects of fog signaling on cell surface morphology. Scanning electron micrographs of Ore-R (A,C,E) and UASfog-expressing (B,D,F) embryos. Inset shows the rounded apical surface of cells in Ore-R and the apical flattening seen in 23% (n=21) of embryos expressing UASfog12. (C,D) Embryos expressing the stronger line, UASfog6, show disrupted, irregular, delayed ventral furrow formation (VF) and regions of apical flattening (asterisk) compared with Ore-R. (E,F) At the onset of germband extension, cells that fail to be internalized round up (asterisk) and lateral folds appear (arrowheads) in UASfog6-expressing embryos.

View larger version (96K): [in a new window]   Fig. 7. The role of adherens junctions in response to fog signaling. (A-G) Staining of ventral (A-D,F) or lateral (E,G) cells showing the position of adherens junctions (zona adherens, ZA) and basal junctions (BJ) stained with anti-ARM antibody (red) and anti-myosin II antibody (M, green). (A-C) Movement of junctions in Ore-R embryos: ZA are located subapically in all cells at the end of cellularization (A) and this subapical position is maintained in lateral cells, while ZA move to the very apical edge of ventral cells at the onset (B) and during (C) gastrulation. (D,E) Control embryos from mat67;mat15 mothers show the same dynamics of ZA movement as seen in Ore-R, with junctions located at the very apical edge of ventral cells (D) and subapically in lateral cells (E) and at the onset of gastrulation. (F,G) Embryos expressing UASfog6 from the mat67;mat15 driver line show an apical shift of junctions in all cells both ventral (F) and lateral (G). Inset in E and G: non-specific background staining has been amplified in the blue channel to reveal the cell outline. (H-M) Localization of myosin II (green) to the cellularization front of wild-type embryos, and embryos in which junction formation has been disrupted (by expression of UASnullo, or by germline clone reduction of arm protein). Neurotactin staining is shown in red (except the arm panel in H where red staining is non-specific cell surface stain). (H) Myosin localizes normally to the cellularization front of wild-type, UASnullo and arm germline clone embryos. (I,J) Myosin localizes normally to the apical surface of ventral cells (arrow) at gastrulation in wild-type (I) and UASnullo (J) embryos. (K-M) Images of the apical surface of ventral cells. Cells are outlined by Nrt staining (red). Myosin (green) is localized throughout the apical surface of wild-type cells (K) but is constricted to a tight ball (arrow) within the cells of UASnullo (L) and arm germline clones (M), leaving large black (non-stained) areas without myosin that are not seen in wild type.

View larger version (20K): [in a new window]   Fig. 8. Model of fog function in controlling cell shape change. (A) The patterning gene twist (twi), a transcription factor, specifies mesodermal fate of the ventral cells. As a consequence of twi expression, these cells activate transcription of fog (arrow), resulting in the production and secretion of fog protein from the apical side of the cell (blue dots). (B) Reception of fog signal on the apical side of the cell results in localized activation of ROK (red asterisk) which in turn activates the contractility of myosin with actin. This local source of actomyosin contractility drives myosin (pink) to the apical side of the cell (arrows). (C) The actin-myosin cytoskeleton is tethered to the cell surface through adherens junctions (orange). The force generated by apically localized contraction of the actin-myosin cytoskeleton therefore pulls down and flattens the domed apical cell surface and draws the adherens junctions up to the apical edge of the cell (arrows). (D) The continued contraction of apical actin-myosin exerts further force on the adherens junctions, pulling them close together, and resulting in the apical constriction of the cells.