Regions within a single epidermal cell of Drosophila can be planar polarised independently
- University of Cambridge, United Kingdom
Figures
Figure 1
Part of the wild type ventral epidermis.
(A) Ventral abdominal epidermis, including the central midline, showing rows of predenticles 1–6. (B) The denticles made later by the same larva. (C) A model: ds expression combined with expression of fj, gives the presumed pattern of Ds activity and explains the orientation of the rows 0–6. Each row points towards the neighbouring cell with the most Ds. The line within a cell (usually sloped) indicates that each cell has different amounts of Ds at its anterior and posterior limits (see Figure 4). In all the figures, one individual is imaged first at mid second stage and later after moulting to third stage. Note almost exact correspondence between predenticles and denticles in all cases. Cell boundaries (Ecad) are shown in red and actin highlighted in green (see ‘Materials and methods’). Cells shaded in blue belong to the posterior compartment (Lawrence and Struhl, 1996). T1 and T2 indicate the two rows of tendon cells, shaded in grey in C. Although tendon cells do not show actin predenticles, they show characteristic actin palisade-like structures (Saavedra et al., 2014). The blue dotted lines are transects to locate the diagrammatic cross sections shown at the right of each figure. Anterior is to the left.
Figure 2 with 1 supplement
Atypical cells.
(A–D) One atypical and multipolar cell, largely in row 4, is shown, in B–D (shaded in magenta). The transects shown as dotted lines in C and G are illustrated in D and H with the presumed amounts of Ds and Fj as well as the presumed activity of Ds. (E–H) One atypical cell of row 2 is shown; labelling as in other figures. See also Figure 2—figure supplement 1.
Figure 2—figure supplement 1
Atypical cells: more examples.
An example (A-D) showing two atypical cells, one in row 2, one in row 4. Even though much of the row 2 cell abuts, not T1 as is ‘typical’, but in other row 2 cells, the polarities of all denticles are always normal (Table 1). The row 4 atypical cell is of interest because it has only a small promontory that abuts another row 4 cell, and yet this small promontory has one posteriorly oriented denticle. Presented as the other figures. Related to Figure 2.
Figure 3 with 1 supplement
Testing the model.
(A–D) An epidermis lacking both the ds and ft genes. The cell shaded in magenta is a row 4 cell which has an atypical disposition. The predenticle and denticle orientation is variable and awry. (E–H) A clone of two cells over-expressing ds (marked in yellow, E and shaded in yellow, F–H) imposes an orientation on all or parts of the neighbouring wild-type cells. Note that one reoriented cell (*) appears to have no direct contact with clonal cells expressing ds but only with neighbours of such cells. We have seen this in other cases. In the adult abdomen such propagation extends to several cell diameters (Casal et al., 2002; Casal et al., 2006) and has also been reported in larvae (Repiso et al., 2010; Donoughe and DiNardo, 2011). See also Figure 3—figure supplement 1.
Figure 3—figure supplement 1
Breaches in rows of tendon cells.
(A–D) An example of the effect on cells of an interruption in the T1 row. This is associated with an unpolarised cell shaded in magenta. (E–G) shows the similar effect of a breach in the T2 row of cells. These results are interpreted in D and H and fit our model of Ds activity very well. Presented as the other figures. Related to Figure 3.
Figure 4
Polarised conduits in the cells.
A hypothetical view of how Ds and Ft polarise cells or parts of cells. All membranes contain both kinds of dimers: these are Ds in the cell x (Dsx), linked to Ft in the neighbouring cell y (Fty), or Ft in the cell x (Ftx) linked to Ds in the neighbouring cell y (Dsy). We proposed (Casal et al., 2006) that intercellular bridges consisting of heterodimers of Ds and Ft are asymmetrically distributed in a polarised cell and determine the polarity of that cell. In the diagram, we indicate a majority of Ds by yellow, and a majority of Ft by rufous. Some membranes contain similar numbers of Ds and Ft and these are shown with alternating blotches of the two colours. Arrows indicate the sign and the paths of the oriented conduits that span between facing and limited areas of membrane. Such conduits can give small parts of cells an individual polarity, as in the atypical cell in row 4 (red arrows). The tendon cells, T1 and T2, largely drive the segmental pattern of Ds activity—they have low Ds activity and therefore the majority of heterodimers formed between a T2 and a row 4 cell are Ds4–FtT2. Similarly at the boundary between a row 2 and a row 3 cell, the majority of the heterodimers are Ft2–Ds3; partly because at the opposite boundary between a T1 and a row 2 cell, the heterodimers are largely FtT1–Ds2. Where row 3 and row 4 cells meet in the wild type, they are imagined to have similar levels of Ds3–Ft4 and Ft3–Ds4 because, at that cell junction similar, but opposite, effects from very low Ds levels in both T1 and T2 tendon cells converge. However, in the red-arrowed region, at the anterior limit of the atypical cell, the heterodimers are mostly Ft3–Ds4. We propose that in this red-arrowed region, the deployment of heterodimers is the outcome of a different comparison made between facing subregions of the anterior and posterior limits of this atypical cell. What is different about this comparison? In this red-arrowed region, the cell's anterior neighbours (row 3) have less Ds than the posterior neighbour (a normal row 4). As a result, at the boundary between row 3 and atypical row 4, the heterodimers will be mostly Ft3–Ds4. However, at the facing boundary, between atypical and normal row 4, there are balancing influences from anterior and posterior directions, as in a boundary between a normal row 3 and a normal row 4, and as a consequence the amounts of Ds4–Ft4 and Ft4–Ds4 heterodimers should be similar.
Tables
Table 1
Atypical cells: quantitation of denticle polarities in relation to neighbouring cells showing the effect of the Ds/Ft system
| Wild type | ds− ft− | ||||||
|---|---|---|---|---|---|---|---|
| Anterior neighbour | Denticle polarity of atypical Row 2 cells† | Posterior neighbour | Anterior neighbour | Denticle polarity of atypical Row 2 cells§ | Posterior neighbour | ||
| Anteriorly | Posteriorly | Anteriorly | Posteriorly | ||||
| T1 cell | 0 | 52 | Row 3 cell | T1 cell | 16 | 21 | Row 3 cell |
| Row 2 cell | 0 | 35 | Row 3 cell | Row 2 cell | 14 | 13 | Row 3 cell |
| Anterior neighbour | Denticle polarity of atypical Row 4 cells‡ | Posterior neighbour | Anterior neighbour | Denticle polarity of atypical Row 4 cells¶ | Posterior neighbour | ||
|---|---|---|---|---|---|---|---|
| Anteriorly | Posteriorly | Anteriorly | Posteriorly | ||||
| Row 3 cell | 110 | 8* | T2 cell | Row 3 cell | 54* | 37 | T2 cell |
| Row 3 cell | 8** | 41 | Row 4 cell | Row 3 cell | 24 | 20** | Row 4 cell |
-
†
Denticles of 29 atypical cells.
-
‡
Denticles of 27 atypical cells. 8 of 8* (and 6 of 8**) were predenticles located in ambiguous positions and their denticles were arbitrarily allocated to those classes favouring the null hypothesis. Fisher exact test p-value: <2.2−16.
-
§
Denticles of 23 atypical cells. Fisher exact test p-value: 0.6135.
-
¶
Denticles of 24 atypical cells. 1 of 54* (and 1 of 20**) were predenticles located in ambiguous positions and their denticles were arbitrarily allocated to those classes disfavouring the null hypothesis. Fisher exact test p-value: 0.7104.
-
Numbers in bold emphasise the main result of the table that is, the effect of neighbours on denticle polarities in the wildtype. This effect does not exist in the mutant larvae.
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Regions within a single epidermal cell of Drosophila can be planar polarised independently
eLife 4:e06303.
https://doi.org/10.7554/eLife.06303
