Leg evagination. Showns are fixed legs from third instar larvae (L)(panel A), white prepupae (0) (B), 2 (C), 4 (D) and 30 hr (F) after white prepupae. The omb domain is in green and the legs are counterstained in red with phallodin, which stains the actin cytoskeleton.

Cartoon models of wing and leg eversion. Panels A-D show a wing disc (en face (A) and cross section (B-D) as it begins to evert (D). Arrows show the direction of eversion. Panels E - J are of legs. Third instar leg discs are shown en face (E) and in cross section (F). Panels G - J are cross sections of legs at progressively later stages in eversion.

Cell rearrangement during evagination. Shown are images from a prepupal leg at various times during a time lapse experiment. The experiment started at about 3 hrs awp. This is the same time lapse experiment that is presented in Supplementary Movie 3. The time in minutes since the beginning of the experiment is shown. The time points chosen were ones that were both informative and where essentially all of the 8 marked cells (and their progeny) were easily visible. Interpretive drawings are shown that identify 8 cells in each of the time points. The magenta cell from the interpretive drawing is marked on the micrographs by a small magenta circle. Two of the 8 cells divided and they are shown in the interpretive drawings as two cells of the same color.

Wing evagination. Shown are 2, 4 and 8 hr prepupal wings as well as a 28 hr pupal wing for comparison. The wings were stained with the 22C10 monoclonal antibody which stains neurons. The locations of the various landmark cells are shown on the 2 hr (and in most cases later) images. The downward arrow points to landmark M1, the downward arrowhead to M2, the upward arrowhead to M3, the upward arrow to M4 and the oblique arrow to M5. The lines in the 4 hr image are examples of the M1 through M4 measurements. The lines in the 8 hr image show how the W1 and W2 distances were obtained. Note the dramatic change in the relative position of landmark M3 as development proceeded.

Changes in the ptc domain during wing evagination: Panels show a wing disc from a third instar larvae (L), a white prepupae and from 2 and 8 hr after white prepuae. All preparations are of fixed material. In all panels green is GFP expressed using ptc-Gal4. In the lower 4 panels 22C10 antibody staining is shown in red. This antibody stains sensory neurons in the wing. In the upper two panels red is actin staining. The positions of the M1 and M2 (middle panels) and M4 and M5 cells are indicated by arrows and arrowheads. Note the large increase in the distance between M1 and M2, and M4 and M5 from 2 to 8 hr. Note also the decrease in the width of the ptc domain.

The omb domain and leg evagination. Panels A-F are from in vivo imaging of a prepupae that is expressing GFP in the omb domain. The panels show the tibia and basitarsus and parts of more distal tarsal segments. The time (in hr) after white prepupae is noted. The same location in the leg at the different time points is shown by arrowheads (mid-tibia), short arrows (tibia/basitarsus border) and long arrows (basitarsus/tarsus border). Note the lengthening and thinning of the omb domain from 3.25 to 5.25 hr. Note also the thickening and increase in cell number after 5.25 hr. Panels G-K show a fixed tibia from wpp, 2, 4, 6 and 8 hr legs. Note the early thinning and elongation followed by thickening. The spiral cell arrangement that is characteristic of pupal legs is visible by 6 hr.

Cell movement during evagination. Panel A shows the initial and final positions of cells that could be followed throughout the movie provided in Supplemental Movie 4. The images were all aligned using the orange square cell (marked by an arrow) as an anchor. This simplified the analysis. The leg as a whole moved during the experiment and the slide needed to be re-centered several times. Cells are identified by the same symbols as in Supplementary Movie 4. The images are of cells in the omb domain of an evaginating tarsal segment. Panel B shows the position of a selected subset of cells every 4th time point (every 20 minutes). The earliest time point is identified by a thicker outline. Subsequent time points are over layed. Panel A shows the marked tendency for cells to move upward toward the anchored cell and both proximally and distally. This is as expected for elongation by cell intercalation. The movement was not consistent or sharply directed as is seen in Panel B. Panels C and D are the starting and final images from the movie. The asterisk marks the cell marked with an orange square in panels A and B.

Cell rearrangement and cell division: Panels A and B are of the same MARCM clone at 3.45 and 8 hr. A single optical plane from a stack is shown and some cells are out of the plane of focus. Panels a and b are interpretive drawings of the clone cells from panels A and B. Note the increase in cell number during this period. Panels C and D show a blow up of region 1 from panel A. The arrow in panel is parallel to the proximal/distal axis. The arrow in panel C points to a cell that has divided in panel D. The sister cells are indicated by arrowheads. Panels C and D were the time points right before and after the cell division. Interpretive drawings of the cells in panels C and D are shown in c and d. Panels E and F are a blow up of the cells from region 2 of panel A before and after these cells divide. Interpretive drawings of this are shown in panels e and f. Panels G, H and I are of a small region of the ptc domain from a tibia at three time points spread out over 30 minutes. Interpretive drawings of these images are shown in panels g, h and i. Note the rearrangement of cells resulting in the green and red cells touching each other in panel I, but not G. At the same time the purple and blue cells lose contact. Panels J and K are of part of a large MARCM clone 1 hr apart. Panels j and k are interpretive drawings. Once again note the change in cell neighbors with the purple and red cells losing contact with one another.

Quantitation of wing and leg evagination. Panel A shows the change in the ratio of wing landmark distances as a function of time (see Methods for more details). Error bars show the standard error of the mean. Note that the relative distance for all of the wing landmark neurons to the distal margin increase over time and the increases are larger for the more proximal landmarks consistent with elongation happening throughout the proximal distal axis. Time is in hr after white prepupae (wpp). The number of wings measured was 5, 8, 6, 11, 6, 6 and 12 for the 1, 2, 3, 4, 5, 7 and 8 hr time points respectively. Panel B show the width of the ptc domain at landmark neurons in the wing at 2, 4 and 8 hrs. The number of cells separating the M1 and M2 landmarks and the M4 and M5 landmarks are also shown. As in panel A the error bars show the standard error of the mean. Each of the W1, W2 and 1to2 points represent 8 to 10 independent measurements. The W4, W5 and 4to5 points represent 6 to 8 indpendent measurements (in some cases a measurement could not be made accurately for a particular time point due to the tissue not being flat at that location). Measurements that are significantly different from the previous measurement are indicated by an asterisks (* < 0.05, ** < 0.01). Note that there is a decrease in the width of the ptc domain from 2 to 4 hrs at all landmarks and an increase in the width from 4 to 8 hrs for several landmarks. Also note the substantial increase in the distance between landmarks. Panel C shows the width of the omb domain in the basitarsus and tibia at 2, 4, 6 and 8 hr. Error bars and significance indicated as in earlier panels. Each point represents counts from 6 - 8 legs. Panel D shows the distribution of division planes in leg clones. The most common division planes were from 46-60 degrees. The angle of the division plane was estimated for seventy nine divisions.