|
|
|
|
|
|
|
||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | ||||
| ||||||||||||||||||||
Files in this Data Supplement:
Fig. S1. ceh-20(mu290) is a recessive allele. nDf16 is a deficiency on linkage group III that removes the ceh-20 gene. The vertical gray line indicates the birthplace of the QR cell. The pink arrows indicate the anterior distance that QR and QR.p descendents migrate in wild-type animals. n>100 cells for each strain. Additional neurons in the body were present in some mutant animals analyzed in this study, raising concern regarding the accuracy of our recorded cell positions. Q.pax cell distributions in this study were analyzed at the end of L1 only in animals without additional neurons. Comparison of these cell positions with those determined by lineage analysis revealed that Q descendant positions determined by scoring at the end of L1 and by lineage analysis were very similar. We lineaged QR in ceh-20(mu290) and unc-62(mu232) animals and recorded the final positions of the QR.pax cells. The distribution of the QR.pax cells in the 20 ceh-20(mu290) animals lineaged, including nine in which extra neurons from the QR lineage persisted, revealed that over 60% of QR.pax cells were at or posterior to V4.a. Four out of 20 unc-62(mu232) animals whose QR cells were lineaged showed division of V5.a. Nevertheless, over 40% of the QR.pax cells in these twenty unc-62(mu232) animals were at or posterior to V4.a. These findings are consistent with the final positions of the QR.pax cells of ceh-20(mu290) and unc-62(mu232) determined by scoring cell positions at the end of L1. Although this selection may slightly bias the distribution of cells, the major conclusion of our paper regarding migration (i.e. ceh-20 and unc-62 may promote anterior migration of Q descendants independently of lin-39 and are required for such cells to respond to mig-13) relies on observations that can be confirmed using other methods. For example, one observation that supports our conclusion is that some QR descendants migrated posteriorly instead of anteriorly in ceh-20(mu290) or unc-62(mu232) double mutants with egl-20. By lineage analysis, we confirmed such posterior migration of QR.p or QR.pa cells in three of twenty ceh-20(mu290); egl-20(mu320) animals and two of twenty-five unc-62(mu232); egl-20(mu320) animals. By contrast, zero of fifteen lin-39(n1760); egl-20(mu320) animals whose QR cells were lineaged showed posterior migration of QR descendants. Another observation that our conclusion relies upon is that the migration of BDU can also be shortened in ceh-20(mu290) and unc-62(mu232) mutants. The additional neurons, when present, were never anterior of V2.p and therefore did not interfere in the analysis of BDU positions. In addition, BDU is easily identified based on its characteristic small size and its central position along the dorsal ventral axis. Before we analyzed QR.pax positions in ceh-20 or unc-62 double mutants with Punc-119::mig-13::gfp, we recognized that such a selection bias may be problematic. Thus, in addition to the 100 animals without extra neurons that we scored, we also recorded the positions of all ectodermal lateral body neurons in the 22 ceh-20(mu290); Punc-119::mig-13 animals and the 17 unc-62(mu232); Punc-119::mig-13::gfp animals with additional neurons. After we identified BDU (distinguishable as described above), ALM [distinguishable by its dorsal position, presence of a ÔhoodÕ, and characteristic position along the anteroposterior (AP) axis], CAN (distinguishable by its oval shape, characteristic position just dorsal to the canal, and characteristic AP position) and HSN (distinguishable by its small size, ventral position, and characteristic AP position), we noted that all of the ectodermal lateral body neurons that were not BDU, ALM, CAN or HSN were never anterior to V3.p. Although it is theoretically possible that the Q.pax cells migrated into the head in the animals, we highly doubt this possibility since there were cells with the morphology of Q descendants in the mid-body region. In ceh-20(mu290) and unc-62(mu232) animals, Vn.a cells could sometimes divide, raising concern about using Vn.x cells as the ruler against which the extent of Q descendant migration was measured. When the Vn.a cells divide, their daughters remain in close proximity to the previous location of the mother cell. As seen in Figs S7 and S8, there is no great rearrangement of the positions of V descendant nuclei along the AP axis even when Vn.a divides. In ceh-20(mu290) and in all double mutants made with this allele (except for the unc-62(mu232) double mutant), the only Vn.a cell that failed to fuse and subsequently divided was V1.a (Table 4, data not shown). In ceh-20(mu290) animals and in all double mutants made with the mu290 allele, all of the QR.pax and QL.pax cells were posterior to V1 and its descendants. Thus, Q.pax cell positions in animals carrying this ceh-20 mutation could be analyzed accurately regardless of Vn.a division. None of the QL.ap cells and only a small percentage of QR.ap cells in animals carrying ceh-20(mu290) migrated near V1.a. In animals in which the cell of interest had migrated near a V1.a cell that had divided, we defined the ÔV1.a positionÕ as that bounded by a dorsoventral line through the center of the V1.aa nucleolus and a dorsoventral line through the the center of the V1.ap nucleolus. For single or double mutants carrying unc-62(mu232), most of the Q.pax cells were positioned between V3.x and V4.x. Notably, the anterior daughter of V3 and V4 rarely divided (Table 4) except in a double mutant with ceh-20(mu290). If the cell(s) of interest migrated near a Vn.a cell that had divided, the ÔVn.a positionÕ was defined as the region bounded by a dorsoventral line through the center of the Vn.aa nucleolus and a dorsal-ventral line through the center of the Vn.ap nucleolus. This method slightly increases the AP distance included as part of the Vn.a position for some Vn.a cells, but it does not invalidate our interpretation. Because the Vn.ax cells remained very close to the previous position of their mother cell Vn.a, the V descendants remain a valid ruler to measure the extent of Q descendant migration.
Fig. S2. ceh-20, unc-62 and lin-39 RNAi shorten the anterior migration of QR descendents. The vertical gray line indicates the birthplace of the QR cell. The pink arrows indicate the anterior distance that QR and QR.p descendents migrate in wild-type animals. n=100 cells for each strain.
Fig. S3. Distribution of QR.pax cells in ceh-20(mu290); unc-62(mu232) animals is similar to that in ceh-20(mu290) or unc-62(mu232) double mutants with mig-13(mu31). In ceh-20(mu290); unc-62(mu232) animals, a few QR.pax cells were positioned posterior to the birthplace of QR, suggesting that QR.p or its descendants migrated in the posterior rather than in the anterior direction. The vertical gray line indicates the birthplace of the QL cell. n=100 cells for each strain.
Fig. S4. Pan-neuronal mig-13 expression shortens the posterior migration of the CAN cell. The embryonic migration of CAN starts from the head and ends just posterior to V3 in wild-type animals. This posterior migration is not altered in mig-13(mu225), lin-39(n1760), ceh-20(mu290) or unc-62(mu232). mig-13 expression in all neurons in a wild-type background or in the mutant backgrounds listed above causes premature termination of this posterior migration. Similar distributions were observed for the left CAN cell in these strains (data not shown). The black arrow indicates the posterior distance that CAN migrations in wild-type animals. n=50 cells for each strain.
Fig. S5. Increased LIN-39 expression in Q cells and their daughters in ceh-20(mu290) animals. Wild-type (A-C,G-I) and ceh-20(mu290) (D-F,J-L) animals were stained with antisera against the LIN-39 protein (Maloof and Kenyon, 1998) and MH27 to outline cells (Francis and Waterston, 1991). Age of animals were as follows: 0-2 hours (before Q cell delamination), 1-4 hours (after Q cell delamination) and 4-6 hours (after Q cell division). Lateral views are shown for all panels except I, for which a ventrolateral view is shown. Anterior is towards the left in all panels. Q cells and lateral hypodermal cells adjacent to the Q cells (V4 or V5) are labeled. The unlabeled cell nucleus that stains brightly between V4 and QR in D-F is P7/8. LIN-39 expression in wild-type QR and daughters is described in the text. For QL, lin-39 may be expressed weakly in QL before delamination (G). After delamination but before division, lin-39 expression was either absent or weak (I). In ceh-20(–) animals, lin-39 was expressed more strongly in QL before (J) and after (K) delamination compared with wild type. After division (L), the QL daughters in some ceh-20(–) animals continued to express lin-39 strongly. In all unc-62(–) animals observed, lin-39 was expressed in QL before and after delamination as well as after division (Table 1). We never observed polarity reversals, but Q cell migrations (before division) were shortened in ceh-20(mu290) and unc-62(mu232) mutant animals. Lineage analysis starting 30 minutes after hatching showed that in wild-type animals, nearly 100% of QR nuclei migrate anteriorly over V4 before dividing (presence of metaphase plate signified division) and nearly 100% of QL nuclei migrate over V5 before dividing. However, in ceh-20(mu290) animals, nuclear migration of QR and QL are shortened: 90% (n=19) of dividing QR nuclei are between the birthplace of Q and V4 at division, and 74% (n=27) of dividing QL nuclei are between the birthplace of Q and V5 at division. unc-62(mu232) animals show a similar defect. In these animals, 86% (n=29) of QR and 80% (n=25) of dividing QL nuclei failed to migrate their full distance before division.
Fig. S6. ceh-20(mu290) and unc-62(mu232) shorten the anterior migration of QL descendents seen in mab-5(e2088) and egl-20(n585) animals. In ceh-20(mu290) and unc-62(mu232) double mutants with egl-20(n585), QR.p descendents can be positioned farther posteriorly than in wild type. The vertical gray line indicates the birthplace of the QL cell. n=100 cells for each strain.
Fig. S7. Vn.a cells divide inappropriately in unc-62(mu232) animals. Nomarski photomicrographs of 6- to 8-hour-old L1 larvae. (A,B). In these wild-type animals, V cells have divided. Their anterior daughters do not divide further. (C,D) Shown are examples of two unc-62(mu232) animals in which V1.a divided (C) or V6.a divided (D). Most unc-62(mu232) animals had at least one Vn.a cell that had divided (96% on right, 78% on left). In these single mutants, Vn.a divisions were observed in all Vn.a cells except V3.a. In ceh-20(mu290); unc-62(mu232) animals, 42% (right) and 52% (left) of V3.a cells divided, suggesting that all Vn.a cells had the potential to divide (Table 4). All doubly mutant animals displayed at least one divided Vn.a cell. Animals are oriented with anterior towards the left.
Fig. S8. Vn.a division occurs before fusion in unc-62(mu232) animals. (A) Nomarski micrograph of a unc-62(mu232) animal carrying jam-1::gfp in the L1 stage. In this animal, V6.a has just divided. (B) Fluorescence micrograph of the same animal as in A showing the outline of the V6.a daughter. As jam-1::gfp is expressed in adherens junctions on the cell membrane, fusion has not yet occurred. Dorsal is upwards and anterior is towards the left.
Fig. S9. V5.a can generate a Q-like lineage in unc-62(mu232) animals. Nomarski micrographs of animals at the end of L1. (A) In wild-type animals, Vn.a cells do not divide further. V4 and V5 daughters are shown. (B). In unc-62(mu232) animals, V5.a can divide further. In this animal, the left V5.a cell (V5L.a) divided in the Q lineage pattern to generate three additional neurons. The left V5.aap, which is labeled as ÔQ.ap from V5L.aÕ, even has the characteristic look of a Q.ap cell as it is slightly larger than the Q.pax cells and displays a prominent nucleolus. The proper QL.ap cell has migrated into the tail, which is out of the field shown here.
| ||||||||||||||||||||
