NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Riddle DL, Blumenthal T, Meyer BJ, et al., editors. C. elegans II. 2nd edition. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory Press; 1997.

Cover of C. elegans II

C. elegans II. 2nd edition.

Show details

Section VICell Migration and other Aspects of Cellular Phenotype

A handful of cells undergo long-range migrations during embryogenesis (Sulston et al. 1983; Hedgecock et al. 1987). The somatic mesoblast M and its contralateral homolog, the right intestinal muscle cell, the gonadal mesoblasts Z1/Z4, and the coelomocyte mother cells all migrate posteriorly along the ventral body wall from their origins in the head to stereotyped positions along the midbody. The head-mesodermal cells migrate dorsally alongside the developing pharynx using UNC-6, possibly that which is secreted from pharyngeal neuron I5, as a repulsive cue (Hedgecock et al. 1990; Wadsworth et al. 1996). Somewhat later, the ALM and CAN neurons migrate posteriorly along the lateral hypodermis from their origins in the head, whereas HSN migrates anteriorly from the tail. More than 20 known genes are required for these embryonic cell migrations (Trent et al. 1983; Hedgecock et al. 1987; Desai et al. 1988; Manser and Wood 1990; Garriga et al. 1993b). Some genes are highly specific, but most mutations affect the migrations of multiple cell types, often including both mesoblasts and neuroblasts.

The navigational programs that regulate cell migration and process outgrowth are parts of larger programs that control all aspects of cellular phenotype including division, differentiation, and death. Genetic studies have identified the regulatory mechanisms that control migration and other cellular phenotypes. For example, whereas egl-5 controls diverse aspects of HSN development including cell migration and neuronal differentiation (e.g., growth cone navigation and neurotransmitter synthesis), egl-43 and unc-86 are required specifically for HSN migration and differentiation, respectively (Desai et al. 1988; Finney et al. 1988). All three genes encode transcription factors expressed in the HSN motor neurons; EGL-5, EGL-43, and UNC-86 are a Hox protein, a zinc-finger protein, and a POU-homeodomain protein, resepectively (Fig. 5) (Clark et al. 1993; Garriga et al. 1993b; Wang et al. 1993; Finney et al. 1988; Finney and Ruvkun 1990; C. Guenther and G. Garriga, unpubl.; see McGhee and Krause, this volume).

A. wingless/wnt1 Signaling Pathway in Q Neuroblast Migrations

The postembryonic development of the paired neuroblasts QR and QL illustrates how migrations are interwoven with cell division, determination, and death in a simple cell lineage (Sulston and Horvitz 1977; Sulston et al. 1980; Hedgecock et al. 1987). Neuroblasts QR and QL each generates a small number of sensory neurons during the L1 stage (Sulston and Horvitz 1977). Starting in midbody, QR descendants migrate anteriorly for stereotyped distances along the right lateral hypodermis, where they divide again or differentiate, whereas homologous QL descendants on the left-hand side remain stationary or migrate posteriorly. Nearly 20 genes have been identified that affect the navigational responses of the Q neuroblasts (Chalfie et al. 1983; Trent et al. 1983; Kenyon 1986; Hedgecock et al. 1987, 1990; Clark et al. 1993; Wang et al. 1993; Harris et al. 1996; Hishida et al. 1996; E.M. Hedgecock, unpubl.). The program of anteriorly oriented movement apparently represents a default program for Q neuroblasts, whereas posteriorly oriented movement is contingent upon a developmental signal normally active only on the left side of the body. Expression of the Hox gene mab-5 in Q cells is both necessary and sufficient for the latter program. In mab-5 loss-of-function (lf) mutants, both neuroblasts follow the default program, but in mab-5 gain-of-function (gf) mutants, both cells undergo posteriorly oriented movements (Kenyon 1986; Salser and Kenyon 1992). The Q migrations share several features in common with sex-myoblast migrations (see below); both cell types evidently employ an integrin-dependent mechanism of cell motility (Fig. 5) (Gettner et al. 1995; E.M. Hedgecock et al.; P.D. Baum and G. Garriga; both in prep.).

The distinct responses of the Q neuroblasts are controlled by the wingless signal-transduction pathway. In mig-5 (lf) mutants, both neuroblasts follow similar default programs, migrating anteriorly like wild-type QR. In mig-5(lf) mab-5(gf) double mutants, both cells follow nondefault programs, placing mig-5 upstream of mab-5 in this signaling pathway (Guo 1995). Similarly in lin-17 mutants, both Q neuroblasts often follow the default program. Moreover, QR, which normally expresses MAB-5, often fails to express MAB-5 in lin-17 mutants, placing lin-17 upstream of mab-5 (Harris et al. 1996). MIG-5 is an ortholog of the Drosophila protein Dishevelled (Guo 1995) (Fig. 5), the first known downstream component in the Wingless signaling pathway (Klingen-smith et al. 1994; Sussman et al. 1994; Theisen et al. 1994; Sokol et al. 1995), and LIN-17 is an ortholog of Drosophila Frizzled (Sawa et al. 1996). A new member of the frizzled gene family Drosophila Dfz2 has recently been shown to encode a wingless receptor (Bhanot et al. 1996). wingless/wnt1 and related genes encode a family of secreted proteins that control cell fates and planar polarity during epithelial development (Klingensmith and Nusse 1994). These proteins can organize cell patterning over large distances through both direct signaling and the relaying of Wingless-induced or related asymmetric polarity cues from cell to cell possibly via the frizzled pathway (Vinson and Adler 1987; Wong and Adler 1993; Theisen et al. 1994). A simple model is that seam cells on the left-hand side secrete a Wingless protein that signals QL and its descendants to remain stationary or navigate posteriorly by expressing MAB-5 and its target genes. winglesshomologs have been identified in C. elegans, but it remains to be determined whether they play a part in Q neuroblast migrations (Kamb et al. 1989; Shackleford et al. 1993; Herman et al. 1995; Sawa et al. 1996).

B. FGF Signaling Pathway in Sex-Myoblast Migration

The hermaphrodite sex myoblasts illustrate how a hierarchy of path and target cues position sex muscles precisely over the developing uterus and vulva (Thomas et al. 1990; Stern and Horvitz 1991; Clark et al. 1992; DeVore et al. 1995). In the late L1 stage, the mesoblast M generates paired sex myoblasts (Sulston and Horvitz 1977). During the L2 stage, these cells migrate anteriorly along the body wall to flank the gonad. During the L3, they divide to generate uterine and vulval muscle cells. During the L4, these muscles differentiate, forming stereotyped attachments to body wall, uterus, and vulva. In the adult, these muscles mediate egg laying.

Ablation of the gonad rudiment reveals that initial anterior movement of the sex myoblasts along the body wall is independent of the gonad (Thomas et al. 1990). However, these cells often undershoot or overshoot their destination in operated larvae, suggesting that a gonad-dependent cue is important for precise positioning. Indeed, sex myoblasts can accurately find the gonad in dig-1 mutant larvae, in which the gonad rudiment is displaced to nearby sites on the body wall (Thomas et al. 1990). From these and related experiments, it was proposed that the anchor and uterine blast cells of the gonad provide a diffusible chemoattractant for refining the final positions of the sex myoblasts (Stern and Horvitz 1991).

In egl-15 and egl-17 mutants, gonadal cues actually repulse the sex myoblasts (Trent et al. 1983; Stern and Horvitz 1991; DeVore et al. 1995). In intact mutants, sex myoblasts halt their migration prematurely and occasionally even reverse direction. If the gonad is ablated, these cells can migrate farther anteriorly, although their final positions are more variable than similarly operated wild-type animals. Two models have been proposed to explain why the gonad repels the sex myoblasts in egl-15 and egl-17 mutants. In the first, the egl-15 and egl-17 mutations remove an attractive signal to reveal an underlying repulsive signal. In a second model, egl-15 and egl-17 mutations transform an attractive signal into a repulsive one. Several other genes cause similar sex myoblast migration defects when mutated alone (e.g., ksr-1 ) or in combination with mild egl-15 alleles (e.g., sem-5 ) (Clark et al. 1992; DeVore et al. 1995; Sundaram and Han 1995).

EGL-15 is an ortholog of the fibroblast growth factor (FGF) receptor, SEM-5 is an ortholog of human Grb2 and Drosophila Drk, which couple receptor tyrosine kinases to Ras activators, and KSR-1 is a novel Raf-like protein (Fig. 5) (Clark et al. 1992; Lowenstein et al. 1992; Simon et al. 1993; Stern et al. 1993; DeVore et al. 1995; Sundaram and Han 1995). FGF signaling pathways have been widely implicated in cell migration and neurite outgrowth. Mutations in the Drosophila FGF receptor gene breathless, for example, disrupt migrations of both neuroglial and tracheal cells (Klambt et al. 1992; Reichman-Fried et al. 1994). Similarly, activation of FGF receptors on cultured neurons promotes neurite outgrowth, acting synergistically with several known cell adhesion molecules (Chao 1992; Williams et al. 1994). Knowledge of the sources and distribution of both ligands and receptors will no doubt be important for understanding the role of FGF signaling in these cell migration programs.

C. A Hormonal Signaling Pathway in Gonadal Migrations

The multistage migrations of the gonadal leader cells illustrate how changes in cell substrate and direction, and transitions between motile and stationary states, can be programmed in postmitotic cells (Kimble and Hirsh 1979; Hedgecock et al. 1987). The gonadal leaders comprise two distinct cell types generated from mesoblasts Z1 and Z4 in the late L1 stage which organize gonadogenesis and polarize germ-line maturation along their distal-proximal axis (Kimble and Hirsh 1979). Depending on nematode species and sex, one or both types of leaders undergo complex, stereotyped migrations along the larval body wall which determine the shape and position of the mature gonad (Chitwood and Chitwood 1974). In C. elegans males, only the proximal leader (linker cell) migrates, whereas in hermaphrodites of this species, only the distal leaders (distal-tip cells) migrate (Kimble and Hirsh 1979; see Schedl, this volume). Each leader cell describes a unique trajectory on the body wall that can be analyzed as a succession of unidirectional migrations on uniform substrates (Fig. 6). Most segments follow the natural axes of the body wall, but the oblique movement of the linker cell near the start of L4 stage is best described as a ventral movement superimposed on a continuous posterior migration.

Figure 6. Migrations of the hermaphrodite distal-tip cells (top) and male linker cell (bottom) on the larval body wall in C.

Figure 6

Migrations of the hermaphrodite distal-tip cells (top) and male linker cell (bottom) on the larval body wall in C. elegans (Kimble and Hirsh 1979; Hedgecock (more...)

In principle, elaborate labeled pathways on the larval body wall might guide the migrations of each gonadal leader. Instead, it appears that these cells navigate by following several global directional cues, e.g., UNC-6, in strict sequence. Changes in cell direction do not correspond to obvious guidepost cells that could provide signals to advance the navigational program. Moreover, when leader cells are generated at ectopic sites on the body wall by reiterated division (e.g., unc-39 ) or displacement of the gonad rudiment (e.g., dig-1 ), they migrate for more or less normal times and distances, not to fixed landmarks (Thomas et al. 1990; E.M. Hedgecock, unpubl.). By inference, the navigational programs of these cells advance either autonomously or in response to hormonal signals available throughout the larva. For example, UNC-5 is expressed in the hermaphrodite distal-tip cells just hours before the L3 molt immediately presaging their dorsal movements (M. Su et al., in prep.). This and other differences in distal-tip and linker cell behavior reflect distinct navigational programs, not sex-specific extracellular signals. In particular, intersexes produced by mosaicism for tra-1 , or mutations in tra-1 that selectively disrupt its activity in soma (e.g., e1488) or gonad (e.g., rh132), reveal that gonadal leaders are indifferent to the sexual phenotype of hypodermis and skeletal muscle which form the migration paths (Hodgkin 1987a; Herman and Hedgecock 1990; Hunter and Wood 1990).

Development of the gonad and soma is closely coordinated over a range of growth conditions in wild-type larvae. In particular, changes in gonadal leader direction or motility during the L2, L3, and L4 larval stages all occur soon after commitment to another molt cycle, as judged from new cuticle synthesis and DNA replication in the hypodermis. Moreover, migration ceases at the dauer molt, indicating that gonadal leaders respond to a general signal for developmental arrest. Mutations at several loci, known as heterochronic genes, can advance or delay stage-specific events in the hypodermis and other somatic tissues, but they have little or no effect on gonad and germ line (Ambros and Horvitz 1984; Ambros, this volume). Two new genes, mig-7 / daf-12 and mig-8 , are required to advance aspects of gonadal development, including the navigational programs of the leader cells (A. Antebi et al., in prep.). In these mutants, leader cells may continue without turning or completely cease migration at various stage transitions. Like previously described heterochronic mutants, mig-7 / daf-12 also delays development of somatic tissues, causing repetitions of earlier events.

mig-7 mutants have proven to be novel alleles of the dauer pathway gene daf-12, which encodes a nuclear hormone receptor most similar to Drosophila DHR96 (Fig. 5) (Riddle et al. 1981; Yeh 1991; Fisk and Thummel 1995; Riddle, this volume). Conceivably, a hormone, acting through DAF-12 and related receptors, synchronizes development of soma and gonad at the commitment to each postembryonic stage. At these transitions, heterochronic gene activities within each tissue select stage-specific transcriptional cascades. The targets of DAF-12 activation or repression are unknown, but they may include tissue-specific transcription factors (e.g., Hox genes) as well as other heterochronic genes. Indeed, several transcription factors have been implicated in the gonadal leader migrations (Fig. 5) (EGL-5/Abdominal-B, Chisholm 1991; Wang et al. 1993; LIN-39/Sex combs reduced, Clark et al. 1993; CEH-18, Greenstein et al. 1994; and VAB-3/PAX-6, Chisholm and Horvitz 1995; Zhang and Emmons 1995; A.D. Chisholm and H.R. Horvitz, in prep.). In some cases, a gonadal focus has been demonstrated for the migration phenotype (Chisholm 1991; Greenstein et al. 1994). Finally, DAF-12 controls, perhaps indirectly, the expression of UNC-5 and other cell surface receptors used in leader cell migration (M. Su et al., in prep.).

Copyright © 1997, Cold Spring Harbor Laboratory Press.
Bookshelf ID: NBK20003
PubReader format: click here to try


  • PubReader
  • Print View
  • Cite this Page

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...