Mutations Affecting Nerve Attachment of Caenorhabditis elegans

Corresponding author: Shin Takagi, Division of Biological Science, Nagoya University Graduate School of Science, Furo-cho, Chikusa-ku, Nagoya 464-8602, Japan., i45116a{at}nucc.cc.nagoya-u.ac.jp (E-mail)

Communicating editor: R. K. HERMAN

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Using a pan-neuronal GFP marker, a morphological screen was performed to detect Caenorhabditis elegans larval lethal mutants with severely disorganized major nerve cords. We recovered and characterized 21 mutants that displayed displacement or detachment of the ventral nerve cord from the body wall (Ven: ventral cord abnormal). Six mutations defined three novel genetic loci: ven-1, ven-2, and ven-3. Fifteen mutations proved to be alleles of previously identified muscle attachment/positioning genes, mup-4, mua-1, mua-5, and mua-6. All the mutants also displayed muscle attachment/positioning defects characteristic of mua/mup mutants. The pan-neuronal GFP marker also revealed that mutants of other mua/mup loci, such as mup-1, mup-2, and mua-2, exhibited the Ven defect. The hypodermis, the excretory canal, and the gonad were morphologically abnormal in some of the mutants. The pleiotropic nature of the defects indicates that ven and mua/mup genes are required generally for the maintenance of attachment of tissues to the body wall in C. elegans.

GENETIC analyses using mutants have revealed genes that have specific roles in animal development. In Caenorhabditis elegans, genes involved in the various stages of neural development, i.e., determination and specification of cell fate, axonal guidance, and formation of synaptic connections, have been identified through analyses of mutant phenotypes (for review see ANTEBI et al. 1997 ; RUVKUN 1997 ). As survival of worms in the laboratory is relatively independent of the nervous system, most of the neural mutants analyzed extensively to date were viable. It is, however, possible that some of the genes (mutation of which leads to larval or embryonic lethality) may also be involved in the formation or maintenance of the nervous system. However, there have been few studies of the neural defects of lethal mutants. Here, we report lethal mutants of C. elegans exhibiting displacement or detachment of nerves.

Proper positioning is essential for the proper function of tissues or organs, and they are usually fixed at appropriate positions through attachment to other tissues or organs. This also applies to C. elegans. Their body wall muscles are anchored to the cuticle (the outer covering of the worm) by a specialized basement membrane and hemidesmosome structures in the hypodermis (a cellular syncytium that covers the worms and secretes the cuticle; FRANCIS and WATERSTON 1991 ; for review see MOERMAN and FIRE 1997 ). It was suggested that direct interaction of the muscles and the hypodermis is important in organizing the cell attachment structure (HRESKO et al. 1994 ). Groups of mutants with defects in muscle attachment (mua; PLENEFISCH et al. 2000 ) and muscle positioning (mup; GOH and BOGAERT 1991 ; MYERS et al. 1996 ; GATEWOOD and BUCHER 1997 ) have been isolated. Analysis of these mutants showed that the cellular origin of defects in establishment and maintenance of body wall muscle position can be either the muscles or the hypodermis (MYERS et al. 1996 ; GATEWOOD and BUCHER 1997 ; see also WILLIAMS and WATERSTON 1994 ). Positioning can be particularly important for the nervous system, since apposition of neuronal processes and their targets is a prerequisite for proper neural transmission. In contrast to the muscle, however, no specialized ultrastructure has been suggested for the attachment of the nervous system of C. elegans, except for a certain class of neurons (CHALFIE and SULSTON 1981 ; FRANCIS and WATERSTON 1991 ). Mutations affecting the positioning of nerves have not been screened systematically in C. elegans.

The ventral cord is a major longitudinal tract of worms, consisting mainly of motor neurons and their axons running in two parallel bundles. The ventral cord is embedded in the thickening of epidermal cells called the ventral hypodermal ridge and covered by basal lamina and, consequently, fixed to the body wall (WHITE et al. 1976 ). Through a screen of larval lethal mutants with morphological defects in the ventral cord, we identified a novel phenotype, Ven (for ventral cord abnormal). Ven mutants exhibit gross anatomical defects in the nerve cords, including their complete detachment from the body wall. Here, we report the screening, mapping, and partial characterization of Ven mutants. The screen led to the identification of mutations in three novel genetic loci, ven-1, ven-2, and ven-3, as well as mutations in previously identified mua/mup genes. All the Ven mutants exhibited the muscle attachment/positioning defects. We will also describe the close association of detachment defects of nerves and muscles and discuss the possible cellular mechanisms leading to nerve detachment.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

General methods, genes, and alleles:
Strains were grown at 20° on NG plates and were maintained as described by BRENNER 1974 . The wild-type strain used was Bristol N2. The strain RW7000 was used in interstrain crosses for sequence-tagged site (STS) mapping (WILLIAMS et al. 1992 ). Strains with the following mutations and rearrangements were also used.

• LG(linkage group)I: bli-4(e937), vab-10(e698), unc-54(e190)

• LGII: mup-1(e2347), mua-1(rh160), dpy-2(e489), dpy-10(e128), let-236(mn88), let-253(mn181), lin-26(mc1), rol-6(e187), unc-4(e120), mix-1(mn29), unc-53(e404, n569), unc-52(e444), mua-9(rh197), mnDf57, mnDf66, mnDf84, mnDf86, mnDf87, mnDf88, mnDf90, mnDf97, mnC1.

• LGIII: daf-2(e1368), dpy-17(e164), lon-1(e185), sma-3(e491), mup-4(ar60), ncl-1(e1865), unc-36(e251), sma-2(e502), unc-32(e189), glp-1(q46), ced-7(n1892), unc-69(e587), vab-7(e1562), tra-1(e1099), mua-2(rh119) qDp3.

• LGIV: lin-22(n372), dpy-13(e184sd), unc-5(e53), bli-6(sc16), mua-5(rh179), lin-45(n2018), egl-19(n582), unc-24(e138), him-8(e1489), dpy-20(e1282), unc-24(e138), unc-31(e928).

• LGV: dpy-11(e224), unc-23(e25), unc-42(e270), sma-1(e30), unc-39(ct73, e257), unc-112(st562, st581), ego-3(om40), unc-61(e228), unc-65(e351), sdc-3(y132), unc-76(e911), dpy-21(e428), yDf8, yDf9, yDf11, yDf12, arDf1, wDf1, nT1[unc-?(n754) let-?](IV;V), nT1[let-?(m435)].

• LGX: dpy-3(e27), lon-2(e670), mup-2(e2346ts), unc-6(e78), dpy-6(e14), unc-3(e151).

OT125: mua-9(rh197)/mnC1, OT117: mua-5(rh179) lin-45(n2018)/unc-5(e53) lin-45(n2018), and OT124: mua-6(rh85) were generated and provided by J. Plenefisch, and GS234: qDp3; ncl-1(e1865) unc-36(e251) glp-1(q46) mup-4(ar60) was generated and provided by E. Bucher.

Germline transformation and integration of extrachromosomal arrays:
The procedure for microinjection and germline transformation followed those of FIRE 1986 and MELLO et al. 1991 . The plasmid pH20 is a promoter trap construct in which a genomic fragment was inserted in the expression vector for green fluorescent protein (GFP), pPD95.75 (A. FIRE et al., personal communication). pH20 drives expression of GFP in almost all neurons (T. ISHIHARA and I. KATSURA, unpublished results; see also PAGE et al. 1997 ). pH20 (0.1 mg/ml) was mixed with the plasmid pBluescript KS(-) (0.3 mg/ml; Stratagene, La Jolla, CA) and injected into adult wild-type hermaphrodites. F₁ animals expressing GFP in neurons were picked onto separate plates, and several lines carrying pH20 as an extrachromosomal array were established. From the most intensely fluorescent line, integrated strains were generated by -ray irradiation with ⁶⁰Co source at a dose of 3600 rads. We screened 500 F₂ progeny of irradiated worms and obtained seven integrants. All the integrated lines were outcrossed at least five times to wild-type worms. None of the integrated lines exhibited any obvious abnormalities. ST2 (ncIs2) II and ST3 (ncIs3) III were used in this study.

Mutant screen:
Two types of screens were carried out. The type I screen was designed to isolate mutations on LGIV. ncIs2 II; dpy-13 +/+ unc-5 IV hermaphrodites were mutagenized with 0.05 M EMS (BRENNER 1974 ). Individual F₁ progeny were placed on separate plates at 20° and were allowed to reproduce. The F₂ progeny on the plates were screened for larval lethality caused by mutations on LGIV, as indicated by the decrease in the adult population of Dpy and/or Unc animals under a dissection microscope, and were then inspected for defects in neural morphology under an epifluorescence microscope (Axioplan with filter set #10, Zeiss, Thornwood, NY). Mutations were maintained by transferring wild-type hermaphrodites to new dishes.

In the type II screen, mutations on all linkage groups were screened. ncIs2 II; him-8 IV hermaphrodites were mutagenized as described above. F₁ progeny were plated singly and F₂ progeny were examined for abnormal neural morphology as described above. To maintain mutations with homozygous lethality, wild-type male siblings of the mutants were picked and mated to DA438 hermaphrodites of the genotype bli-4 I; rol-6 II; daf-2 vab-7 III; unc-31 IV; dpy-11 V; lon-2 X, and the wild-type progeny were picked individually. Wild-type hermaphrodites whose siblings exhibited the Ven phenotype were transferred individually.

In the type II screen, in addition to the larval lethal mutants with the abnormal ventral cord, we recovered five mutants with an increased number of neurons on the lateral body wall. Four of these mutations (nc31, nc32, nc33, nc34) failed to complement lin-22(n372) (WRISCHNIK and KENYON 1997 ). The other mutation, nc1, that caused the generation of many neurons on each side of the body in adult males and the decrease of four neurons in hermaphrodites was shown to be an allele of pry-1 (MALOOF et al. 1999 ).

Genetic mapping:
Three-factor analysis, STS mapping, deficiency mapping, and complementation test: Mutations recovered in the type II screen were mapped to a chromosome by identifying the genetic markers of DA438 remaining in the lines maintained as described in the previous section and/or by using polymorphic sequence-tagged sites (WILLIAMS et al. 1992 ). Some mutations were mapped by three-factor mapping or deficiency mapping using standard genetic techniques.

Complementation tests were carried out essentially as follows. For mutations a and b recovered in the type II screen, failure of complementation was indicated by the production of glowing F₁s (the cross progeny) exhibiting the Ven phenotype derived from mating between a/+; ncIs2 males and b/+ hermaphrodites, where a/+ and b/+ indicated trans-balanced mutations a and b, respectively.

To facilitate complementation tests of mutations on LGII, ncIs2 located in cis to the mutation was sometimes removed by recombination, and ncIs3 III was introduced to facilitate visualization of the nervous system.

mua-1 II (nc11, nc12, nc13): nc11, nc12, and nc13 were recovered in the type II screen and failed to complement mua-1(rh160).

ven-3 II (nc30): nc30 was recovered in the type II screen and was found to be linked to ncIs2 II. Three-factor mapping placed nc30 between dpy-10 and unc-53. The map position indicated that nc30 is not an allele of mua-1 or ven-2 on the same LG. nc30 complemented mua-9; ncIs2 nc30/+ + males were crossed to mua-9/mnC1 hermaphrodites, and no Ven mutants were found among the glowing F₁s.

ncIs2 + nc30 /+ dpy-10 +; him-8 males were crossed to unc-4 mnDf88/mnC1 dpy-10 unc-52 hermaphrodites, and about half of the glowing F₁s died during embryonic development. In a similar cross with unc-4 mnDf84/mnC1 dpy-10 unc-52 or unc-4 mnDf97/mnC1 dpy-10 unc-52 hermaphrodites, most of the glowing F₁s were normal except for a few dead embryos. As we observed a few glowing dead F₁ embryos even in the crosses between ncIs2 males and hermaphrodites carrying the deficiencies, we presumed that mnDf84 and mnDf97 both complemented nc30, whereas mnDf88 failed to complement nc30. nc30 complemented let-236 and let-253, which are the known let loci in the interval defined by the deficiencies; ncIs2 nc30/+ + males were crossed to let-236 unc-4/mnC1 dpy-10 unc-52, or let-253 unc-4/mnC1 dpy-10 unc-52 hermaphrodites, and no Ven mutants were found among the glowing F₁s in either cross.

ven-2 II (nc29): nc29 was recovered in the type II screen. nc29 was placed between stp98 (+1.84) and maP1 (+4.128) on LGII by STS mapping; of 23 mutants carrying stP36, 12 also had stP98 but not maP1, and 11 had maP1 but not stP98. nc29 was balanced with mnC1, and three-factor analysis indicated that the mutation is mapped to the right of unc-53 or linked tightly to unc-53.

nc29 complemented mua-1(nc11); the glowing F₁ progeny between ncIs2 + nc29/+ dpy-10 +; him-8 males and nc11 +/+ dpy-10 hermaphrodites were all non-Ven. nc29 complemented mua-9; males derived from the cross between + nc29 +/ dpy-10 + unc-53; ncIs3 hermaphrodites and N2 males were individually crossed to mua-9/mnC1 hermaphrodites, and glowing F₁ progeny were all wild-type.

nc29/mnC1 males were crossed to hermaphrodites with the following genotypes: unc-4 mnDf57/mnC1 dpy-10 unc-52, mnDf66/mnC1 dpy-10 unc-52, unc-4 mnDf86/mnC1 dpy-10 unc-52, unc-4 mnDf87/mnC1 dpy-10 unc-52, unc-4 mnDf90/mnC1 dpy-10 unc-52. About a quarter of glowing F₁s arrested as embryos with mnDf66 and mnDf90, but not with mnDf57, mnDf86, or mnDf87. These results indicated that neither mnDf66 nor mnDf90 complemented nc29. Deficiency mapping placed nc29 in the interval between +3.04 and +3.05. nc29 complemented mix-1, which is the only known let locus mapped to this interval; males of the genotype nc29/mnC1 dpy-10 unc-52; ncIs3 were crossed to hermaphrodites of the genotype unc-4 mix-1/mnC1 dpy-10 unc-52, and wild-type adult F₁ cross-progeny were picked individually. Some F₁s segregated wild-type worms and lethal Ven larvae, but no paralyzed Dpy larvae, i.e., mnC1 homozygotes, indicating that the genotype of these viable F₁s was + nc29 +/ unc-4 + mix-1.

mup-4 III (nc20, nc21, nc22, nc23, nc24, nc35): nc20 was recovered in the type I screen and was later shown to be linked to LGIII. The other mutations were recovered in the type II screen. They failed to complement each other, and nc20 failed to complement mup-4(ar60).

mua-5 IV (nc14, nc15, nc17, nc18, nc19): nc17, nc18, and nc19 were recovered in the type I screen, and nc14 and nc15 were isolated in the type II screen. They failed to complement each other, and nc18 failed to complement mua-5 (rh179).

ven-1 V (nc25, nc26, nc27, nc28): nc25 was recovered in the type I screen and was later shown not to be linked to LGIV, but to LGV. The other mutations were recovered in the type II screen. Three-factor mapping placed ven-1 between sma-1 and unc-76. nc26, nc27, and nc28 all failed to complement nc25; males of the genotype ncIs2; nc27/dpy-11 or ncIs2; nc28/+ were crossed to hermaphrodites of the genotype nc25/sma-1 unc-76, and Ven worms were found among glowing F₁s in either cross. ncIs2; nc26/dpy-11 males were crossed to nc25/nc25 hermaphrodites, and Ven worms were found among glowing F₁s.

For complementation tests of ven-1(nc25) with deficiencies, Df/+ hermaphrodites were crossed to ncIs2/+; ven-1(nc25)/+ males, and the presence of Ven worms was scored among glowing F₁s. Males of the genotype ncIs2/+; ven-1(nc25)/+ were crossed to hermaphrodites of the following genotypes, and no Ven worms were found among cross-progeny: nDf42 V/nT1[unc-?(n754) let-?] (IV;V), unc-42 arDf1 V/nT1[unc-?(n754) let-?] (IV;V), unc-42 yDf9 V/nT1[let-?(m435)] (IV;V), yDf11 V/nT1[let-?(m435)] (IV;V), unc-42 yDf12 V/nT1[unc-?(n754) let-?] (IV;V); dpy-6. Males of the genotype ncIs2/+; ven-1(nc25)/+ were crossed to hermaphrodites of the genotype yDf8 V/nT1[unc-?(n754) let-?] (IV;V), and Ven worms were found among glowing F₁s. There were viable escapers among ven-1/yDf8 worms.

unc-23, which is the only locus on LGV previously shown to exhibit the Mua phenotype (PLENEFISCH et al. 2000 ), has been mapped to the left of sma-1 and is distinct from ven-1. Complementation tests of ven-1(nc25) with closely located loci, unc-39, unc-112, unc-61, unc-65, sdc-3, and ego-3, were carried out by mating ncIs2; him-8/+; nc25/+ males with the following hermaphrodites, and no Ven worms were found among glowing F₁s: unc-39(ct73); unc-61; unc-65; unc-112(st562)/dpy-11 unc-23; unc-112(st581)/unc-39(e257); unc-76 wDf1/unc-61 dpy-21; sdc-3/unc-76; ego-3 unc-76/nT1[unc-?(n754) let-?].

mua-6 X (nc16): nc16 was recovered in the type II screen and failed to complement mua-6 (rh85).

Histochemical procedures:
Indirect immunofluorescence histochemistry was used to stain worms for GABA following the procedure described by MCINTIRE et al. 1992 . As primary and secondary antibodies, rabbit anti-GABA antiserum (Sigma, St. Louis) and Texas red-conjugated donkey anti-rabbit antiserum (Sigma) were used, respectively. 4',6-diamidino-2-phenylindole (1 mg/ml) was included in the solution for the secondary antibody. Stained worms were mounted with PBS (10.4 mM Na₂PO₄, 1.8 mM KH₂PO₄, 137 mM NaCl, 4.4 mM KCl, pH 7.2) containing 90% glycerol and 1,4-diazobicyclo-2,2-octane (DABCO, 0.6 mg/ml) and observed with an epifluorescence microscope (Axioplan, Zeiss) using Zeiss filter set #15.

To visualize muscles, worms of mixed stages were gathered in 1% bovine serum albumin, fixed with 4% paraformaldehyde in PBS for 24 hr at 4°, and partially broken with a microhomogenizer (Wheaton) in PBS. Then, the worms were incubated in rhodamine-conjugated phalloidin (phalloidin-RITC, 0.02 units/µl, Wako) in Tris-Cl buffer (10 mM, pH 7.4) containing 150 mM NaCl, 1% Triton X-100, 1% Tween 20, and 3% skim milk for 24 hr. The worms were washed with PBS containing 1% Tween 20 and observed with the epifluorescence microscope as described above.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Screen for neuroanatomical larval lethal mutants:
To facilitate the identification of mutants with abnormal neural morphology, ST2(ncIs2), an integrant strain expressing GFP in almost the entire nervous system, was mutagenized and screened. We focused our attention mainly on the morphology of the ventral cord, one of the major nerve bundles in the worms. Mutants that arrested during larval stages with severely disorganized ventral nerve cords (Ven phenotype) were isolated (Fig 1). In the type I screen targeting mutations mainly on LGIV, 5 mutations (nc17, nc18, nc19, nc20, and nc25) were recovered from 3500 F₁ colonies. In the type II screen targeting mutations on the entire LGs, 16 mutations (nc11, nc12, nc13, nc14, nc15, nc16, nc21, nc22, nc23, nc24, nc26, nc27, nc28, nc29, nc30, and nc35) were recovered among 2000 F₁ colonies.

View larger version (95K): In this window In a new window Download PPT slide   Figure 1. Disorganized nerves of mutants. In worms carrying ncIs2 or ncIs3, the entire nervous system was visualized by the expression of GFP. Lateral views of wild-type (A) and mutant (B–J) larvae are shown. The ventral side is to the bottom. The mutants are (B) mua-1(nc11), (C) mup-4(nc20), (D) mua-5(nc19), (E) mua-6(nc16), (F) ven-2(nc29), (G) ven-1(nc25), (H) ven-3(nc30), (I) mup-1(e2347), and (J) mup-2(e2346ts). In the wild-type L1 worm (A), the ventral cord is positioned along the ventral midline, and somata of the motoneurons are embedded in the cord. In mutants (B–J), ventral cords are shifted laterally. (K) In an adult ven-1(nc25) mutant hermaphrodite, part of the ventral cord is displaced laterally (asterisk), but remains attached to the body wall. (L) In an adult ven-1(nc25) mutant, small nerves are also disorganized (arrows). Bars in A–J and L are 50 µm; bar in K is 20 µm.

Mapping of mutations conferring Ven phenotype:
All mutations were shown to be recessive to wild type. The mutations were mapped by classical mapping methods and sometimes by STS mapping (see MATERIALS AND METHODS for details). The results of genetic mapping are summarized in Table 1.

The mutations were classified into seven complementation groups on all LGs except LGI as summarized in Table 2 and Fig 2. Six mutations defined three novel loci, which we named ven-1, ven-2, and ven-3. Four mutations for ven-1 and one each for ven-2 and ven-3 were recovered. Fifteen of the mutations were assigned to four previously defined loci: three alleles of mua-1; six alleles of mup-4; five alleles of mua-5; one allele of mua-6. mua and mup loci were originally defined by the mutant phenotypes, fragile muscle attachments (PLENEFISCH et al. 2000 ) and muscle positioning abnormal (GATEWOOD and BUCHER 1997 ), respectively.

View larger version (17K): In this window In a new window Download PPT slide   Figure 2. Genetic map positions of the genes. Each linkage group is depicted in a horizontal line. The genes identified in the present screen are shown above each chromosome. Previously reported loci exhibiting the Mua/Mup defects are shown below the line (GOH and BOGAERT 1991 ; MYERS et al. 1996 ; GATEWOOD and BUCHER 1997 ; PLENEFISCH et al. 2000 ; J. PLENEFISCH, personal communication). Genetic markers used for mapping are shown below in slanted positions. For LGV, the positions of deficiencies are also shown below. For LGII and LGIII, the central region is enlarged and shown on the right.

As the mutations included alleles of four of the mua/mup genes, we were interested in whether previously isolated mua/mup mutants also exhibit the Ven phenotype. First, we examined the nervous system of worms with an existing allele of mua-1(rh160) in the ncIs3 genetic background. The rh160 ventral cord was defective, similar to that in nc11, nc12, and nc13 (data not shown). Next, we examined the mutants for other mua/mup loci that were not isolated in the present screen. By electron microscopic examination, GOH and BOGAERT 1991 reported that the nervous system of mup-1(e2347) was disorganized. We introduced ncIs3 into mup-1(e2347) and found that the ventral cord was detached from the body wall (Fig 1I) in living worms. We also found the Ven defect in mup-2(e2346ts) in the ncIs3 genetic background raised at 20° (Fig 1J). In the ncIs3 genetic background, mua-2(rh119), unc-23(e25), and vab-10(e698), which have been shown to have the Mua defects (PLENEFISCH et al. 2000 ), also exhibited the Ven defect (data not shown). Therefore, the Ven phenotype seems common to mua/mup mutants.

Phenotypes of mutants:
Disorganized nerve cords, distorted alae, meandering excretory canals: In general, Ven mutants arrested at larval stages (Table 3). The arrested larvae seldom moved and often lay in the shape of a coil or circle on agar plates. Viewed under a fluorescence microscope, the ventral cord of mutant worms was abnormal (Fig 1; Table 3). The dorsal cord, another major nerve tract of worms, was also abnormal. Part of the cords was often displaced laterally. In some cases, disorganization of the nerve structure including the defasciculation and/or displacement of cell bodies of neurons was observed. We sometimes observed that the displaced nerves were floating free in the body when worms moved and bent the body, indicating that the cords were detached from the body wall. Some defasciculations appeared to be caused by detachment of parts of nerve bundles from the cords. In worms surviving to late larval stages, the arrangement of the nervous system on the lateral body wall was also disorganized. Some mutant worms moved fairly well, despite the disorganization of their ventral cords.

We noticed that tissues other than nerves were also affected in the mutants. The excretory canal is a cellular process of the excretory cell extending longitudinally along the body wall. Alae are special cuticular structures generated by the seam cells, a row of hypodermal cells constituting the lateral body wall. Inspection with Nomarski optics revealed that the morphology of these structures was sometimes abnormal in Ven mutants (Fig 3).

View larger version (167K): In this window In a new window Download PPT slide   Figure 3. Abnormal excretory canal and hypodermis of mutants. The excretory canals (A and B) and alae (C and D) were visualized under Nomarski optics. (A) In a wild-type L2 larva, the excretory canal (arrows) runs straight longitudinally on the lateral body wall. (B) In a mua-5(nc18) L2 larva, the canal meanders (arrows) and sometimes makes a loop (arrowhead). (C) The typical appearance of the alae (arrows) in a wild-type adult worm is shown. (D) The alae in a ven-1(nc25) adult worm. The alae are bifurcated (arrowhead) and extended into an abnormal position (arrows). Bars in A and B are 10 µm; bars in C and D are 25 µm.

ven-1(nc25, nc26, nc27, nc28): All alleles exhibited similar phenotypes (Table 3). Many worms homozygous for these mutations did not grow beyond the size of L1/L2 wild-type larvae. There were also escapers that reached adulthood and laid eggs, and the strains could be maintained as homozygotes. Some escapers had a normal ventral cord. In most of the arrested larvae, the ventral cord and/or the dorsal cord was defective (Fig 1G). Some had floating ventral cords. In others, the ventral cord remained attached to the body wall but was misrouted (Fig 1K and Fig L) or defasciculated. Sometimes, the ventral cord terminated midway on the ventral midline and shifted laterally and then extended straight on longitudinally. Alae were sometimes in abnormal positions or were disorganized. The excretory canal sometimes stalled or was misrouted. The gonadal morphology was sometimes abnormal.

ven-2 (nc29): Most of the mutant worms were arrested at a fixed stage, which appeared to correspond to L2 as judged by the body size and shape of the gonad. There were no adults (Table 3). Unlike the mutants of the other loci, nc29 mutants did not arrest in the shape of a coil or circle, and some mutants moved rather well. The ventral cord was displaced and was usually bifurcated along the posterior body, but was often normal along the anterior body (Fig 1F). Sometimes, four to five neuronal somata were aggregated at the branching point of bifurcation. No defect was observed in the dorsal cord (Table 4). nc29 mutants had ALMR/L and BDUR/L, but the somata of ALMs were sometimes slightly misplaced dorsally or ventrally, and the process of ALMs was sometimes bent. The mutants lost the other neurons on the lateral body wall including all postdeirid neurons. The excretory canal often meandered, and body shape was sometimes abnormal.

More than 70% of nc29/Df worms were arrested as embryos (see MATERIALS AND METHODS), suggesting that nc29 is not a null allele. More than 80% of embryos were arrested as amorphous cell clumps and could not be staged. For the rest of the embryos, the arrested stage ranged from the comma stage to late embryos.

ven-3 (nc30): Many mutants were arrested at the L1 stage, but some were arrested as embryos (mostly at the threefold stage) and few survived to adulthood (Table 3). In the arrested larvae, the ventral cord or dorsal cord was often disorganized and detached from the body wall (Fig 1H). The excretory canal sometimes meandered, and alae were sometimes in abnormal positions or bifurcated. The intestine was sometimes dilated.

About 70% of nc30/Df worms were arrested as embryos (see MATERIALS AND METHODS), suggesting that nc30 is not a null allele. More than 70% of embryos were arrested as amorphous cell clumps and could not be staged. For the rest of the embryos, the arrested stage ranged from the comma stage to late embryos.

mua-1(nc11, nc12, nc13): Of the three alleles, nc12 was the strongest (Table 3). Worms homozygous for the mutation were arrested at various larval stages, beginning from the L2 stage. No arrested embryos were observed (Table 4). Some escapers reached adulthood and laid eggs. In mutant worms, part of or the entire ventral cord was sometimes displaced to the lateral body wall (Fig 1B). Motoneurons of the cord were sometimes displaced with the cord. The ventral nerve cord was sometimes defasciculated. In contrast to this, no abnormal dorsal cord was observed (Table 4). The mutant larvae also exhibited defects in tissues other than the nervous system: the excretory canal often meandered; alae were sometimes distorted, misplaced, or missing locally; the gonadal morphology was sometimes abnormal.

mup-4 (nc20, nc21, nc22, nc23, nc24, nc35): nc22 was the strongest allele and nc23 was the weakest allele (Table 3). Worms homozygous for strong alleles were arrested as embryos or L1 and did not move, remaining in a folded position like a "pretzel," as reported previously for the other alleles of the gene (GATEWOOD and BUCHER 1997 ). In strong alleles, the ventral or dorsal nerve cords were detached from the body wall, often along the entire length (Fig 1C). In nc23, there were some adult escapers, which had no Ven defect and laid eggs.

mua-5 (nc14, nc15, nc17, nc18, nc19): All the alleles were of similar strength, with nc14 appearing slightly stronger than the others (Table 3). Most of the worms were arrested at early larval stages in the shape of a coil. Few survived to the later larval stages. We have observed, although very rarely (<1%), that nc19 worms survived, growing into dumpy adults without nerve defects, and laid eggs.

Ventral cords and dorsal cords were often displaced onto the lateral body wall, sometimes along the entire length, and were sometimes defasciculated. Motoneurons in the ventral midline were often missing (Fig 1D). Cell bodies expressing GFP were often found in the displaced ventral cord, suggesting that motoneurons of ventral cords are mispositioned together with the cord. To test this possibility, we stained worms with an anti-GABA antibody (Fig 4). GABAergic neurons on the ventral midline were often missing and ectopic GABAergic neurons were present in the mispositioned ventral cord, indicating that some of the laterally placed neurons were GABAergic motoneurons.

View larger version (23K): In this window In a new window Download PPT slide   Figure 4. Immunofluorescence pictures of neurons stained with the anti-GABA antibody. (A) In the wild-type worm, the GABAergic fibers (large arrow) and GABAergic motoneurons (small arrow) in the ventral cord are located on the ventral midline. (B) In a mua-5(nc18) worm, GABAergic fibers (large arrow) are displaced and somata of GABAergic neurons are found on the lateral body wall (small arrow).

The sensory processes in the head were sometimes disorganized, and the excretory canal often meandered (Fig 3B).

mua-6 (nc16): Most mutant worms were arrested at early larval stages in the shape of a circle or rod and did not move. The ventral cord was often detached from the body wall (Fig 1E). The gonad was sometimes distorted or mispositioned. The excretory canal sometimes meandered.

Attachment of body wall muscles is affected in Ven mutants:
Genetic mapping revealed that some of the mutations are alleles of mua/mup genes that have been previously identified on the basis of defects in attachment or in the position of body wall muscle. To determine the relationships between the Ven and Mua/Mup phenotypes, we examined whether the Ven mutants isolated in the present study exhibited the Mua/Mup phenotype. In wild-type worms, laterally located body wall muscle cells form four longitudinal bands; two in the dorsal and two in the ventral quadrant of the body. Phalloidin-RITC stained them as two parallel bands when viewed laterally (Fig 5A). All mutants for the newly identified loci ven-1, ven-2, and ven-3 showed the Mua/Mup phenotype when stained with phalloidin-RITC. That is, part of the ventral body wall muscle was shifted dorsally as a strip, sometimes to the position of the dorsal body wall, or part of the dorsal body wall muscle was sometimes shifted ventrally (Fig 5C, Fig D, and Fig E). They often appeared to be detached from the body wall. The phalloidin-RITC staining also revealed that mutants for the newly isolated alleles of mua-1, mua-5, mua-6, and mup-4 exhibited the Mua/Mup phenotype (Fig 5B; data not shown).

View larger version (56K): In this window In a new window Download PPT slide   Figure 5. Muscle defects of the mutants. The displacement of the ventral nerve cord coincided with that of body wall muscles in mutants. (A–E) Muscles are stained with phalloidin-RITC. (A) Lateral view of a wild-type worm showing two longitudinal bands of the body wall muscles in the dorsal and ventral quadrants of the body. In mutants, part of the body wall muscles is detached from the body wall and shifted dorsally. (B–E) In mutants, the body wall muscles are displaced (arrows). (B) mua-1(nc13), (C) ven-1(nc25), (D) ven-2(nc29), and (E) ven-3(nc30). (F–J) The nerves are visualized in the same specimen as shown in (E–J). Arrows indicate the displacement of nerve cords.

To examine how the neural and muscular defects are correlated, mutant worms carrying the neuronal GFP marker, ncIs2 or ncIs3, were stained with phalloidin-RITC. In many mutants, whenever the ventral cord was displaced, the body wall muscle at the same position was displaced. The extent and direction of the displacement were similar in both tissues, suggesting that the defects are associated (Fig 5, G–J; data not shown). To confirm this quantitatively, we scored the displacement of the nerve cords and body wall muscles on the dorsal and the ventral side of the body in individual worms (Table 4). The results showed a high correlation between the displacement of nerve cords and the displacement of muscles on the same side of the body.

The analysis also revealed that the penetrance for the defects of tissues on the ventral side of the body and that on the dorsal side of the body is different and that the difference is characteristic for each gene (Table 4). The dorsal cords and the dorsal body wall muscles were often displaced in ven-3 and mua-5 worms, whereas those tissues were rarely affected in ven-2, mua-1, and mua-6 worms. Conversely, the penetrance of the displacement of the ventral tissues was high in ven-2, mua-1, and mua-6 worms, whereas that in ven-3 and mua-5 worms was relatively low. In ven-1 and mup-4 worms, the ventral tissues were affected more frequently than were the dorsal tissues.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Screening with a pan-neural marker revealed mutants with a novel neural phenotype,
Ven: Using a strain expressing a pan-neuronal GFP marker, we identified a novel mutant phenotype, Ven, in which nerve cords are displaced or detached from the body wall. In C. elegans, many mutations causing neuroanatomical defects in the nervous system have been reported, most of which affect a particular process of neural development, such as generation of neurons, establishment of neuronal subtype, or axonal growth, and affect a particular subset of neurons (for review see ANTEBI et al. 1997 ; RUVKUN 1997 ). In contrast, in Ven mutants, the nervous system is grossly affected, and the neural defect appears irrespective of neuronal type.

Our genetic screen through <5000 haploid genomes identified 21 mutations with the Ven phenotype in seven genetic loci including three novel loci. The screen also enabled us to isolate alleles of lin-22 and pry-1 (see MATERIALS AND METHODS), which had an increased number of lateral body wall neurons, indicating that the strain expressing a pan-neuronal GFP marker is useful for screening of gross morphological abnormalities of the nervous system on the lateral body wall as well as in the nerve cords.

Three ven loci were identified:
We have identified three novel genetic loci defined by the Ven phenotype. Although ven-1 and ven-2 mutants were defective in muscular attachment and were larval lethal, they were somewhat different from the previously isolated mua mutants; mua mutants have been identified based on the phenotype of progressive loss of motility during larval development (PLENEFISCH et al. 2000 ). In contrast, ven-2 mutants arrested at a fixed larval stage, and ven-1 mutants had many escapers that showed very slight locomotory defects. Whereas mua mutations were suppressed by unc-54 mutations, which affect the contraction of skeletal muscles (PLENEFISCH et al. 2000 ), our preliminary experiments showed that unc-54 mutations do not suppress ven-1 mutations (G. SHIOI and S. TAKAGI, unpublished results). These observations suggested that the primary targets of the mutations of ven-1 and ven-2 genes may not be muscular attachment, whereas it was proposed that the mua mutations primarily affect muscular attachment.

Genetic analysis using deficiencies suggested that ven-2(nc29) and ven-3(nc30) are not null alleles; the worms with the mutations over the deficiencies arrested at embryonic stages. Mutations in pat genes cause embryonic arrest at the twofold stage (WILLIAMS and WATERSTON 1994 ), and pat mutants have disorganized muscles and develop muscle position defects after arrest. Though some ven-2/Df and ven-3/Df worms arrested at the twofold stage, many arrested at earlier stages. Thus, although the null phenotype of ven-2 and ven-3 remains to be determined, these genes are probably distinct from pat genes. ven-2(nc29) was mapped very close to mix-1, and mutants of these genes shared many phenotypes (LIEB et al. 1998 ). It is, however, unlikely that nc29 is allelic to mix-1, because nc29 complemented mix-1(mn29).

Mutations in mua/mup genes confer the Ven phenotype:
Four of the loci responsible for the Ven phenotype proved identical to the previously identified mua/mup loci, mua-1, mua-5, mua-6, and mup-4, which were defined originally by abnormal muscle attachment or muscle positioning of the mutants (GATEWOOD and BUCHER 1997 ; PLENEFISCH et al. 2000 ). By staining muscles, we showed that all the mutants recovered in the present screen for the Ven defect exhibited muscle defects defined as Mua or Mup. Using the pan-neuronal marker, we have also shown that mup-1, mup-2, mua-2, unc-23, and vab-10 mutants have disorganized nerve cords. Thus, although neural defects in mua/mup mutants have not been examined extensively, the Ven phenotype appears common to mua/mup mutants. The relatively small number of genomes screened and the failure of recovering mutations in some mua/mup genes indicated that the present screen was insufficient to identify all the genes in the class. Probably, more ven or mua/mup genes remain to be discovered.

ven and mua/mup mutations affect multiple tissues on the body wall:
In mutants exhibiting the Ven phenotype, we demonstrated morphological defects in tissues other than nerves or muscles: the excretory canal was sometimes abnormal in most of the mutants isolated in this screen. Abnormal alae similar to those reported for mup-2 mutants (MYERS et al. 1996 ) were sometimes observed in mua-1, ven-1, and ven-3 mutants. Thus, it seems a common feature of ven and mua/mup mutations that the terminal phenotypes are expressed both in tissues composing and in those attached to the body wall.

Ven phenotype may be caused by defects in muscles, hypodermis, or basal laminae:
How the mutations cause the ventral cord to detach from the body wall remains to be determined. Mutations might affect molecular components directly involved in the attachment of nerves, or they might affect the integrity of the tissues involved in attachment of nerves.

By double labeling of the nervous system and muscles, we showed the spatial coincidence of the defects in these two tissues. This suggested that either the defects are consequences of common regional defects or the defects of the nerve and of the muscle have a causal relationship. One possibility is that detachment of the muscle leads to detachment of the nerves. The Ven defect caused by a mutation in mup-2, which encodes the muscle contractile protein troponin T (MYERS et al. 1996 ), appears to be a secondary consequence of the muscle defect.

Alternatively, nerves and muscles may share cellular mechanisms for attachment that are defined by some of the ven and mua/mup genes. The ventral cord runs in apposition to the thickening of the ventral hypodermis called the ventral ridge, and both are surrounded by basal laminae (WHITE et al. 1976 ). The muscle is also attached to the hypodermis via the basal laminae. Therefore, defects in the hypodermis and/or basal laminae may lead directly to displacement of both nerves and muscles. Previous studies indicated that the hypodermis is the primary target for mup-4 and mua-3 mutations (BUCHER and GREENWALD 1991 ; GATEWOOD and BUCHER 1997 ).

Furthermore, the nerves and muscles may use common molecules for attachment to the body wall. Previous studies on mua/mup mutants failed to detect abnormalities in the known components of the muscle attachment complex, and the molecular mechanisms responsible for Mua defects remain to be determined (GATEWOOD and BUCHER 1997 ; PLENEFISCH et al. 2000 ). Although the morphological features of the attachment differ between the muscles and the nerves, it is possible that ven and mua/mup genes encode some unknown molecules expressed in the hypodermis or basal laminae necessary to attach both tissues to the body wall.

We have shown that the excretory canal was sometimes affected in Ven mutants. Similar to nerves, the excretory canal is in contact with hypodermis and the basal lamina, which are known to play important roles in the guidance of migrating cells, and mutations affecting canal elongation are also known to affect migration of axons and cell bodies of neurons (HEDGECOCK et al. 1987 , HEDGECOCK et al. 1990 ; MANSER et al. 1997 ; TAKAGI et al. 1997 ). Interestingly, the abnormal excretory canal meandered but was not detached from the body wall. It is also noted that in some specimens of ven-1 mutants, the nerves were displaced but remained attached to the body wall. These observations suggest that the genes may also be required for the proper guidance of cell migration.

Attachments were differentially affected between the dorsal side and the ventral side of the body:
We have revealed gene-specific differences in the penetrance of displacement defects between the ventral tissues and the dorsal tissues. The observed differences might imply that some genes, such as mua-1 and ven-1, function mainly on the ventral side of the body, whereas others, such as ven-2, function mainly on the dorsal side of the body. Alternatively, the differences might reflect the developmental period when the function of particular genes would be required; during the embryo elongation stage, the body shape of C. elegans shows drastic changes from a ball of cells into its final worm shape. As an embryo lies with its dorsal side out during the early phase of the elongation, dorsal nerves and muscles are on the concave surface of the body wall. They would be more vulnerable to the mechanical stress in the longitudinal direction caused by the elongation, compared with the tissues on the convex surface of the ventral side of the body. Therefore, slight attachment defects would lead to more severe consequences for the dorsal tissues than for the ventral tissues during the elongation period. More detailed phenotypic analysis of mutant embryos would be necessary to elucidate what causes these differences. Analysis on the spatio-temporal requirements of the genes would also be informative.

In summary, our study revealed the pleiotropic nature of the defects in ven and mua/mup mutants. ven and mua/mup mutants would provide a unique opportunity to study the genetic cascade underlying cellular interactions required for the establishment and maintenance of the proper positioning of many tissues. Developmental analyses and mosaic analyses are required to clarify the cellular mechanisms of the detachment of tissues for each mutant. Molecular analysis of the relevant genes is beginning to give a partial image of this phenomenon. mup-4 and mua-3, which have been suggested to encode large transmembrane proteins with EGF motifs and low-density lipoprotein-like motifs (GATEWOOD and BUCHER 1997 ; J. PLENEFISCH, personal communication), may be involved in the cell-extracellular matrix interaction. Future molecular analyses of ven and mua/mup genes would be a promising approach for understanding this phenomenon.

	ACKNOWLEDGMENTS

We thank John Plenefisch for communicating unpublished observations prior to publication and Siegfried Hekimi for comments on the manuscript. We also thank John Plenefisch, Elizabeth Bucher, and Ryuji Hosono for strains; Naoko Inoue for her work in the early phase of the screen; Andrew Fire for pPD95.75; and past and present members of our laboratories for discussion and advice throughout this work. Some strains were provided by the Caenorhabditis Genetic Center, which is funded by the National Institute for Health National Center for Research Resources. This work was supported by grants from the Ministry of Education, Science and Culture, Japan (H.F., S.T.) and Core Research for Evolution Science and Technology (CREST) of Japan Science and Technology Corporation (JST) (H.F.).

Manuscript received June 5, 2000; Accepted for publication December 19, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ANTEBI, A., C. R. NORRIS, E. M. HEDGECOCK and G. GARRIGA, 1997 Cell and growth cone migrations, pp. 583–610 in C. elegans II, edited by D. R. RIDDLE, T. BLUMENTHAL, B. J. MAYER and J. R. PRIESS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

BRENNER, S., 1974  The genetics of Caenorhabditis elegans.. Genetics 77:71-94[Abstract/Free Full Text].

BUCHER, E. A. and I. GREENWALD, 1991  A genetic mosaic screen of essential zygotic genes in Caenorhabditis elegans.. Genetics 128:281-292[Abstract].

CHALFIE, M. and J. SULSTON, 1981  Developmental genetics of the mechanosensory neurons of Caenorhabditis elegans.. Dev. Biol. 82:358-370[Medline].

FIRE, A., 1986  Integrative transformation of Caenorhabditis elegans.. EMBO J. 5:2673-2680[Medline].

FRANCIS, R. and R. H. WATERSTON, 1991  Muscle cell attachment in Caenorhabditis elegans.. J. Cell Biol. 114:465-479[Abstract/Free Full Text].

GATEWOOD, B. K. and E. A. BUCHER, 1997  The mup-4 locus in Caenorhabditis elegans is essential for hypodermal integrity, organismal morphogenesis and embryonic body wall muscle position. Genetics 146:165-183[Abstract].

GOH, P. Y. and T. BOGAERT, 1991  Positioning and maintenance of embryonic body wall muscle attachments in C. elegans requires the mup-1 gene. Development 111:667-681[Abstract].

HEDGECOCK, E. M., J. G. CULOTTI, D. H. HALL, and B. D. STERN, 1987  Genetics of cell and axon migrations in Caenorhabditis elegans.. Development 100:365-382[Abstract].

HEDGECOCK, E. M., J. G. CULOTTI, and D. H. HALL, 1990  The unc-5, unc-6, and unc-40 genes guide circumferential migrations of pioneer axons and mesodermal cells on the epidermis in C. elegans. Neuron 2:61-85.

HRESKO, M. C., B. D. WILLIAMS, and R. H. WATERSTON, 1994  Assembly of body wall muscle and muscle cell attachment structures in Caenorhabditis elegans.. J. Cell Biol. 124:491-506[Abstract/Free Full Text].

LIEB, J. D., M. R. ALBRECHT, P. T. CHUANG, and B. J. MEYER, 1998  MIX-1: an essential component of the C. elegans mitotic machinery executes X chromosome dosage compensation. Cell 92:265-277[Medline].

MALOOF, J. N., J. WHANGBO, J. M. HARRIS, G. D. JONGEWARD, and C. KENYON, 1999  A Wnt signaling pathway controls hox gene expression and neuroblast migration in C. elegans.. Development 126:37-49[Abstract].

MANSER, J., C. ROONPRAPUNT, and B. MARGOLIS, 1997  C. elegans cell migration gene mig-10 shares similarities with a family of SH2 domain proteins and acts cell nonautonomously in excretory canal development. Dev. Biol. 184:150-164[Medline].

MCINTIRE, S. L., G. GARRIGA, J. WHITE, D. JACOBSON, and H. R. HORVITZ, 1992  Genes necessary for directed axonal elongation or fasciculation in C. elegans.. Neuron 8:307-322[Medline].

MELLO, C. C., J. M. KRAMER, D. STINCHCOMB, and V. AMBROS, 1991  Efficient gene transfer in C.elegans: extrachromosomal maintenance and integration of transforming sequences. EMBO J. 10:3959-3970[Medline].

MOERMAN, D. G., and A. FIRE, 1997 Muscle: structure, function, and development, pp. 417–470 in C. elegans II, edited by D. R. RIDDLE, T. BLUMENTHAL, B. J. MAYER and J. R. PRIESS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

MYERS, C. D., P. Y. GOH, T. S. ALLEN, E. A. BUCHER, and T. BOGAERT, 1996  Developmental genetic analysis of troponin T mutations in striated and nonstriated muscle cells of Caenorhabditis elegans.. J. Cell Biol. 132:1061-1077[Abstract/Free Full Text].

PAGE, B. D., W. ZHANG, K. STEWARD, T. BLUMENTHAL, and J. R. PRIESS, 1997  ELT-1, a GATA-like transcription factor, is required for epidermal cell fates in Caenorhabditis elegans embryos. Genes Dev. 11:1651-1661[Abstract/Free Full Text].

PLENEFISCH, J. D., X. ZHU, and E. M. HEDGECOCK, 2000  Fragile skeletal muscle attachments in dystrophic mutants of Caenorhabditis elegans: isolation and characterization of the mua genes. Development 127:1197-1207[Abstract].

RUVKUN, G., 1997 Patterning the nervous system, pp. 543–582 in C. elegans II, edited by D. R. RIDDLE, T. BLUMENTHAL, B. J. MAYER and J. R. PRIESS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

TAKAGI, S., C. BENARD, J. PAK, D. LIVINGSTONE, and S. HEKIMI, 1997  Cellular and axonal migrations are misguided along both body axes in the maternal-effect mau-2 mutants of Caenorhabditis elegans.. Development 124:5115-5126[Abstract].

WHITE, J. G., E. SOUTHGATE, J. N. THOMSON, and S. BRENNER, 1976  The structure of the ventral nerve cord of Caenorhabditis elegans.. Philos. Trans. R. Soc. Lond. B Biol. Sci. 275:327-348[Medline].

WILLIAMS, B. D. and R. H. WATERSTON, 1994  Genes critical for muscle development and function in Caenorhabditis elegans identified through lethal mutations. J. Cell Biol. 124:475-490[Abstract/Free Full Text].

WILLIAMS, B. D., B. SCHRANK, C. HUYNH, R. SHOWNKEEN, and R. H. WATERSTON, 1992  A genetic mapping system in Caenorhabditis elegans based on polymorphic sequence-tagged sites. Genetics 131:609-624[Abstract].

WRISCHNIK, L. A. and C. J. KENYON, 1997  The role of lin-22, a hairy/enhancer of split homolog, in patterning the peripheral nervous system of C. elegans.. Development 124:2875-2888[Abstract].

This article has been cited by other articles:

				H. Ohta, M. Fujiwara, Y. Ohshima, and T. Ishihara ADBP-1 Regulates an ADAR RNA-Editing Enzyme to Antagonize RNA-Interference-Mediated Gene Silencing in Caenorhabditis elegans Genetics, October 1, 2008; 180(2): 785 - 796. [Abstract] [Full Text] [PDF]

				W.-M. Woo, E. C. Berry, M. L. Hudson, R. E. Swale, A. Goncharov, and A. D. Chisholm The C. elegans F-spondin family protein SPON-1 maintains cell adhesion in neural and non-neural tissues Development, August 15, 2008; 135(16): 2747 - 2756. [Abstract] [Full Text] [PDF]

				L. Jia and S. W. Emmons Genes That Control Ray Sensory Neuron Axon Development in the Caenorhabditis elegans Male Genetics, July 1, 2006; 173(3): 1241 - 1258. [Abstract] [Full Text] [PDF]

				H. C. Korswagen, D. Y.M. Coudreuse, M. C. Betist, S. van de Water, D. Zivkovic, and H. C. Clevers The Axin-like protein PRY-1 is a negative regulator of a canonical Wnt pathway in C. elegans Genes & Dev., May 15, 2002; 16(10): 1291 - 1302. [Abstract] [Full Text] [PDF]