Genetics, Vol. 163, 905-915, March 2003, Copyright © 2003

DIM-1, a Novel Immunoglobulin Superfamily Protein in Caenorhabditis elegans, Is Necessary for Maintaining Bodywall Muscle Integrity

Teresa M. Rogalskia, Mary M. Gilberta, Danelle Devenporta, Kenneth R. Normana, and Donald G. Moermana
a Department of Zoology, University of British Columbia, Vancouver, British Columbia V6T 1Z4, Canada

Corresponding author: Donald G. Moerman, University of British Columbia, 6270 University Blvd., Vancouver, British Columbia V6T 1Z4, Canada., moerman{at}zoology.ubc.ca (E-mail)

Communicating editor: P. ANDERSON


*  ABSTRACT
*TOP
 *ABSTRACT
*MATERIALS AND METHODS
 *RESULTS
 *DISCUSSION
 *LITERATURE CITED

The UNC-112 protein is required during initial muscle assembly in C. elegans to form dense bodies and M-lines. Loss of this protein results in arrest at the twofold stage of embryogenesis. In contrast, a missense mutation in unc-112 results in viable animals that have disorganized bodywall muscle and are paralyzed as adults. Loss or reduction of dim-1 gene function can suppress the severe muscle disruption and paralysis exhibited by these mutant hermaphrodites. The overall muscle structure in hermaphrodites lacking a functional dim-1 gene is slightly disorganized, and the myofilament lattice is not as strongly anchored to the muscle cell membrane as it is in wild-type muscle. The dim-1 gene encodes two polypeptides that contain three Ig-like repeats. The short DIM-1 protein isoform consists entirely of three Ig repeats and is sufficient for wild-type bodywall muscle structure and stability. DIM-1(S) localizes to the region of the muscle cell membrane around and between the dense bodies, which are the structures that anchor the actin filaments and may play a role in stabilizing the thin rather than the thick filament components of the sarcomere.

BODYWALL muscle in Caenorhabditis elegans is an excellent in vivo system in which to study the assembly and maintenance of integrin-containing adhesion structures and to identify components that are involved in these processes. In C. elegans the myofilament lattice is anchored to the muscle cell membrane and the adjacent basement membrane by dense bodies and M-lines. The cell extracellular matrix adhesion sites of both dense bodies and M-lines are remarkably similar to mammalian focal adhesion plaques (WATERSTON 1988 Down; FRANCIS and WATERSTON 1991 Down; MOERMAN and FIRE 1997 Down; HRESKO et al. 1999 Down). The nematode and mammalian structures contain many of the same components (BURRIDGE and CHRZANOWSKA-WODNICKA 1996 Down), and, in both cases, integrin-ECM interactions are required to initiate assembly and to stabilize existing adhesion complexes (HRESKO et al. 1994 Down; YAMADA and GEIGER 1997 Down). Genetic analysis of these focal adhesion analogs in C. elegans has identified three proteins that affect integrin organization, UNC-52/perlecan, UNC-112, and PAT-4/ILK (ROGALSKI et al. 1993 Down; HRESKO et al. 1994 Down; WILLIAMS and WATERSTON 1994 Down; ROGALSKI et al. 2000 Down; MACKINNON et al. 2002 Down).

In C. elegans ßPAT-3/{alpha}PAT-2 integrin heterodimers link both the dense body and the M-line components to the underlying basement membrane (FRANCIS and WATERSTON 1985 Down; GETTNER et al. 1995 Down; B. WILLIAMS, personal communication). The UNC-112 protein affects the organization of these integrin heterodimers. UNC-112 is homologous to a human protein of unknown function called Mig-2 (WICK et al. 1994 Down) and shares a short region of homology with talin and other members of the FERM superfamily of proteins (CHISHTI et al. 1998 Down). The UNC-112 protein colocalizes with ßPAT-3 integrin throughout development and is required for the correct spatial distribution of this protein in the muscle cell membrane (ROGALSKI et al. 2000 Down). UNC-112 is not required for the initial polarization of integrin to the muscle cell membrane or for its clustering into nascent attachments. Instead, UNC-112 is needed for the subsequent localization of the nascent attachments into an ordered array within the basal membrane.

In the absence of the UNC-112 protein, actin and myosin filaments do not attach to the muscle cell membrane (WILLIAMS and WATERSTON 1994 Down; ROGALSKI et al. 2000 Down). Embryos homozygous for null mutations in the unc-112 gene exhibit a Pat (paralyzed, arrested at twofold) terminal phenotype. They arrest elongation at the twofold stage of embryonic development and have severely disorganized bodywall muscle (WILLIAMS and WATERSTON 1994 Down). In contrast, animals homozygous for the unc-112(r367) missense mutation are viable, although they do have severely disorganized bodywall muscle (BEJSOVEC et al. 1984 Down; ROGALSKI et al. 2000 Down). When raised at 20°, these mutants move well as young larvae, but gradually slow down as they progress through the L3 and L4 stages, and are small, thin, and paralyzed as adults. The paralysis observed in these mutants is due to the detachment of the myofilament lattice from the muscle cell membrane.

In an attempt to identify proteins that may interact with UNC-112, we undertook a screen for suppressors of the paralyzed phenotype of unc-112(r367) animals. Since the r367 mutation results in the production of an altered protein product (ROGALSKI et al. 2000 Down), it is a good candidate for this type of analysis. All of the suppressor mutations characterized in this study are alleles of the dim-1 (disorganized muscle) gene. These alleles are null mutations and are able to suppress the paralyzed phenotype of unc-112(r367) animals when either heterozygous or homozygous. The overall muscle structure in hermaphrodites lacking a functional dim-1 gene is slightly disorganized, and the myofilament lattice is not as strongly anchored to the muscle cell membrane as it is in wild-type muscle. We have determined that the dim-1 gene utilizes two separate promoters to potentially produce long (640 amino acid) and short (324 amino acid) DIM-1 protein isoforms that contain immunoglobulin (Ig)-like repeats. We have also determined that the short DIM-1 protein is sufficient for wild-type muscle structure and stability and that this protein localizes near the basal muscle cell membrane. Several Ig-domain-containing muscle proteins have been identified in vertebrates, but none of these appear to be orthologs of either DIM-1 protein.


*  MATERIALS AND METHODS
*TOP
 *ABSTRACT
 *MATERIALS AND METHODS
*RESULTS
 *DISCUSSION
 *LITERATURE CITED

Isolation of intragenic revertants and intergenic suppressors of unc-112(r367):
Mutagenesis was done using 0.05 M ethyl methanesulfonate (EMS) in M9 buffer as described by SULSTON and HODGKIN 1988 Down. Dominant or semidominant revertants were obtained by screening the F1 progeny of mutagenized Unc-112 hermaphrodites for animals that moved as adults. All of the revertants were maintained as homozygous strains and were determined to carry either closely linked or unlinked mutations by outcrossing and examining the F2 progeny for the presence or absence of paralyzed unc-112 (r367) animals. The raDf2, raDf4, and raDf7–raDf12 deficiencies were isolated in a similar manner using either 0.8% formaldehyde (MOERMAN and BAILLIE 1981 Down) or UV/psoralen (YANDELL et al. 1994 Down) as the mutagen. Revertant strains carrying mutations that could not be made homozygous (i.e., always segregated paralyzed progeny) were assumed to carry a deficiency of the dim-1 gene.

Mapping the dim-1(ra102) suppressor mutation:
The ra102 mutation was positioned relative to the dpy-7 and dpy-6 loci on LG X using standard genetic three-factor mapping techniques. The disorganized muscle phenotype exhibited by dim-1(ra102) homozygous animals was followed in the mapping crosses using polarized light microscopy. The following protocol was used to determine that the dim-1 gene is deleted by the uDf1 deficiency. Wild-type hermaphrodites with the genotype uDf1/szT1[lon-2(e678)] X; +/szT1(I) were mated with dim-1(ra102)/ 0 males, and hermaphrodite outcross progeny were examined for the Dim-1 muscle phenotype. Dim-1 hermaphrodites were found among the progeny of this cross, indicating that the uDf1 deficiency deletes the dim-1 gene.

Test for suppression of unc-112(r367):
The following protocol was used to show that the dim-1(ra102) mutation suppresses the paralyzed phenotype of homozygous unc-112(r367) hermaphrodites. Wild-type hermaphrodites with the genotype unc-112(r367)/+; dim-1(ra102)/+ were obtained and their progeny were examined for the presence of heterozygously suppressed (slow) unc-112(r367); dim-1(ra102)/+ animals. The presence of these animals would indicate that the ra102 mutation is an intergenic suppressor of the unc-112(r367) mutation. Similarly, slow (suppressed) animals were observed among the progeny of hermaphrodites with the genotype uDf1/+; unc-112(r367)/+, indicating that the uDf1 deficiency suppresses the Unc-112 paralyzed phenotype.

Test for suppression of unc-112(st562):
Hermaphrodites with the genotype sqt-3(e24)unc-112(st562)/+ +; dim-1(ra102) were obtained and their progeny were examined for either adult Sqt-3 animals or arrested sqt-3(e24)unc-112(st562) Pat embryos. No Sqt-3 adult progeny were observed, but these hermaphrodites did segregate ~25% arrested Pat embryos. This is the expected result for lack of suppression of the unc-112(st562) lethal phenotype by the dim-1(ra102) mutation.

Phenotypic characterization of bodywall muscle:
Living worms were examined using both polarized light microscopy and Nomarski differential interference contrast optics. Nematodes were mounted on slides in M9 buffer with Sephadex G-100 (Pharmacia, Piscataway, NJ) beads to support the coverslip and to retain as much of the fragile muscle structure as possible (HEDGECOCK et al. 1990 Down). Polarized light images of wild-type and mutant bodywall muscle were obtained with a Ziess Axiophot photomicroscope (D-7082 Oberkochen) and photographed on Kodak TMAX-400 35-mm film.

Complementation tests with transgenic strains carrying cosmid arrays:
The following procedure was used to determine that the yEx10 cosmid array carries a wild-type copy of the dim-1 gene. Males with the genotype unc-31(e928)IV; him-5(e1490)V; xol-1(y9)/0(X); yEx10[C14G10(IV) + W07E7(X)] were obtained from the TY1123 strain (RHIND et al. 1995 Down) and mated to unc-31(e928)IV; dim-1(ra102)X hermaphrodites. Wild-type males from this cross [genotype: unc-31(e928)IV; him-5(e1490)/+V; dim-1(ra102)/0(X); yEx10[C14G10(IV) + W07E7(X)]] were then mated to unc-31(e928)IV; dim-1 (ra102)X hermaphrodites. Wild-type hermaphrodite progeny from this cross [genotype: unc-31(e928)IV; dim-1(ra102)X; yEx10[C14G10(IV) + W07E7(X)]] were examined using polarized light microscopy. These animals exhibited wild-type bodywall muscle, indicating that the dim-1 gene is located on the W07E7 cosmid. Several of the wild-type male progeny from the first cross were also examined in this manner and were found to have wild-type bodywall muscle as well. A similar protocol was used to show that the C37E11 cosmid does not carry the dim-1 gene. Animals with the genotype unc-31(e928)IV; dim-1(ra102)X or dim-1(ra102)/0(X); yEx3[C14G10(IV) + C37E11(X)] exhibit the Dim-1 muscle phenotype.

PCR and sequence analysis:
The PCR strategy used to determine if a particular cosmid was deleted by any of the raDf2, raDf4, raDf6, raDf7, or raDf9–raDf12 deficiencies is described in ROGALSKI et al. 2000 Down. Approximately four to six arrested raDf homozygous animals were used per PCR reaction. The arrested deficiency homozygotes were obtained from parents with the genotype dpy-7(sc27ts) +/+ raDf.

The PCR strategy used to identify DNA rearrangements or sequence alterations in the C18A11.7 open reading frame (ORF) is also described in ROGALSKI et al. 2000 Down. The primer sets used cover most of the region between exons 1 and 6 and exons 7 and 12 of the dim-1 gene. Four overlapping DNA fragments covering the entire unc-112 gene were amplified for sequencing from adult unc-112(r367ra233) hermaphrodites as described by ROGALSKI et al. 2000 Down. One DNA fragment containing exon 3 from the unc-112 gene was amplified for sequencing from unc-112(r367ra236) and unc-112(r367ra202) homozygous adult hermaphrodites. DNA fragments containing the ra233, ra236, ra202, ra102, ra111, ra203, ra204, ra214, and ra219 mutations were independently amplified and sequenced at least twice.

The yk377b7, yk184a3, yk577c6, and yk628d8 cDNAs were obtained as {lambda}ZapII clones from the C. elegans cDNA project (kindly provided by Y. Kohara, National Institute of Genetics, Mishima, Japan; expressed sequence tagged accession nos. CO8191, CO9860, C34310, C45546, AV194354, AV200573, and AV203498). The entire cDNA inserts of these clones were sequenced on one strand to confirm the intron/exon boundaries predicted by the Genefinder program. The nucleotide and protein sequence have been submitted to GenBank/EMBL/DDBJ under accession nos. AF500489 and AY095447.

DIM-1 homologous proteins were identified using the BLAST search algorithm (ALTSCHUL et al. 1990 Down). Sequence alignments were constructed using the Clustal W program (THOMPSON et al. 1994 Down; available at http://dot.imgen.bcm.tmc.edu).

Construction of genomic clones for microinjection:
The dim-1(S) clone was constructed by cloning a 5.95-kb PCR fragment into the BamHI site of the Bluescript plasmid. This construct carries only dim-1 DNA downstream of the BamHI site in exon 6. The dim-1(S)::GFP DNA construct was made using the technique described by CASSATA et al. 1998 Down and HOBERT et al. 1999 Down. The green fluorescent protein (GFP)-encoding sequences from the pPD117.01 plasmid (kindly provided by A. Fire, S. Xu, J. Ahnn and G. Seydoux, Carnegie Institute, Baltimore) were introduced into the dim-1 gene just before the stop codon. Large DNA fragments were amplified using the expand long template PCR system (Boehringer Mannheim GmbH, Mannheim, Germany, catalog no. 1681834).

Microinjections of C. elegans:
The dim-1 plasmid DNAs (~0.1 µg/ml) were co-injected with the pRF4 rol-6(su1006dm) plasmid (~86 µg/ml) into the gonad syncytium of dim-1(ra102) hermaphrodites as described by MELLO and FIRE 1995 Down. The following transgenic strains were obtained: DM7031: dim-1 (ra102); raEx31[rol-6(su1006dm) + dim-1(S)::GFP]; and DM7036: dim-1(ra102); raEx36[rol-6(su1006dm) + dim-1(S)]. The DM5124: +/+; raEx31[rol-6(su1006dm) + dim-1(S):: GFP] strain was constructed in the following manner: Rol hermaphrodites from the DM7031 strain were mated with wild-type males and an outcross [dim-1(ra102)/+; raEx31] animal was identified using polarized light microscopy. A strain was established from an F2 hermaphrodite that segregated only progeny with wild-type bodywall muscle.

Construction of dim-1 strains carrying the unc-112::GFP transgenic array:
The raEx16[unc-112::GFP; rol-6(su1006)] array was introduced into the the dim-1(ra102) strain by mating hemizygous dim-1(ra102)/0 males to +/+; raEx16[unc-112:: GFP; rol-6(su1006)] hermaphrodites. An outcross Rol hermaphrodite with the genotype dim-1(ra102)/+; raEx16[unc-112::GFP; rol-6(su1006)] was identified, and several of its Rol progeny were selected and allowed to reproduce. The DM2706 strain was established from an animal that segregated only Dim-1 progeny [i.e., had the genotype dim-1(ra102); raEx16 [unc-112::GFP; rol-6(su1006)]].

Generation of polyclonal antisera:
A 550-bp BamHI/EcoRI restriction fragment from the yk377b7 cDNA clone was subcloned into the glutathione S-transferase (GST) expression vector pGEX-1 (SMITH and JOHNSON 1988 Down) to generate the pDM#708 fusion clone. The GST-DIM-1(Asp260-Phe443) fusion protein was purified as described by SMITH and JOHNSON 1988 Down. The DD1 polyclonal antiserum was generated against this GST-DIM-1 fusion protein using the protocol described by MULLEN et al. 1999 Down.

Western hybridization analysis:
Protein extracts were prepared from wild-type and mutant strains in the following manner: Mixed stage populations were washed off plates and pelleted by centrifugation. The worm pellets were resuspended in 2x Laemmli sample buffer at a 1:1 ratio. The samples were boiled for 5 min, sonicated for 30 sec, boiled again for 5 min, and loaded onto 10% polyacrylamide gels. Proteins were resolved by SDS-PAGE and transferred to 0.45-µm-pore Hybond-ECL nitrocellulose filters (Amersham, Buckinghamshire, UK) by the electrophoretic method of TOWBIN et al. 1979 Down. Transfer was for 30–45 min at 10 V in a trans-blot SD electrophoretic transfer cell (Bio-Rad, Richmond, CA). The Western hybridizations were performed as previously described (ROGALSKI et al. 1993 Down). The rabbit polyclonal serum DD1 was diluted 1:5000–1:10,000, and the horseradish peroxidase-labeled goat anti-rabbit IgG secondary antibody (Amersham) was diluted 1:10,000. The filters were incubated with enhanced chemiluminescence (ECL) detection reagents (Amersham) and exposed to Kodak XAR film.

Antibody staining:
Adult hermaphrodites were stained using the freeze-fracture procedure described by ROGALSKI et al. 2000 Down. For immunofluorescence staining, the mouse monoclonal antibody MH25 (FRANCIS and WATERSTON 1985 Down, FRANCIS and WATERSTON 1991 Down) and the rabbit polyclonal GFP serum (Molecular Probes, catalog no. A-6455) were diluted 1:50. The secondary antibodies, Alexa 568-labeled donkey anti-mouse IgG and Alexa 488-labeled donkey anti-rabbit IgG (Molecular Probes, catalog no. A-11029 and A-11031) were diluted 1:200.

Microscopy:
Confocal images were collected using the Bioradiance 2000 system (Bio-Rad) attached to a Zeiss Axiovert S100 compound microscope. Optical sections were collected at 0.2-µm intervals.


*  RESULTS
*TOP
 *ABSTRACT
 *MATERIALS AND METHODS
 *RESULTS
*DISCUSSION
 *LITERATURE CITED

Isolation of intragenic revertants and intergenic suppressors of unc-112(r367):
A total of 34 homozygous revertant strains were isolated by selecting well-moving animals from the progeny of EMS-mutagenized unc-112(r367) hermaphrodites. Of these 34 strains, 17 have been characterized: 3 carry intragenic revertants in the unc-112(r367) gene and 14 carry unlinked suppressors of r367. The three intragenic revertants isolated in this study appear to be wild type, at least at a gross morphological level (see below). The entire unc-112 coding region from animals carrying one of these reversion mutations, unc-112(r367ra233), was sequenced, and the nucleotide alteration responsible for reversion of the r367 phenotype was identified. The mutation responsible for the Unc-112 phenotype changes the Thr85 codon to an Ile codon (ROGALSKI et al. 2000 Down). The ra233 reversion mutation is a G-to-A transition that would result in the Glu95 codon (aag) being changed to a Lys codon (aaa). Both the r367 and ra233 lesions are located in exon 3 of the unc-112 gene. We sequenced this exon from the other two intragenic revertant strains, unc-112(r367ra202) and unc-112(r367ra236), and found that both carried the same nucleotide alteration as the unc-112(r367ra233) hermaphrodites.

Null mutations in the dim-1 gene suppress the paralysis exhibited by unc-112(r367) mutant hermaphrodites:
All 14 of the intergenic suppressor mutations characterized are alleles of the dim-1 gene. Hermaphrodites homozygous for a dim-1 mutation on its own appear as wild type under the dissecting microscope. However, the bodywall muscle appears disorganized when viewed under polarized light. The myofilament lattice in these mutants is not as strongly anchored to the muscle cell membrane as it is in wild-type animals, and as a result some, but not all, bodywall muscle cells in dim-1 hermaphrodites have regions where the myofilament lattice has detached from the membrane. In addition, some bodywall muscle cells in these animals exhibit aberrantly organized A-bands that we refer to as the chevron phenotype (see Fig 1).



View larger version (156K):
In this window
In a new window
Download PPT slide
  		Figure 1. Polarized light micrographs showing disorganized bodywall muscle structure in mutant hermaphrodites. (A) A single adult, wild-type bodywall muscle cell. The bright A-bands (stubby arrow) alternate with darker I-bands. Dense bodies are seen as bright dots within the dark I-bands (curved arrow). (B) A bodywall muscle cell in an unc-112(r367) hermaphrodite. No organized filaments are apparent, only bright clumps (double arrowhead) and short fibers (arrows). (C) A bodywall muscle cell from a dim-1(ra102) mutant animal. This cell exhibits the chevron pattern. Three short A-bands on the left side of C (short arrows) appear to terminate near an A-band of the opposite orientation (long arrows). Dense bodies are seen in the normal and abnormal dark I-bands (curved arrow). (D) The partially suppressed bodywall muscle structure of an unc-112(r367); dim-1(ra102) hermaphrodite. The A-bands are not clearly defined, and dense bodies are not easily seen (curved arrow). Regions of some A-bands appear normal with M-lines still visible (small arrow) while other regions are clumped (double arrowhead). In places it appears that the ends of two A-bands have fused together (*).

Hermaphrodites with the genotype unc-112(r367); dim-1(ra102) are indistinguishable from wild type in overall size and fecundity. However, their movement is noticeably slower than that of wild-type animals. Hermaphrodites with the genotype unc-112(r367); dim-1 (ra102)/+ do not move as well as the homozygous dim-1(ra102) suppressed animals and eventually become paralyzed as older adults. They also retain eggs and hatched larvae, and as a result, have a reduced brood size.

To determine whether suppression of the unc-112 (r367) paralyzed phenotype was due to loss of dim-1 function, a deficiency that uncovers this gene was introduced into the unc-112(r367) mutant strain. Hermaphrodites with the genotype uDf1/+; unc-112(r367) are not paralyzed and are identical in phenotype to unc-112 (r367); dim-1(ra102)/+ hermaphrodites. Therefore, a reduction in the amount of the dim-1 gene product suppresses the unc-112(r367) phenotype. In addition, the phenotype of dim-1(ra102)/uDf1 animals is the same as that of homozygous dim-1(ra102) animals. These results suggest that ra102 is probably a null mutation, and this has been confirmed by identifying the ra102 sequence alteration (see below). Although loss of dim-1 gene function is able to suppress the paralyzed phenotype of r367 mutant hermaphrodites, it is incapable of suppressing the embryonic lethality of the null allele unc-112(st562).

Polarized light microscopy was used to examine and compare the sarcomere structure in the bodywall muscles of wild-type, unc-112(r367), unc-112(r367ra233), dim-1(ra102), and unc-112(r367); dim-1(ra102) adult hermaphrodites (Fig 1). In wild-type bodywall muscle (Fig 1A), polarized light generates a pattern of alternating dark I-bands and light (birefringent) A-bands. This pattern is not seen in the bodywall muscle cells of unc-112(r367) hermaphrodites. In these animals (Fig 1B), bright clumps fill the bodywall muscle cells. Hermaphrodites homozygous for unc-112(r367ra233) are indistinguishable from wild-type animals in appearance and movement when viewed under the dissecting microscope. The bodywall muscle in these revertant animals is also indistinguishable from wild-type muscle when viewed under polarized light (data not shown). Thus, the ra233 second-site reversion mutation appears to restore wild-type function to the unc-112(r367) gene.

Bodywall muscle cells in the dim-1(ra102) animal (Fig 1C) show either parallel rows of alternating A- and I-bands similar to wild type or a nested chevron pattern of bands. The nested chevrons are formed when A-bands and I-bands that extend the full length of the muscle cell intersect at an acute angle with much shorter A-bands and I-bands. This chevron pattern has been observed in bodywall muscle cells in the posterior region of wild-type animals on rare occasions. The dim-1 chevron pattern is novel because it is seen in several cells along the length of a muscle quadrant and is associated with sarcomere fragility in all bodywall muscles. The overall muscle structure in these mutant animals appears slightly disorganized when compared to wild type. The edges of the sarcomeres are frayed and the myofilament lattice is easily detached from the membrane by applying pressure to the coverslip. The pattern of dim-1(ra102); unc-112(r367) muscle (Fig 1D) resembles the chevron pattern, but the bands appear more ragged, and in places the filaments have pulled away from the ends of the cell-forming clumps. The myofilament structure in these suppressed animals is much more organized than the myofilament structure in the unc-112 (r367) mutant animals, but is not as organized as, and is more fragile than, the myofilament structure of either wild-type or dim-1(ra102) hermaphrodites. Thus, dim-1(ra102) suppresses only some of the filament fragility caused by the unc-112(r367) mutation.

Hermaphrodites homozygous for the unc-112(r367) mutation have a much-reduced brood size (~16 progeny/hermaphrodite; n = 30) compared to that of wild-type (~300 progeny/hermaphrodite). Examination using Normarski optics revealed that embryos and oocytes in mutant hermaphrodites were often mislocalized due to a breach in the myoepithelial sheath that surrounds the gonad, and in a few hermaphrodites, one or both arms of the gonad were short or twisted (data not shown). In contrast, the gonads of unc-112(r367ra233), unc-112(r367); dim-1(ra102), and dim-1(ra102) hermaphrodites appear normal when examined by Normarski optics, and the brood sizes of these animals are comparable to that of wild-type animals.

dim-1 gene encodes novel proteins containing Ig-like repeats:
We initially positioned the dim-1 locus to the right of dpy-7 on LG X. To define the location of this gene on the physical map we isolated several deficiencies that include the dim-1 locus and correlated their breakpoints with the physical map (data available at http://www.wormbase.org). The results obtained localized the dim-1 gene to a region of five to seven cosmids, and rescue with a transgenic array placed the dim-1 gene on the sequenced C18A11 cosmid. We have determined that the dim-1 gene corresponds to the C18A11.7 open reading frame (C. ELEGANS GENOME SEQUENCING CONSORTIUM 1998; data available from GenBank/EMBL/DDBJ under accession no. U39667) by identifying the sequence alterations corresponding to six dim-1 alleles (see Fig 2). The ra111 mutation is a 183-bp deletion and the remaining five alleles are GC-AT point mutations. The ra102, ra215, and ra203 mutations alter splice donor or acceptor sites, and ra219 and ra204 change tryptophan codons (tgg) to stop codons (tga) in exons 10 (Trp455) and 11 (Trp565), respectively.



View larger version (24K):
In this window
In a new window
Download PPT slide
  		Figure 2. The dim-1 gene encodes two polypeptides. A line drawing representing the genomic DNA fragment containing the C18-A11.7/dim-1 ORF is shown. The 12 exons are represented as boxes. The gene structure shown here has been confirmed by sequencing the full-length yk577c6 cDNA clone (data are available from GenBank/EMBL/DDBJ under accession no. AY095447). The sequence alterations corresponding to six dim-1 mutations are identified on the gene map. The cDNA analysis predicts the two DIM-1 polypeptides shown here. The rectangle represents the novel amino acid sequence of DIM-1, and the ovals represent the Ig-like repeats.

Our sequencing results for the cDNA clones yk577c6 and yk628d8 revealed that the C18A11.7 ORF consists of 12 exons with a rather large (3.5 kb) intron separating exons 6 and 7 (see Fig 2). The yk577c6 and yk628d8 clones encode identical 2.4-kb cDNAs that extend from the putative start codon in exon 1 to ~300 bp downstream of the stop codon in exon 12. In addition, the 5' end of the yk577c6 cDNA contains 11 nucleotides that match the sequence of the SL1 splice leader (KRAUSE and HIRSH 1987 Down). Our sequencing results for the yk184a3 cDNA clone identified a second, smaller, SL1-spliced dim-1 transcript starting at exon 7. Both transcripts have the same 3' end but have different 5' start sites, and both appear to be trans-spliced to SL1. The first 316 amino acids of the full-length protein (encoded by exons 1–6; Fig 2) are not similar to any known protein, whereas the remaining 324 amino acids comprise three Ig-like repeats (encoded by exons 7–12; Fig 2) similar to those found in Ig-domain-containing muscle proteins. The smaller dim-1 transcript encodes a protein containing the three Ig-like repeats (Fig 2).

A search of the GenBank/EMBL/DDBJ database identified orthologs of the short DIM-1 isoform in two other nematode species (C. briggsae and Dirofilaria immitis) but not in flies or mammals (Fig 3). No ortholog of the long isoform has been found in any organism. There is a one-amino-acid difference between the predicted sequence of the CB027N19.k ORF from the closely related nematode C. briggsae and the sequence of the short DIM-1 protein. The Ala residue at position 543 in the C. elegans sequence in Fig 3 is a Ser residue in the C. briggsae ortholog. The second C. briggsae dim-1 ortholog (CB027N19.l) is only 83% identical to either the dim-1 or the CB027N19.k sequences. The D. immitis ortholog shares 76% identity and 84% similarity to the dim-1 gene.



View larger version (84K):
In this window
In a new window
Download PPT slide
  		Figure 3. Comparison of the predicted amino acid sequences of DIM-1(L), the C. briggsae CB02-7N9.k (CbDIM-1a), and CB027N-19.l (CbDIM-1b) proteins (data are available from GenBank/EMBL/DDBJ under accession no. AC084-455), and the D. immitis DIM-1 ortholog (DiDIM-1; GenBank/EMBL/DDBJ accession no. AX7-8091). The long DIM-1 protein encoded by the yk577c6 cDNA (GenBank/EMBL/DDBJ accession no. AY095447) consists of animo acids 1–640, and the short DIM-1 isoform encoded by the yk184a3 cDNA (GenBank/EMBL/DDBJ accession no. AF500489) contains animo acids 317–640. The CbD-IM-1a, CbDIM-1b, and DiDIM-1 proteins are 99, 83, and 76% identical to the short DIM-1 polypeptide. The amino acid sequences were aligned using the Clustal W program. Identical amino acids are shaded, and similar amino acids are boxed.

The genomic regions around the dim-1 loci in both C. elegans and C. briggsae have been sequenced by the C. elegans Genome Sequencing Consortium (data available at http://www.wormbase.org). A comparison of the annotated genes in these two regions reveals some similarity in their organization, although interesting differences are observed. One difference is that a tandem duplication of the last six exons of the dim-1 gene (encoding the short protein isoform) has occurred in C. briggsae accounting for the presence of two dim-1(S) orthologs in this nematode. Another difference is that the first six exons of the C. elegans dim-1 gene are not present in the sequenced genomic DNA in this region of C. briggsae.

The short DIM-1 protein is sufficient for wild-type muscle structure and stability:
We determined that the 3.5-kb region between exons 6 and 7 of the dim-1 gene is sufficient for expression of the short DIM-1 protein isoform and that this protein isoform is sufficient for wild-type muscle structure and stability. A plasmid containing only the dim-1 DNA sequences downstream of exon 6 was co-injected with the pRF4[rol-6] plasmid into dim-1(ra102) hermaphrodites. One stable transgenic line was obtained, but only after reducing the concentration of the dim-1 plasmid DNA to ~0.1 ng/µl. The raEx36[dim-1(S); rol-6(su1006)] array produces a functional protein that rescues the disorganized bodywall muscle phenotype of homozygous dim-1(ra102) mutant hermaphrodites.

The short DIM-1 protein localizes near the muscle cell membrane:
Two strategies were employed to address when and where the dim-1 gene products are expressed and localized. First, a polyclonal antiserum, DD1, was generated against amino acids 260–443 of the long DIM-1 protein. Unfortunately, this serum is not able to detect either of the DIM-1 proteins in situ. However, it does detect a protein of ~35,000 Mr on Western blots of wild-type worm lysates (Fig 4). The size of this protein is similar to the predicted size of the short DIM-1 isoform. BINI et al. 1997 Down have shown previously that the short DIM-1 protein is present in worm homogenates. The ~71,000-Mr long isoform predicted by the yk577c6 cDNA sequence is not detected here, nor was it detected in the previous study (BINI et al. 1997 Down). Protein extracts prepared from homozygous dim-1 mutants were also analyzed by Western blotting. In the cases of dim-1 (ra111), dim-1(ra102), dim-1(ra203), dim-1(ra204), and dim-1(ra215), the 35,000-Mr band is completely absent, whereas in the dim-1(ra219) extract a ~15,000-Mr truncated protein is detected. These results confirm that the DD1 serum is specific to the dim-1 gene product. Fig 4 shows a Western blot of protein extracts from wild-type, dim-1(ra102), and dim-1(ra219) animals reacted with the DD1 antibody.



View larger version (40K):
In this window
In a new window
Download PPT slide
  		Figure 4. The DD1 polyclonal antibody recognizes a DIM-1 polypeptide. DD1 recognizes a 35,000-Mr protein on Western blots of wild-type extracts (N2; lanes 1 and 3). This protein is absent in the extract from the dim-1(ra102) mutant animals (lane 2). A 15,000-Mr truncated protein is detected in the extract from the dim-1(ra219) mutant animals (lane 4).

The second strategy that we employed was to express a GFP-tagged short DIM-1 protein from a transgenic array. The plasmid constructed for these experiments was identical to the dim-1(S) rescuing plasmid described earlier but with the addition of GFP-encoding sequences just before the stop codon in exon 12. The DD1 serum and an anti-GFP polyclonal serum both recognize a GFP-tagged short DIM-1 isoform on Western blots of protein extracts from a transgenic dim-1(ra102) strain carrying the raEx31[dim-1(S)::GFP; rol-6(su1006)] array (data not shown). However, the GFP-tagged DIM-1 protein produced by this array is not completely functional since it is unable to rescue the disorganized muscle phenotype of dim-1(ra102) mutant hermaphrodites.

We have been able to determine the expression pattern of the short DIM-1 isoform using the dim-1::GFP translational fusion produced by the raEx31[dim-1(S):: GFP; rol-6(su1006)] transgenic array. The in situ localization of this protein in a wild-type background was determined by indirect immunofluorescence using an anti-GFP polyclonal serum (Fig 5). An antibody was required because the native fluorescence of the DIM-1 (S)::GFP protein is very faint. DIM-1(S)::GFP is expressed throughout development from the ~1.5-fold stage of embryogenesis and appears to be limited to bodywall muscles. Anti-GFP immunofluorescence was not observed in any of the other muscle types.



View larger version (47K):
In this window
In a new window
Download PPT slide
  		Figure 5. The short DIM-1 protein is expressed in bodywall muscle and localizes near the basal cell membrane. (A–C) A single bodywall muscle cell in an adult +/+; raEx31[dim-1 (S)::GFP; rol-6(su1006)] hermaphrodite stained with the ßPAT-3 integrin mAb, MH25 (A), an anti-GFP polyclonal antibody (B), or both antibodies (C). The MH25 staining is shown in red and the anti-GFP staining is shown in green. In this animal, ßPAT-3 integrin is localized in the basal cell membrane at dense bodies, M-lines, and the regions of contact between muscle cells (A). The DIM-1::GFP protein is also localized near the basal cell membrane in stripes that run parallel with, and in the spaces between, the dense bodies (B). When the two staining patterns in the wild-type muscle cell are merged (C), it appears that the distribution of the DIM-1(S)::GFP protein and ßPAT-3 integrin do not overlap extensively. (D) Part of one bodywall muscle quadrant from a dim-1(ra102); raEx31[dim-1(S)::GFP; rol-6(su1006)] adult hermaphrodite stained with an anti-GFP polyclonal antibody. Most of the DIM-1::GFP protein is found in clumps in the cytoplasm of the muscle cells in this dim-1(ra102) mutant hermaphrodite. A–C are shown at a higher magnification than D. Bar, 5 µm.

UNC-112 and ßPAT-3 integrin colocalize at muscle cell dense bodies, at M-lines, and at the regions of contact between adjacent muscle cells. Here we show that the GFP-tagged DIM-1(S) isoform is localized at the basal membrane region of bodywall muscle cells, but does not appear to colocalize with ßPAT-3 integrin or, by inference, with UNC-112. Fig 5 shows the results obtained when +/+; raEx31[dim-1(S)::GFP; rol-6 (su1006)] adult hermaphrodites are stained with MH25, an mAb that recognizes ßPAT-3 integrin, and with an anti-GFP polyclonal serum that recognizes the DIM-1(S)::GFP protein. Fig 5A shows MH25 staining localized to the dense bodies, M-lines, and muscle cell boundaries in an adult bodywall muscle cell. Fig 5B shows the anti-GFP staining pattern in the same cell. This antibody localizes near the membrane in stripes that run parallel to, and in the spaces between, the dense bodies. When the two staining patterns are merged (Fig 5C), it appears that, although the DIM-1(S)::GFP protein and ßPAT-3 integrin are both located at the basal membrane, their distribution overlaps only slightly.

Fig 5D shows the anti-GFP staining pattern in an adult dim-1(ra102) hermaphrodite carrying the raEx31 [dim-1(S)::GFP; rol-6(su1006)] array. Although some DIM-1::GFP protein localizes to the basal cell membrane in the mutant animal, it is clearly less organized than in wild type (compare Fig 5B with Fig 5D). Most of the GFP-tagged DIM-1 protein is found in clumps within the cytoplasm of the muscle cells (Fig 5D). These data reveal that the GFP-tagged short DIM-1 polypeptide does not localize properly in the dim-1(ra102) mutant background.

DIM-1 is not required for UNC-112 localization:
The raEx16[unc-112::GFP; rol-6(su1006)] transgenic array produces a functional UNC-112::GFP fusion protein that is expressed in bodywall, vulval, uterine, spermathecal, and anal sphincter/depressor muscle cells. In bodywall muscle this protein localizes to dense bodies, M-lines, and the regions of contact between adjacent muscle cells (ROGALSKI et al. 2000 Down; Fig 6A). In adult dim-1(ra102) animals, the UNC-112::GFP protein is properly localized to its position at the muscle cell membrane, but is slightly disorganized compared to wild type (compare Fig 6A and Fig 6B). UNC-112::GFP distribution in L2 and L3 dim-1(ra102) larvae is indistinguishable from wild type (data not shown), and it is not until animals become late L4 larvae or adults that UNC-112::GFP begins to appear disorganized. Thus, the severity of disorganization increases with the age of the mutant hermaphrodites and is the result of the general disruption of the myofilament lattice rather than a result of the specific requirement for DIM-1 in UNC-112 localization. This is also the case for ßPAT-3 integrin (data not shown).



View larger version (92K):
In this window
In a new window
Download PPT slide
  		Figure 6. UNC-112::GFP localization does not require the DIM-1 polypeptides. UNC-112:: GFP fluorescence in the bodywall muscle of an adult +/+; raEx16[unc-112::GFP; rol-6(su1006)] hermaphrodite (A) and an adult dim-1(ra102); raEx16[unc-112::GFP; rol-6(su1006)] hermaphrodite (B). UNC-112::GFP localizes to dense bodies, M-lines, and adhesion plaques in both animals. However, the placement of these structures is disorganized in the Dim-1 mutant compared to the placement in wild type. The extent of disorganization is variable between cells (compare the cell at the botton right of B to the cell just above it).


*  DISCUSSION
*TOP
 *ABSTRACT
 *MATERIALS AND METHODS
 *RESULTS
 *DISCUSSION
*LITERATURE CITED

Suppressor analysis can be useful for identifying genes with mild or unusual phenotypes that would not easily be found by standard genetic means. This study describes the identification of one such gene, dim-1. Loss or reduction of dim-1 gene function can suppress the severe muscle disruption and paralysis exhibited by hermaphrodites homozygous for the r367 missense mutation in the unc-112 gene. The only phenotype observed in animals lacking a functional dim-1 gene is a slight disorganization of the bodywall muscle when viewed under polarized light. The dim-1 gene encodes two polypeptides that contain three Ig-like repeats. In the absence of these proteins, dense bodies and M-lines assemble and the myofilament lattice attaches to these structures. However, the attachment of the myofilament lattice to the muscle cell membrane is more fragile than normal, and as a result, Dim-1 muscle can be more easily disrupted than wild-type muscle. An additional phenotype observed is the occasional mispositioning of the dense bodies and M-lines in the muscle cell membrane.

All known alleles of the dim-1 gene have been isolated as suppressors of unc-112(r367). Interestingly, all of the sequence alterations that have been identified to date are null mutations in the 3' half of this gene and affect both dim-1 gene products. This includes the 6 alleles identified here as well as another 14 described by E. GILCHRIST, G. MULLEN, T. M. ROGALSKI and D. G. MOERMAN (unpublished results). Apparently elimination of just the large DIM-1 protein is not sufficient to suppress the unc-112(r367) phenotype. Either just the short DIM-1 protein or, perhaps, both DIM-1 proteins must be removed for suppression. The data presented in this article reveal that the short DIM-1 protein is sufficient for wild-type bodywall muscle structure and stability in the absence of the long protein isoform.

Using a DIM-1(S)::GFP translational fusion we have discovered that the short DIM-1 protein isoform is expressed in all bodywall muscle cells throughout development. This fusion protein does not have wild-type function, but does localize near the muscle cell membrane surrounding the dense bodies in the presence of the wild-type protein. We have been unable to confirm this localization because the polyclonal antibody that we made does not detect the DIM-1 protein in situ. However, on the basis of the data that we have obtained, we tentatively conclude that the wild-type DIM-1(S) protein localizes near the basal membrane in bodywall muscle cells. The DIM-1 protein does not contain a transmembrane region so it is probably not inserted in the membrane. Most likely, it binds to one of the other proteins associated with the membrane. Interestingly, DIM-1(S) localizes just to the region around and between the dense bodies, which are the structures that anchor the actin filaments. The localization of the long DIM-1 isoform has not been determined. In fact, this protein isoform is not detected on Western blots of wild-type worm lysates with the DD1 serum, even though this serum should be able to detect an intact DIM-1(L) protein. Also, no orthologs of the DIM-1(L) protein are in either C. briggsae or D. immitis. At this time, we do not know whether the long isoform of DIM-1 is made at all, whether it is cleaved post-translationally to produce a short protein isoform, or whether it is degraded or lost in our protein preparations.

The DIM-1 proteins contain three immunoglobulin-like repeats or domains. Ig domains have been classified into four groups, termed V, C1, C2, and I, on the basis of their sequence and their ß-sheet conformation (HARPAZ and CLOTHIA 1994 Down; SMITH and XUE 1997 Down). An analysis of the Ig domains of DIM-1 by the SMART analysis program (LETUNIC et al. 2002 Down; available at http://www.dylan.embl-heidelberg.de) reveals that the first Ig domain of DIM-1 belongs to the C2 class. The two remaining Ig domains, although similar to Ig's in structure, do not fit into any of the four categories.

The Ig protein module is found in a diverse group of proteins including antibodies, cell adhesion molecules, cell surface receptors, and muscle proteins (SMITH and XUE 1997 Down). DIM-1 is a member of the intracellular muscle branch of the immunoglobulin superfamily of proteins. The founding member of this branch is the C. elegans UNC-22/twitchin polypeptide (BENIAN et al. 1989 Down), which localizes to A-bands (MOERMAN et al. 1988 Down) and is thought to function in the regulation of muscle contraction (MOERMAN et al. 1982 Down). Since these studies were completed many intracellular muscle proteins have been identified. In general these proteins are associated either with the M-line and thick filaments or with the Z-disc and thin filaments. The vertebrate M-line contains M-protein, myomesin (FURST and GAUTEL 1995 Down), the carboxy-terminal portion of titin (LABEIT and KOLMERER 1995 Down), and skelemin (PRICE and GOMER 1993 Down). Other members of this family include C-protein (EINHEBER and FISCHMAN 1990 Down), telokin (GALLAGHER and HERRING 1991 Down), and insect projectin (AYME-SOUTHGATE et al. 1995 Down). Most of these proteins, including the C. elegans proteins UNC-22/twitchin and UNC-89, are thought to interact with myosin through their Ig and FnIII domains (BENIAN et al. 1989 Down, BENIAN et al. 1993 Down, BENIAN et al. 1996 Down). This interaction has been directly demonstrated for titin (LABEIT et al. 1992 Down), telokin (SHIRINSKY et al. 1993 Down), C-protein (OKAGAKI et al. 1993 Down), and myomesin (OBERMANN et al. 1997 Down). An exception is skelemin, which has been reported to interact with ß-integrin subunits in nonmuscle cells (REDDY et al. 1998 Down). In addition, proteins that bind thin filaments and {alpha}-actinin have also been identified. The kettin protein in Drosophila melanogaster binds to both actin and {alpha}-actinin (LAKEY et al. 1993 Down). The mammalian myotilin (SALMIKANGAS et al. 1999 Down), palladin (PARAST and OTEY 2000 Down; BANG et al. 2001 Down), and myopalladin (BANG et al. 2001 Down) proteins all colocalize at Z-discs with {alpha}-actinin, and both myotilin and myopalladin have been shown to directly interact with {alpha}-actinin.

At the present time, the short DIM-1 polypeptide is found only in nematodes and is not an ortholog of any of the Ig-domain-containing muscle proteins identified in other organisms. At the protein sequence level, DIM-1 is no more similar to any of the thick filament-associated proteins than it is to any of the thin filament-associated proteins. In overall structure, DIM-1 is most similar to palladin since both proteins have only three Ig domains. However, palladin also has a unique, proline-rich, amino- terminal region. Our data tentatively place the short DIM-1 polypeptide at the muscle cell membrane adjacent to the structures that anchor the thin filaments. If this subcellular localization is correct, then the primary role of DIM-1 may be in stabilizing the thin rather than the thick filament components of the sarcomere. Whatever the role of this protein it appears to be either redundant or relatively minor and may not be required in vertebrate muscle.

The data presented here argue against a direct interaction between UNC-112 and DIM-1(S) in wild-type muscle function. The localization studies reveal very little overlap between the two proteins, and UNC-112 does not require DIM-1 for its localization and assembly into dense bodies and M-lines. All of the data that we have obtained lead us to propose a model whereby the inappropriate expression of one or both DIM-1 proteins could be responsible for the phenotype observed in unc-112(r367) mutant animals. The r367 missense mutation in the UNC-112 protein may affect the localization of the DIM-1 proteins such that they interfere with one or more of the components that anchor the myofilament lattice. This, in turn, could cause the lattice to rip away from the muscle cell membrane when stressed. Removal of the DIM-1 proteins would then suppress this phenotype. If this model is correct, then the major effect of the r367 mutation is on the localization of one or both DIM-1 polypeptides. However, this is probably not the only effect of this mutation since removal of the DIM-1 gene products in the r367 background results in a phenotype that is worse than the Dim-1 phenotype in a wild-type background. The ragged appearance of the myofilament lattice and the increased fragility of the bodywall muscle in unc-112(r367); dim-1(ra102) animals compared to that of dim-1(ra102) animals may be a direct result of the alteration in the UNC-112 protein.

MACKINNON et al. 2002 Down have identified an interaction between UNC-112 and PAT-4/ILK, another protein that is also found in the dense bodies and M-lines. These two proteins are required during initial muscle assembly when they play a role in recruiting and stabilizing attachment proteins to form dense bodies and M-lines (ROGALSKI et al. 2000 Down; MACKINNON et al. 2002 Down). When either of these proteins is absent, a myofilament lattice fails to form and animals arrest at the twofold stage of embryogenesis. In contrast, neither of the DIM-1 proteins is required for assembly of the myofilament lattice; however, the short isoform does play a role in maintaining a strong connection between the myofilament lattice and the basal cell membrane during growth.


*  FOOTNOTES

Sequence data from this article have been deposited with the DDBJ/EMBL/GenBank Data Libraries under accession nos. AY095447 and AF500489. Back


*  ACKNOWLEDGMENTS

We thank Dr. Barbara Meyer, Dr. Ann Rose, Dr. David Baillie, and Dr. Ben Williams for providing strains, Dr. Michelle Hresko for providing the MH25 antibody, Dr. Yuji Kohara for providing cDNAs, Dr. Andy Fire for providing GFP plasmids, and the C. elegans Genome Sequencing Consortium for dim-1 sequence data. Some nematode strains used in this work were provided by the Caenorhabditis Genetics Center, which is funded by the National Institutes of Health National Center for Research Resources. M.M.G. was supported by a scholarship from the Heart and Stroke Foundation of Canada. This work was funded by grants from the Canadian Institutes for Health Research, the Natural Sciences and Engineering Research Council of Canada, and the Health Research Foundation of British Columbia to D.G.M.

Manuscript received September 18, 2002; Accepted for publication December 4, 2002.


*  LITERATURE CITED
*TOP
 *ABSTRACT
 *MATERIALS AND METHODS
 *RESULTS
 *DISCUSSION
 *LITERATURE CITED

ALTSCHUL, S. F., W. GISH, W. MILLER, E. W. MYERS, and D. J. LIPMAN, 1990  Basic local alignment search tool. J. Mol. Biol. 215:403-410.[Medline]

AYME-SOUTHGATE, A., R. SOUTHGATE, J. SAIDE, G. M. BENIAN, and M. L. PARDUE, 1995  Both synchronous and asynchronous muscle isoforms of projectin (the Drosophila bent locus product) contain functional kinase domains. J. Cell Biol. 128:393-403.[Abstract/Free Full Text]

BANG, M. L., R. E. MUDRY, A. S. MCELHINNY, K. TROMBITAS, and A. J. GEACH et al., 2001  Myopalladin, a novel 145-kilodalton sarcomeric protein with multiple roles in Z-disc and I-band protein assemblies. J. Cell Biol. 153:413-427.[Abstract/Free Full Text]

BEJSOVEC, A., D. EIDE and P. ANDERSON, 1984 Genetic techniques for analysis of nematode muscle, pp. 267–273 in Molecular Biology of the Cytoskeleton, edited by G. BORISY, D. CLEVELAND and D. MURPHY. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

BENIAN, G. M., J. E. KIFF, M. NECKELMANN, D. G. MOERMAN, and R. H. WATERSTON, 1989  Sequence of an unusually large protein implicated in regulation of myosin activity in C. elegans. Nature 342:45-50.[Medline]

BENIAN, G. M., S. W. L'HERNAULT, and M. E. MORRIS, 1993  Additional sequence complexity in the muscle gene unc-22, and its encoded protein, twitchin, of Caenorhabditis elegans.. Genetics 134:1097-1104.[Abstract]

BENIAN, G. M., T. L. TINLEY, X. TANG, and M. BORODOVSKY, 1996  The Caenorhabditis elegans gene unc-89, required for muscle assembly, encodes a giant modular protein composed of Ig and signal transduction domains. J. Cell Biol. 132:835-848.[Abstract/Free Full Text]

BINI, L., H. HEID, S. LIBERATORI, G. GEIER, and V. PALLINI et al., 1997  Two-dimensional gel electrophoresis of Caenorhabditis elegans homogenates and identification of protein spots by microsequencing. Electrophoresis 8:557-562.

BURRIDGE, K. and P. CHRZANOWSKA-WODNICKA, 1996  Focal adhesions, contractility, and signaling. Annu. Rev. Cell Dev. Biol. 12:463-519.[Medline]

CASSATA, G., H. KAGOSHIMA, R. F. PRETOT, G. ASPOCK, and G. NIKLAUS et al., 1998  Rapid expression screening of Caenorhabditis elegans homeobox open reading frames using a two-step polymerase chain reaction promotor-gfp reporter construction. Gene 212:127-135.[Medline]

Genome sequence of the nematode C. elegans: a platform for investigating biology. (1998) Science 282:2012-2017.[Abstract/Free Full Text]

CHISHTI, A. H., A. C. KIM, S. M. MARFATIA, M. LUTCHMAN, and M. HANSPAL et al., 1998  The FERM domain: a unique module involved in the linkage of cytoplasmic proteins to the membrane. Trends Biochem. Sci. 23:281-282.[Medline]

EINHEBER, S. and D. A. FISCHMAN, 1990  Isolation and characterization of a cDNA clone encoding avian skeletal muscle C-protein: an intracellular member of the immunoglobulin superfamily. Proc. Natl. Acad. Sci. USA 87:2157-2161.[Abstract/Free Full Text]

FRANCIS, G. R. and R. H. WATERSTON, 1985  Muscle organization in C. elegans: localization of proteins implicated in thin filament attachment and I-band organization. J. Cell Biol. 101:1532-1549.[Abstract/Free Full Text]

FRANCIS, G. R. and R. H. WATERSTON, 1991  Muscle cell attachment in Caenorhabditis elegans.. J. Cell Biol. 11:465-479.

FURST, D. O. and M. GAUTEL, 1995  The anatomy of a molecular giant: how the sarcomere cytoskeleton is assembled from immunoglobulin superfamily molecules. J. Mol. Cell Cardiol. 27:951-959.[Medline]

GALLAGHER, P. J. and B. P. HERRING, 1991  The carboxyl terminus of the smooth muscle myosin light chain kinase is expressed as an independent protein, telokin. J. Biol. Chem. 266:23945-23952.[Abstract/Free Full Text]

GETTNER, S. N., C. KENYON, and L. F. REICHARDT, 1995  Characterization of ß-Pat-3 heterodimers, a family of essential integrin receptors in C. elegans.. J. Cell Biol. 129:1127-1141.[Abstract/Free Full Text]

HARPAZ, Y. and C. CLOTHIA, 1994  Many of the immunoglobulin superfamily domains in cell adhesion molecules and surface receptors belong to a new structural set which is close to that containing variable domains. J. Mol. Biol. 238:528-539.[Medline]

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.

HOBERT, O., D. G. MOERMAN, K. A. CLARK, M. C. BECKERLE, and G. RUVKIN, 1999  A conserved LIM protein that affects muscle adherens junction integrity and mechanosensory function in the nematode Caenorhabditis elegans.. J. Cell Biol. 144:45-57.[Abstract/Free Full Text]

HRESKO, M. C., B. D. WILLIAMS, and R. H. WATERSTON, 1994  Assembly of bodywall muscle and muscle cell attachment structures in Caenorhabditis elegans.. J. Cell Biol. 24:491-506.

HRESKO, M. C., L. A. SCHRIEFER, P. SHRIMANKAR, and R. H. WATERSTON, 1999  Myotactin, a novel hypodermal protein involved in muscle-cell adhesion in Caenorhabditis elegans.. J. Cell Biol. 146:659-672.[Abstract/Free Full Text]

KRAUSE, M. and D. HIRSH, 1987  A trans-spliced leader sequence on actin mRNA in C. elegans. Cell 49:753-761.[Medline]

LABEIT, S. and B. KOLMERER, 1995  Titans: giant proteins in charge of muscle ultrastructure and elasticity. Science 270:293-296.[Abstract/Free Full Text]

LABEIT, S., M. GAUTEL, A. LAKEY, and J. TRINICK, 1992  Towards a molecular understanding of titin. EMBO J. 5:1711-1716.

LAKEY, A., S. LABEIT, M. GAUTEL, C. FERGUSON, and D. P. BARLOW et al., 1993  Kettin, a large modular protein in the Z-disc of insect muscles. EMBO J. 12:2863-2871.[Medline]

LETUNIC, I., L. GOODSTADT, N. J. DICKENS, T. DOERKS, and J. SHULTZ et al., 2002  Recent improvements to the SMART domain-based sequence annotation resource. Nucleic Acids Res. 30:242-244.[Abstract/Free Full Text]

MACKINNON, A. C., H. QADOTA, K. R. NORMAN, D. G. MOERMAN, and B. D. WILLIAMS, 2002  C. elegans PAT-4/ILK functions as an adaptor protein with integrin adhesion complexes. Curr. Biol. 12:1-20.[Medline]

MELLO, C., and A. FIRE, 1995 DNA transformation, pp. 452–482 in Caenorhabditis elegans: Modern Biological Analysis of an Organism, edited by H. F. EPSTEIN and D. C. SHAKES. Academic Press, San Diego.

MOERMAN, D. G. and D. L. BAILLIE, 1981  Formaldehyde mutagenesis in the nematode C. elegans. Mutat. Res. 80:273-279.[Medline]

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

MOERMAN, D. G., S. PLURAD, R. H. WATERSTON, and D. L. BAILLIE, 1982  Mutations in the unc-54 myosin heavy chain gene of Caenorhabditis elegans that alter contractility but not muscle structure. Cell 29:773-781.[Medline]

MOERMAN, D. G., G. M. BENIAN, R. J. BARSTEAD, L. A. SCHRIEFER, and R. H. WATERSTON, 1988  Identification and intracellular localization of the unc-22 gene product of Caenorhabditis elegans.. Genes Dev. 2:93-105.[Abstract/Free Full Text]

MULLEN, G. P., T. M. ROGALSKI, J. A. BUSH, P. RAHMANI GORGI, and D. G. MOERMAN, 1999  Complex patterns of alternative splicing mediate the spatial and temporal distribution of perlecan/UNC-52 in Caenorhabditis elegans.. Mol. Biol. Cell 10:3205-3221.[Abstract/Free Full Text]

OBERMANN, W. M., M. GAUTEL, K. WEBER, and D. O. FURST, 1997  Molecular structure of the sarcomeric M band: mapping of titin and myosin binding domains in myomesin and the identification of a potential regulatory phosphorylation site in myomesin. EMBO J. 16:211-220.[Medline]

OKAGAKI, T., F. E. WEBER, D. A. FISCHMAN, K. T. VAUGHN, and T. MIKAWA et al., 1993  The major myosin-binding domain of skeletal muscle MyBP-C (C protein) resides in the COOH-terminal, immunoglobulin C2 motif. J. Cell Biol. 123:619-626.[Abstract/Free Full Text]

PARAST, M. M. and C. A. OTEY, 2000  Characterization of palladin, a novel protein localized to stress fibers and cell adhesions. J. Cell Biol. 150:643-656.[Abstract/Free Full Text]

PRICE, M. G. and R. H. GOMER, 1993  Skelemin, a cytoskeletal M-disc periphery protein contains motifs of adhesion/recognition and intermediate filament proteins. J. Biol. Chem. 268:21800-21810.[Abstract/Free Full Text]

REDDY, K. B., P. GASCARD, M. G. PRICE, E. V. NEGRESCU, and J. E. FOX, 1998  Identification of an interaction between the m-band protein skelemin and beta-integrin subunits. Colocalization of a skelemin-like protein with beta1- and beta3-integrins in non-muscle cells. J. Biol. Chem. 273:35039-35047.[Abstract/Free Full Text]

RHIND, N. R., L. M. MILLER, J. B. KOPCZYNSKI, and B. J. MEYER, 1995  xol-1 acts as an early switch in the C. elegans male/hermaphrodite decision. Cell 80:71-82.[Medline]

ROGALSKI, T. M., B. D. WILLIAMS, G. P. MULLEN, and D. G. MOERMAN, 1993  The products of the unc-52 gene in Caenorhabditis elegans are homologous to the core protein of the mammalian basement membrane heparan sulfate proteoglycan. Genes Dev. 7:1471-1484.[Abstract/Free Full Text]

ROGALSKI, T. M., G. P. MULLEN, M. M. GILBERT, B. D. WILLIAMS, and D. G. MOERMAN, 2000  The unc-112 gene in Caenorhabditis elegans encodes a novel component of cell-matrix adhesion structures required for integrin localization in the muscle cell membrane. J. Cell Biol. 150:253-264.[Abstract/Free Full Text]

SALMIKANGAS, P., O. M. MYKKANEN, M. GRONHOLM, L. HEISKA, and J. KERE et al., 1999  Myotilin, a novel sarcomeric protein with two Ig-like domains, is encoded by a candidate gene for limb-girdle muscular dystrophy. Hum. Mol. Genet. 8:1329-1336.[Abstract/Free Full Text]

SHIRINSKY, V. P., A. V. VOROTNIKOV, K. G. BIRUKOV, A. K. NANAEV, and M. COLLINGE et al., 1993  A kinase-related protein stabilizes unphosphorylated smooth muscle myosin minifilaments in the presence of ATP. J. Biol. Chem. 268:16578-16583.[Abstract/Free Full Text]

SMITH, D. B. and K. S. JOHNSON, 1988  Single-step purification of polypeptides expressed in Escherichia coli as fusions with glutathione-S-transferase. Gene 67:31-40.[Medline]

SMITH, D. K. and H. XUE, 1997  Sequence profiles of immunoglobulin and immunoglobulin-like domains. J. Mol. Biol. 274:530-545.[Medline]

SULSTON, J., and J. HODGKIN, 1988 Methods, pp. 587–606 in The Nematode Caenorhabditis elegans, edited by W. B. WOOD. Cold Spring Harbor Laboratory Press, Cold Spring Harbor,