Functions of the Caenorhabditis elegans Regulatory Myosin Light Chain Genes mlc-1 and mlc-2

Corresponding author: Philip Anderson, Department of Genetics, University of Wisconsin, 445 Henry Mall, Madison, WI 53706., andersn{at}facstaff.wisc.edu (E-mail).

Communicating editor: R. K. HERMAN

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Caenorhabditis elegans contains two muscle regulatory myosin light chain genes, mlc-1 and mlc-2. To determine their in vivo roles, we identified deletions that eliminate each gene individually and both genes in combination. Functions of mlc-1 are redundant to those of mlc-2 in both body-wall and pharyngeal muscle. mlc-1(0) mutants are wild type, but mlc-1(0) mlc-2(0) double mutants arrest as incompletely elongated L1 larvae, having both pharyngeal and body-wall muscle defects. Transgenic copies of either mlc-1(+) or mlc-2(+) rescue all defects of mlc-1(0) mlc-2(0) double mutants. mlc-2 is redundant to mlc-1 in body-wall muscle, but mlc-2 performs a nearly essential role in the pharynx. Approximately 90% of mlc-2(0) hermaphrodites arrest as L1 larvae due to pharyngeal muscle defects. Lethality of mlc-2(0) mutants is sex specific, with mlc-2(0) males being essentially wild type. Four observations suggest that hermaphrodite-specific lethality of mlc-2(0) mutants results from insufficient expression of the X-linked mlc-1(+) gene in the pharynx. First, mlc-1(0) mlc-2(0) double mutants are fully penetrant L1 lethals in both hermaphrodites and males. Second, in situ localization of mlc mRNAs demonstrates that both mlc-1 and mlc-2 are expressed in the pharynx. Third, transgenic copies of either mlc-1(+) or mlc-2(+) rescue the pharyngeal defects of mlc-1(0) mlc-2(0) hermaphrodites. Fourth, a mutation of the dosage compensation gene sdc-3 suppresses hermaphrodite-specific lethality of mlc-2(0) mutants.

CONVENTIONAL myosins (myosins II) are hexameric, consisting of two myosin heavy chains (MHCs) and four myosin light chains (MLCs). One molecule of regulatory MLC (rMLC) and one of essential MLC are bound noncovalently to the neck region of each globular myosin head. Light chains provide structural support to the neck region of myosin, thereby amplifying conformational changes associated with myosin motor functions (RAYMENT et al. 1993 ; HOLMES 1997 ). Regulatory MLCs (rMLCs) are named so because they regulate activity of the myosin ATPase. In molluscan muscle, rMLCs inhibit interactions between myosin and actin. Calcium binding to myosin relieves this inhibition, thereby activating the myosin ATPase and provoking muscle contraction (SZENT-GYORGYI et al. 1973 ; KENDRICK-JONES et al. 1976 ). In vertebrate smooth muscle and nonmuscle cells, phosphorylation of certain rMLC residues by myosin light chain kinase (MLCK) activates the myosin ATPase, whereas phosphorylation of other residues by protein kinase C (PKC) inhibits the ATPase (SELLERS 1991 ; TAN et al. 1992 ; GALLAGHER et al. 1997 ). Phosphorylation of rMLCs by p34^cdc2 inhibits the myosin ATPase in nonmuscle cells and likely regulates the timing of cytokinesis by inhibiting contractile ring activity during mitosis (SATTERWHITE and POLLARD 1992 ).

In addition to their role in regulating myosin ATPase activity, rMLCs also regulate thick filament assembly in smooth muscle and nonmuscle cells (TRYBUS 1991 ; TAN et al. 1992 ). Phosphorylation of the rMLCs by MLCK destabilizes a folded myosin conformation, favoring an extended conformation that is competent for thick filament assembly (CRAIG et al. 1983 ). Although the extent to which the rMLCs function in thick filament assembly in vivo has not been established (SOMLYO et al. 1981 ; HOROWITZ et al. 1994 ), rMLC phosphorylation provides an appealing mechanism for rapid redistribution of myosin filaments in transient cellular functions.

MLCs appear to play a more limited role in vertebrate skeletal and cardiac muscle. Calcium regulation of the myosin ATPase is mediated by the thin filament troponin-tropomyosin complex (HOLMES 1995 ; GULICK and RAYMENT 1997 ). Actin-binding and actin-activated ATPase activity of skeletal myosin heavy chains stripped of all light chains is nearly normal (WAGNER and GINIGER 1981 ; SIVARAMAKRISHNAN and BURKE 1982 ; LOWEY et al. 1993 ). Such heavy chains can, however, inhibit myosin movement of actin filaments in vitro (LOWEY et al. 1993 ). Phosphorylation of rMLCs appears to increase isometric force and the rate of force production in skeletal muscle (SWEENEY et al. 1993 ).

Our current understanding of the rMLCs comes primarily from their in vitro manipulation. Genetic analysis will be important to reveal their in vivo functions. For example, the Drosophila muscle regulatory MLC gene, MLC-2, is required for normal myogenesis (WARMKE et al. 1992 ). MLC-2 null mutants undergo gastrulation, germ band extension, and germ band shortening but die late in embryogenesis, presumably due to defects in embryonic musculature. Myofibrillar structure of the indirect flight muscles is disrupted in MLC-2 null mutant heterozygotes, demonstrating that MLC-2 stoichiometry is important for the ultrastructure of this specialized muscle. MLC-2 constructs having substitutions at the MLCK phosphorylation sites rescue the lethality of MLC-2 null mutants (TOHTONG et al. 1995 ), consistent with a limited role of the rMLCs in skeletal muscle. Indirect flight muscle is structurally normal in such mutants, but their flight is impaired, and the power output of isolated flight muscle fibers is reduced (TOHTONG et al. 1995 ). The Drosophila nonmuscle rMLC gene, spaghetti-squash (squ), is essential for cytokinesis (KARESS et al. 1991 ). squ¹ mutants die as pupae with a large number of polyploid cells. Thus, this nonmuscle rMLC likely regulates myosin contractile ring functions. Mutants expressing squ alleles having altered activating phosphorylation sites are nearly indistinguishable from squ null alleles (JORDAN and KARESS 1997 ), which demonstrates an essential role of rMLC phosphorylation in nonmuscle cells. Dictyostelium rMLC null mutants (mlcR^-) have similar cytokinesis defects (CHEN et al. 1994 ). Myosin is localized aberrantly in mlcR^- cells, and purified mlcR^- myosin has abnormal disassembly properties and decreased ATPase activity in vitro (CHEN et al. 1994 ). Surprisingly, the cytokinesis defects of mlcR^- mutants are rescued by a mutant rMLC that cannot be phosphorylated, suggesting that rMLC phosphorylation plays a more limited role in Dictyostelium (OSTROW et al. 1994 ).

The nematode Caenorhabditis elegans, like many invertebrates, exhibits both actin-linked and myosin-linked regulatory systems (LEHMAN and SZENT-GYORGYI 1975 ; HARRIS and EPSTEIN 1977 ). The precise role, however, of these calcium regulatory systems in modulating actomyosin interactions in vivo remains unknown. Two C. elegans regulatory MLC genes, mlc-1 and mlc-2, were previously described (CUMMINS and ANDERSON 1988 ). MLC-1 and MLC-2 are nearly identical, being distinguished by a single conservative amino acid substitution. Despite this similarity, mlc-1 and mlc-2 have striking differences outside of their protein coding regions. For example, mlc-1 has a very long 3' untranslated region, whereas mlc-2 does not. mlc-2 is trans-spliced, but mlc-1 is not. Such differences in gene structure might reflect differences in their expression. To determine whether mlc-1 and mlc-2 have tissue-specific functions, and to establish their roles in muscle and possibly nonmuscle cells, we isolated mutations that eliminate each gene individually and both genes in combination. We describe here the isolation and phenotypic characterization of such mlc-1 and mlc-2 deletion mutants.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains and genetic analysis:
Conditions for growth and maintenance of C. elegans are described by BRENNER 1974 . sup-10(n184) deletes sup-10, mlc-1 (GREENWALD and HORVITZ 1980 ) and deletes ~70 kb of DNA between sup-10 and mlc-1 (C. WHITE, J. M. LEVIN, D. ALBERTSON, H. R. HORVITZ and P. ANDERSON, unpublished results). n184 does not affect mlc-2 (see Figure 1). sup-10(n245) deletes most of the region between mlc-1 and sup-10 (GREENWALD and HORVITZ 1980 ) but does not affect either mlc-1 or mlc-2. The "left" endpoint of n245 is ~500 bp rightward of the 3' end of the mlc-1 transcribed region (see Figure 1). We used n245 as a control to distinguish n184 phenotypes caused by deletion of mlc-1 from those caused by deletion of sup-10 or genes between mlc-1 and sup-10. Both n184 and n245, which were initially isolated as suppressors of unc-93(e1500), were outcrossed and segregated as single mutants before analysis. To examine whether dosage compensation mutations suppress the hermaphrodite-specific lethality of mlc2(r1133), we crossed r1133/0 males with sdc3(y129); unc-3(e151) hermaphrodites. sdc3(y129) mlc2(r1133) homozygotes were identified among the F2 progeny by their dumpy phenotype (y129 homozygotes are Dpy), by their failure to segregate Unc-3 offspring (unc-3 is linked to mlc-2), and by a PCR test diagnostic for r1133 homozygotes. sdc3(y129) is a weak sdc-3 allele that is semiviable when homozygous and exhibits a partial dosage compensation defect (DELONG et al. 1993 ).

View larger version (15K): In this window In a new window Download PPT slide   Figure 1. Map of the mlc-1 and mlc-2 genomic region. Boxes represent mlc-1 and mlc-2 exons, with unfilled regions indicating 5'- and 3'-untranslated regions. Arrows indicate the direction of mlc-1 and mlc-2 transcription. Like many C. elegans mRNAs (BLUMENTHAL and STEWARD 1997 ), the 5' end of mlc-2 mRNA is trans-spliced to SL1. rP2::Tc1 is an insertion of the transposable element Tc1 near mlc-2 (RUSHFORTH et al. 1993 ). The endpoints of deletions sup-10(n184), sup-10(n245), mlc-2(r1133), mlc-2(r1142), and mlc-1,2(r1141) are shown. Plasmids TR#234, which contains only mlc-1(+), and TR#233, which contains only mlc-2(+), were used for transformation rescue experiments.

Isolation and sequencing of mlc deletions:
We established multiple independent populations of strain TR1690 [genotype mut2(r459); dpy-19(n1347); rP2::Tc1] and screened them by PCR for spontaneous deletions in the vicinity of rP2::Tc1 (ZWAAL et al. 1993 ). mut2(r459) activates excision of Tc1 (COLLINS et al. 1987 ), and dpy19(n1347) marks the presence of mut2 (FINNEY 1987 ). rP2::Tc1 is a previously described insertion of the transposable element Tc1 (PLASTERK and VAN LUENEN 1997 ) into noncoding sequences 174 bp downstream of mlc-2 (RUSHFORTH et al. 1993 ). Amplification primers used to identify deletions flanked rP2::Tc1 and annealed outside of mlc-1 and mlc-2 coding regions. To amplify deletion-containing molecules, reactions were heated to 94° for 2 min and processed through 30 cycles of 94° for 0.5 min, 57° for 1 min, and 72° for 3 min, followed by a 10-min incubation at 72°. We subdivided populations in which deletion molecules were detected and isolated the mlc-2 or mlc-1,2 deletions using a sib-selection protocol, as previously described (RUSHFORTH et al. 1993 ). We sequenced the novel mlc deletion junctions using standard protocols after PCR amplification and cloning of appropriate genomic fragments. mlc2(r1133 and r1142) delete mlc-2 coordinates 1069 through 2590 and 44 through 2589, respectively. mlc-1,2(r1141) deletes mlc-1 nucleotide 1658 through mlc-2 nucleotide 2588 [coordinates of CUMMINS and ANDERSON 1988 ]. To reduce unrelated mutations, mlc deletions were outcrossed and resegregated eight times after their initial isolation. Lethal and semilethal mlc mutations were maintained as balanced heterozygotes using mnDp1(XV), which carries the right end of linkage group X [including mlc-1(+) and mlc-2(+)] attached to linkage group V (HERMAN et al. 1976 ).

RNA extractions and Northern analysis:
We prepared RNA as previously described (CUMMINS and ANDERSON 1988 ). For northern transfers, 16 µg of total RNA was denatured with glyoxal and dimethylsulfoxide and electrophoresed through 1.2% agarose gels in 0.01 M NaH₂PO₄ (pH 7.0). Samples were transferred to Zeta-Probe blotting membrane (Bio-Rad Laboratories, Richmond, CA) and hybridized with plasmids TR#115 and T7/T3-18-103 radiolabeled by primer extension of random hexanucleotides. Plasmid TR#115 contains a 5.8-kb genomic EcoRV-BglII fragment that includes all of mlc-1 and mlc-2 (CUMMINS and ANDERSON 1988 ). Plasmid T7/T3-18-103 (a gift of Mike Krause) contains a portion of the act-1 gene (KRAUSE et al. 1989 ) and was used as a normalization control.

Growth properties of mlc mutants:
All experiments were performed at 20°, unless otherwise noted. Brood sizes were measured by transferring hermaphrodites daily and counting their offspring. To quantify r1133 hermaphrodite survivorship, eggs collected during a 12- to 24-hr period were examined daily for a month, and those that did not develop beyond L1 were counted. To quantify r1133 male survivorship relative to that of hermaphrodites, r1133 or N2 males (10 to 15 per plate) were mated to groups of five dpy-11(e224); mlc-2(1133) hermaphrodites and transferred daily to fresh plates. Male and hermaphrodite offspring developing to at least L4 were counted over a 10-day period. Hatching times were measured by collecting eggs during a brief period and determining the length of time until they hatched as L1 larvae (BYERLY et al. 1976 ). The length of hermaphrodite postembryonic development was measured as the time from hatching until the onset of egg laying.

Transformation rescue:
Transgenic animals were generated by microinjecting plasmid DNAs into the distal gonadal syncytium of young adult N2 hermaphrodites (MELLO and FIRE 1995 ). Plasmid TR#233 contains a 3.7-kb EcoRV-ApaI genomic fragment that includes only mlc-2(+). Plasmid TR#234 contains a 4.5-kb SmaI-BglII genomic fragment that includes only mlc-1(+). These clones were coinjected with plasmid pRF4, which carries rol6(su1006) and provides a dominant marker indicating successful transformation (KRAMER et al. 1990 ). Regulatory MLC and pRF4 plasmid DNAs were coinjected into N2 animals at concentrations of 10 and 100 µg/ml, respectively. Multiple transgenic lines carrying each rMLC plasmid were established. In all cases, the transgenes behaved as extrachromosomal arrays. Transgenic males were crossed to mlc-2(r1133); mnDp1(XV)/+(V) or mlc-1,2(r1141); mnDp1(XV)/+(V) hermaphrodites, and roller male offspring were crossed again to the same hermaphrodite parent. Homozygous mlc-2(r1133) or mlc-1,2(r1141) transgenic lines were established from the progeny of these crosses. For each transgenic line, PCR was used to confirm the presence of the mlc deletion, the mlc plasmid DNA, and the absence of mnDp1.

In situ hybridization:
In situ hybridization to detect mlc-1 and mlc-2 mRNAs was performed as previously described (ALBERTSON et al. 1995 ). Digoxygenin-dUTP-labeled mlc-1 and mlc-2 antisense hybridization probes were prepared by asymmetric PCR (SEYDOUX and FIRE 1995 ). To generate transcript-specific probes, we selected amplification primers from the mlc-1 and mlc-2 unique 3'-untranslated regions. For the mlc-1 probe, we used primer MLC1-2498 (5'TGCACCACAAGTAACTGCTC3') to PCR amplify a ~500-nt fragment using plasmid TR#234 cut with BsaHI as a template. For the mlc-2 probe, we used primer MLC2-2398 (5'GCACTAATCCATTGAAAGAT3') to PCR amplify an 85-nt fragment using plasmid TR#233 cut with HpaII as a template. PCR reactions (25 µl) included 25 pmol of primer, 0.4 µg of linearized plasmid DNA, and 5.0 µl of DIG DNA labeling mix (Boehringer Mannheim, Indianapolis) containing 1 mM dATP, dCTP, and dGTP; 0.65 mM dTTP; and 0.3 mM digoxygenin-dUTP in a standard PCR-reaction buffer. Samples were heated to 100° for 5 min, 1.25 units of Taq polymerase was added, and samples were processed through 35 cycles of 94° for 0.5 min, 55° (mlc-1 probe) or 50° (mlc-2 probe) for 1 min, and 72° for 1 min, and completed by a 10-min incubation at 72°. Samples were precipitated and washed with ethanol, and resuspended in 300 µl of hybridization solution.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

sup-10(n184) deletes mlc-1:
mlc-1 and mlc-2 are located immediately adjacent to each other and are transcribed divergently, with their 5' ends separated by 2.6 kb (see Figure 1) (CUMMINS and ANDERSON 1988 ). We mapped mlc-1 and mlc-2 to the right end of the X chromosome by in situ hybridization to embryonic metaphase chromosomes and by Southern blot analysis of a series of X chromosome deficiencies and duplications (C. WHITE, J. M. LEVIN, D. ALBERTSON, H. R. HORVITZ and P. ANDERSON, unpublished results). We identified a deletion of mlc-1 as part of our molecular analysis of this chromosomal region. The muscle-affecting gene sup-10 is located ~70 kb rightward of mlc-1 (see Figure 1). Among 46 spontaneous and mutagen-induced sup-10 mutations that we analyzed, one of them, sup-10(n184), proved to be a deletion that removes both sup-10 and mlc-1 (see Figure 1). The left endpoint of sup-10(n184) breaks within the mlc-1 second exon, and most of the MLC-1 coding sequences are deleted by n184. As discussed below, sup-10(n184) exhibits only minor growth and reproductive abnormalities. As a control for these phenotypes, we analyzed sup-10(n245) (GREENWALD and HORVITZ 1980 ). sup-10(n245) deletes most of the region between mlc-1 and sup-10, but it does not affect either mlc-1 or mlc-2 (see Figure 1). The left endpoint of n245 is ~500 bp rightward of the mlc-1 transcribed region. sup-10(n245) provides a control to distinguish n184 phenotypes caused by deletion of mlc-1 from those caused by deletion of sup-10 or other genes located between mlc-1 and sup-10. The right endpoints of both n184 and n245 are located an unknown distance rightward of sup-10.

Isolation of mlc2 and mlc-1,2 deletions:
Since none of the characterized sup-10 mutations deleted both mlc-1 and mlc-2, and because we wanted to identify unambiguous mlc-2 null alleles, we isolated several mlc-2 and mlc-1 mlc-2 deletions by site-selected deletion of rP2::Tc1. rP2::Tc1 (P is polymorphism) is a Tc1 insertion 175 bp downstream of the mlc-2 transcribed region (RUSHFORTH et al. 1993 ). It was isolated as one of several Tc1 insertions within or near mlc-2, all of which proved to be non-null alleles of mlc-2 (see DISCUSSION). We adapted a site-selected deletion protocol (ZWAAL et al. 1993 ) in combination with sib-selection to identify deletions emanating from rP2::Tc1. We isolated three mlc deletions using this procedure (see Figure 1). mlc-2(r1133) and mlc-2(r1142) delete only mlc-2, whereas mlc-1,2(r1141) deletes both mlc-1 and mlc-2. To determine the precise deletion endpoints, we sequenced the novel junction of each mutation. mlc-2(r1133) and mlc-2(r1142) delete 1523 and 2546 bp, respectively (see MATERIALS AND METHODS). The deleted material includes all of mlc-2 and either 555 bp (r1133) or 1580 bp (r1142) of the intergenic region between mlc-2 and mlc-1. mlc-1,2(r1141) deletes 4245 bp that remove all of mlc-2 and the 5' half of mlc-1 (see Figure 1). Because mlc-1,2(r1141) is lethal when homozygous (see below), it was initially isolated and subsequently maintained as a heterozygote.

Northern analysis of mlc-1 and mlc-2 deletions:
Because mlc-1 and mlc-2 are transcribed divergently and because their 5' ends are separated by only 2.6 kb, regulatory elements important for their expression likely reside within the intergenic region. To establish that r1133, which deletes a portion of the intergenic region, does not affect expression of mlc-1(+) and that n184 does not affect expression of mlc-2(+), we measured by Northern blots the abundance of mlc mRNAs in mlc-2(r1133) and sup-10(n184) (see Figure 2). As expected, mlc-1 mRNA is absent in sup-10(n184), whereas the size and abundance of mlc-2 mRNA is normal (lane 3). Similarly, mlc-2 mRNA is absent in mlc-2(r1133), whereas the size and abundance of mlc-1 mRNA is normal (lane 2). sup-10(n245) affects neither mlc-1 nor mlc-2 mRNAs (lane 4). We conclude that sup-10(n184) and mlc-2(r1133) are true null alleles of their respective genes.

View larger version (35K): In this window In a new window Download PPT slide   Figure 2. Northern analysis wild-type (N2) and mlc deletions. The mlc-1 mlc-2 genomic region is diagrammed above an autoradiogram of a Northern blot hybridized with probes that detect mlc-1, mlc-2, and actin (control) mRNAs. Solid boxes of mlc-1 and mlc-2 correspond to protein-coding regions; open boxes correspond to 5' and 3' untranslated regions. The arrows indicate the direction of transcription.

Phenotype of mlc-1(0) mutants:
sup-10(n184) homozygotes [abbreviated below as mlc-1(0), where 0 indicates a null allele] are essentially wild type in phenotype. They exhibit normal motility (both larval and adult), egg laying, and pharyngeal pumping, which are phenotypes indicative of normal body-wall, vulval, and pharyngeal muscle function. As judged by polarized light microscopy, mlc-1(0) body-wall muscle ultrastructure is normal (see Figure 3). mlc-1(0) males exhibit abnormal mating behavior and only rarely succeed in cross-fertilizing hermaphrodites. However, the control deletion sup-10(n245), which is mlc-1(+) mlc-2(+) (see above), is similarly affected. Thus, the male mating defect of n184 is not due to deletion of mlc-1. Most aspects of mlc-1(0) growth, development, and reproduction are normal (see Table 1). The brood size of mlc-1(0) is about two-thirds that of N2, but sup-10(n245) is similarly affected. A small, but probably insignificant, proportion of mlc-1(0) offspring arrest development as either embryos or young larvae. The lengths of embryonic and postembryonic development of mlc-1(0) mutants are not significantly different from wild type at either 20° or 25°. Numerous indicators of embryonic and postembryonic nonmuscle actomyosin function, such as the multiple cell migrations required for normal gonadal morphogenesis, egg laying, and muscle cell positioning (ANTEBI et al. 1997 ), are normal in mlc-1(0) mutants.

View larger version (143K): In this window In a new window Download PPT slide   Figure 3. Wild-type (N2), mlc-2(r1133), and sup-10(n184) body-wall muscle. Each photograph is a polarized light micrograph of a single, adult, body-wall muscle cell. Representative A bands (A), I bands (I), and dense bodies (db) of wild type are shown.

Phenotype of mlc-2(0) mutants:
Although mlc-1(0) mutants are essentially wild type, deletion of mlc-2 is semilethal. mlc-2(r1133) homozygotes exhibit an Eat (eating abnormal) phenotype, and about 90% of r1133 hermaphrodites die as L1 or L2 larvae. Newly hatched r1133 larvae are fully elongated and morphologically normal (see Figure 4). They exhibit vigorous motility, indicative of near-normal body-wall muscle function, with occasional animals appearing slightly sluggish. Pharyngeal pumping, however, is irregular and feeble. The morphology and polarized light phenotype of r1133 pharynxes are normal, but r1133 larvae pump less frequently, their pumps are generally longer than normal, and the pharyngeal lumen frequently fails to open completely. Pumping in r1133 larvae is, however, somewhat variable. Feeble pumps are often interspersed with brief strong pumps, and pharyngeal pumping is nearly normal in a small number of animals. Arrested r1133 L1 larvae can remain alive and motile without significant growth for up to 4 wk. These animals gradually deteriorate, become less motile, and eventually die in a starved, sickly state. We conclude that mlc-2 performs an important and nearly essential function in pharyngeal muscle.

View larger version (89K): In this window In a new window Download PPT slide   Figure 4. Nomarski photomicrographs of wild- type (N2), mlc-2(r1133), sup-10(n184), and mlc-1,2 (r1141) L1 larvae.

Embryonic development of mlc-2(r1133) is approximately normal. A small but probably insignificant percentage (4.8%) of r1133 zygotes arrest as dead embryos. We did not investigate the terminal phenotypes of the few r1133 embryonic lethals. For those that hatch, the length of embryonic development is indistinguishable from wild type (see Table 1). Embryonic twitching of r1133 body-wall muscle is normal. r1133 homozygotes segregating from r1133/+ (heterozygous) or r1133/r1133 (homozygous) mothers are indistinguishable. Thus, maternally inherited MLC-2, if such exists, does not contribute significantly to the phenotype of mlc-2(0) mutants.

Remarkably, about 10% of mlc-2(r1133) homozygotes grow to be nearly normal adults (see Table 1). Such escapers exhibit normal motility and egg laying. Pharyngeal pumping in r1133 adults may be less frequent than that of wild type but only slightly so. As judged by polarized light microscopy, r1133 adult body-wall muscle ultrastructure is normal (see Figure 3), as is pharyngeal muscle, vulval muscle, and diagonal muscles of the male tail (data not shown). Although the rare r1133 adults are nearly normal, certain parameters of their growth and reproduction are abnormal (see Table 1). The brood size of r1133 adults is about two-thirds that of wild type. Postembryonic development of r1133 is protracted, requiring an average of 8 days from hatching to sexual maturity. The length of postembryonic development of r1133 is highly variable, with some individuals requiring 3 wk or more to reach sexual maturity. For most of this time, r1133 animals appear as thin, pale, and somewhat starved L1 larvae. Once animals grow to a size comparable to L2 or L3 larvae of wild type, however, they usually develop into healthy-looking adults within a day or two. As described above for mlc-1(0) mutants, several indicators of embryonic and postembryonic nonmuscle actomyosin function are apparently normal in r1133 escapers.

The semilethal phenotype of mlc-2(r1133) is sex specific. Whereas only 11% of r1133 hermaphrodites grow to adulthood, about 90% of r1133 males develop into normal adults. The high survivorship of r1133 males was evident as an unexpected sex ratio of adult crossprogeny when we crossed r1133 males with dpy-11(e224); mlc-2(r1133) hermaphrodites. Ninety percent of cross-progeny adults (279/310, recognized by their non-Dpy phenotype) were male. This deviation from an expected 1:1 sex ratio is due to frequent L1 arrest of cross-progeny hermaphrodites relative to cross-progeny males. When N2 control males were mated with dpy-11(e224); mlc-2(r1133) hermaphrodites, 41% (177/432) of cross-progeny adults were males, reflecting a high survival (but <100% survival) of r1133 males. r1133 males mate normally and cannot be distinguished from wild type.

The phenotype of mlc-2(r1133) is more severe at 25°. Indeed, r1133 is effectively lethal at 25°. Over two-thirds (21/30) of r1133 raised at 25° were sterile as adults. The remaining animals had few (<10) progeny each. r1133 adults raised at 25° have abnormal gonads and are sickly and lethargic compared to N2 raised at 25° or r1133 raised at 20°. Stocks of r1133 homozygotes cannot be propagated at 25°.

Dosage compensation contributes to hermaphrodite-specific lethality of mlc-2(0) mutants:
We investigated whether the hermaphrodite-specific lethality of mlc-2(r1133) results from sex-specific differences in mlc1(+) expression. One possible source of sex-specific differences in mlc-1 expression is dosage compensation. Dosage compensation reduces the rate at which individual X-linked genes are transcribed in hermaphrodites relative to males, thereby compensating for the fact that hermaphrodites contain two X chromosomes whereas males contain only one (MEYER 1997 ). We reasoned that perhaps dosage compensation reduces mlc-1(+) expression in hermaphrodites below the level normally found in males. In wild-type animals, which are mlc-1(+) mlc-2(+), sex-specific differences in mlc-1(+) expression may be of little consequence. In mlc-2(0) mutants, however, perhaps mlc-1(+) expression in the hermaphrodite pharynx is insufficient, resulting in the Eat phenotype that we observe.

To test this, we constructed a mlc-2(0); sdc-3 double mutant. Although sdc-3 is an essential gene, the weak (and therefore viable) allele sdc-3(y129) exhibits a partial defect in dosage compensation and increased transcription of X-linked genes in hermaphrodites (DELONG et al. 1993 ). Whereas mlc-2(r1133) cannot be propagated at 25°, the sdc-3(y129); mlc-2(r1133) double mutant can be maintained as a self-fertile stock at 25°. Although mlc-2(r1133); sdc-3(y129) grows slowly at 25°, its growth is not substantially different than that of sdc-3(y129) single mutants. Thus, sdc-3(y129) suppresses the hermaphrodite-specific lethality of mlc-2(r1133). This result does not establish that sdc-3 suppression of mlc-2(0) hermaphrodite lethality is due to elevated mlc-1(+) expression, but data presented below demonstrate that elevated expression of mlc-1(+) is sufficient to suppress r1133 lethality.

Phenotype of mlc-1,2(0) mutants:
mlc-1,2(r1141), which deletes both mlc-1 and mlc-2, is unconditionally lethal, exhibiting defects of both pharyngeal and body-wall muscle. r1141 homozygotes arrest as nearly paralyzed L1 larvae. Such animals are severely Unc, moving poorly or not at all. Motility of the anterior of r1141 larvae is slightly better than that of the posterior. Pharyngeal pumping of r1141 is irregular and feeble, similar to that described above for mlc-2(r1131). Arrested larvae persist without significant growth for as long as 4 wk. Embryonic development of r1141 is slightly protracted (15.9 hr ± 2.3, n = 9 compared to 13.2 hr ± 0.6, n = 19 for r1141/+ and +/+ scored together), but a high proportion of r1141 homozygotes hatch. Embryonic twitching of r1141 commences on schedule at about 1.5-fold elongation but is somewhat less vigorous than that of wild type. The motility defects of r1141 embryos become progressively more severe, with late embryos, which normally roll vigorously within the egg, being nearly paralyzed. r1141 homozygotes exhibit mild elongation defects. Hatched L1 larvae are not fully elongated and have morphological irregularities along their body length ("lumpy dumpy" phenotype; see Figure 4). The elongation defects of r1141, however, are relatively mild compared to the twofold arrest of severely muscle-defective mutants (WILLIAMS and WATERSTON 1994 ; MOERMAN and FIRE 1997 ). r1141 is fully recessive, and homozygotes exhibit approximately the same phenotype at both 20° and 25°. r1141-arrested larvae are slightly larger at 25° compared to 20°.

From the above results, we conclude that: (1) mlc-1 and mlc-2 perform redundant functions in body-wall muscles, (2) functions of mlc-1 are redundant to those of mlc-2 in pharyngeal muscle, but mlc-2 performs an important and nearly essential role in the pharynx, and (3) mlc-1 and mlc-2 likely perform redundant functions in vulval muscles. mlc-1(0) and mlc-2(0) single mutants do not exhibit egg-laying defects, and our expectation is that a mlc-1,2(0) deletion would be egg-laying defective. This cannot be directly demonstrated, however, due to the larval lethality of the mlc-1,2(r1141).

Transformation rescue of mlc-2(0) and mlc-1,2(0) mutants:
To confirm that the phenotypes described above are due to rMLC defects, and to investigate rMLC rescue of the mlc-2(0) and mlc-1,2(0) mutant phenotypes, we transformed genomic copies of either mlc-1(+) or mlc-2(+) (plasmids TR#233 and TR#234; see Figure 1) into N2 and established extrachromosomal arrays. Genomic mlc-1(+) or mlc-2(+) clones were coinjected with a plasmid that marks extrachromosomal arrays with a dominant allele of rol-6 (KRAMER et al. 1990 ). Extrachromosomal arrays were then crossed into mlc-2(r1133) and mlc-1,2(r1141) using the roller marker to identify array-containing cross-progeny, which were then scored for their mlc phenotypes. Both mlc-1(+) and mlc-2(+) individually rescued all phenotypes of both mlc-2(0) and mlc-1,2(0) mutants. Array-containing transformants have large brood sizes, high rates of hatching, and apparently normal postembryonic development. Motility, pharyngeal pumping, and egg laying of transgenic adults is normal at both 20° and 25°, as is the ultrastructure of body-wall muscle when viewed by polarized light microscopy (data not shown). Because of the high copy number of typical extrachromosomal arrays (STINCHCOMB et al. 1985 ), levels of expression of transgenic mlc-1(+) and mlc-2(+) are likely elevated relative to that of the endogenous genes. We conclude that either MLC-1 or MLC-2 is sufficient for all aspects of mlc-1 and mlc-2 function. The striking difference between the mlc-1(0) and mlc-2(0) mutant phenotypes, therefore, likely reflects differing levels of MLC-1 vs. MLC-2 expression in the hermaphrodite pharynx.

In situ localization of mlc-1 and mlc-2 mRNAs:
The results described above predict that both mlc-1 and mlc-2 are expressed in body-wall, pharyngeal, and possibly vulval muscles. To confirm this localization, we performed in situ hybridization to mlc-1 and mlc-2 mRNAs using gene-specific antisense probes. We prepared single-stranded probes from the unique 3' untranslated region of each gene. Hybridization of the mlc-2 probe to mlc-2(r1133) and the mlc-1 probe to sup-10(n184) yielded no signal (Figure 5A and Figure 5D, respectively), confirming that these probes are mRNA specific. Hybridization of mlc probes to wild type demonstrated that both mlc-1 and mlc-2 are expressed in body-wall muscles (Figure 5B and Figure 5E), pharyngeal muscles (Figure 5C and Figure 5F), and vulval muscles (Figure 5B and Figure 5G). Both mlc-1 and mlc-2 mRNAs are present in all regions of the pharynx (procorpus, metacorpus, isthmus, and terminal bulb). We did not detect mlc-1 or mlc-2 mRNA in several of the minor muscle groups, such as the intestinal, uterine, anal depressor, and sphincter muscles, but these tissues were not well-preserved by our fixation methods. Similarly, we did not detect mlc mRNAs in nonmuscle cells, although such expression might be below detectable levels.

View larger version (73K): In this window In a new window Download PPT slide   Figure 5. Localization of mlc-1 and mlc-2 mRNAs in mlc-1(0) and mlc-2(0) mutants. mlc-2(r1133) (A–C) and sup-10(n184) (D–G) were hybridized with antisense mlc-1-specific probes (B, C, and D) or mlc-2-specific probes (A, E, F, and G). Body-wall muscles (bw) are indicated with solid arrows in B and E. The procorpus (pc), metacorpus (mc), isthmus (i), and terminal bulb (tb) regions of the pharynx are labeled in C and F. Vulval muscles (v) are indicated with dotted arrows in B and with solid arrows in G.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

mlc-1 and mlc-2 encode nearly identical regulatory myosin light chains (CUMMINS and ANDERSON 1988 ). To investigate the in vivo functions of these rMLCs, we identified or isolated deletions that remove mlc-1 and mlc-2 individually and both genes in combination. A deletion that removes both mlc-1 and mlc-2 is lethal when homozygous and exhibits both body-wall and pharyngeal muscle defects. We conclude that rMLC function is required in both of these muscle types for normal contraction. Three observations demonstrate that mlc-1 and mlc-2 are functionally redundant in body-wall muscle. First, in situ hybridization shows that both mlc-1 and mlc-2 are expressed in body-wall muscle. Second, both mlc-1(0) and mlc-2(0) single mutants have normal body-wall muscle. The unaffected mlc gene in each mutant is sufficient for normal rMLC function. Third, transgenic copies of either mlc-1(+) or mlc-2(+) rescue the body-wall muscle defects of a deletion that removes both mlc-1 and mlc-2.

mlc-2 performs an important and nearly essential role in the pharynx. mlc-2(0) mutants have an Eat (eating abnormal) phenotype. Pharyngeal morphology and muscle birefringence of mlc-2(0) mutants is normal, but pharyngeal contractions are weak and attenuated. Motility of mlc-2(0) larvae is normal, but arrested larvae do not increase in size and eventually die of apparent starvation after several weeks. The phenotype of mlc-2(0) mutants is much more severe than that of our previously isolated mlc2::Tc1 insertions (RUSHFORTH et al. 1993 ). Despite having Tc1 insertions within mlc-2 exons, these previously isolated mutations are non-null alleles. They express 10–40% of the wild-type quantity of mlc-2 mRNA, with Tc1 having been removed by splicing (RUSHFORTH et al. 1993 ). Such splicing leaves small insertions and deletions within mature mlc-2 mRNAs, but their weak phenotype indicates that the altered MLC-2 proteins are functional.

Remarkably, about 10% of mlc-2(0) homozygotes grow into nearly normal adults. Such escapers grow slowly and appear starved as larvae, but adults are essentially indistinguishable from wild type. Other eating abnormal mutants also have phenotypes more severe in larvae than in adults (AVERY 1993 ). Presumably, ingesting, concentrating, and grinding food are more difficult for young larvae, whose pharynxes are smaller than those of adults. As the pharynx grows larger during larval development, mlc-2(0) mutants likely eat more effectively. The critical differences that cause 10% of mlc-2(0) mutants to survive are unknown. Perhaps escapers are simply larvae that express slightly more mlc-1(+) than their siblings.

Although function of mlc-2 is required in the pharynx, two observations suggest that mlc-1 and mlc-2 are partially redundant in the pharynx and that mlc-2(0) larvae arrest development due only to an insufficient quantity of MLC-1. First, in situ hybridization demonstrates that both mlc-1 and mlc-2 mRNAs are expressed in the pharynx. Second, expression of either MLC-1 or MLC-2 from transgenes is sufficient to rescue the pharyngeal defects of mlc-2(0) mutants. As judged by transformation rescue of mlc-2(0) and mlc-1,2(0) mutants, we detect no functional differences between mlc-1(+) and mlc-2(+). This result suggests that the single, conservative, amino acid substitution that distinguishes MLC-1 and MLC-2 is not significant with regard to MLC function. If it were significant, and if MLC2 performed a unique qualitative function, mlc-2(0) mutants should not be rescued by elevated expression of mlc-1(+). We believe, therefore, that the mlc-2(0) pharyngeal defects result from lowering, but not eliminating, expression of rMLCs in the pharynx. Similar quantitative effects have been observed in Drosophila, where null alleles of MLC-2, a muscle rMLC, exhibit a dominant flightless phenotype due to abnormalities of indirect flight muscles in heterozygotes (WARMKE et al. 1992 ). Perhaps the C. elegans pharynx is similarly sensitive to modest perturbations of rMLC levels.

The hermaphrodite-specific lethality of mlc-2(0) mutants may result from quantitative differences in mlc-1(+) expression in males vs. hermaphrodites. Whereas only about 10% of mlc-2(0) hermaphrodites survive to adulthood, nearly all mlc-2(0) males develop as normal adults. Elevated mlc-1(+) expression, such as likely occurs in the presence of mlc-1(+) transgenes or in a sdc-3(y129) genetic background, suppresses the hermaphrodite-specific lethality of mlc-2(0) mutants. Dosage compensation in C. elegans reduces the per-gene rate of X-linked transcription in hermaphrodites relative to males, such that the overall level of expression is approximately equal in the two sexes (MEYER 1997 ). Dosage compensation, however, may not be precise for every locus, and expression of certain X-linked genes may be different in hermaphrodites and males. mlc-1 may be one such gene. Pharyngeal expression of mlc-1(+) may be lower in hermaphrodites than in males. In wild-type animals, which are mlc-1(+) mlc-2(+), such sex-specific differences in gene expression would be of little consequence. In mlc-2(0) mutants, however, mlc-1(+) levels in the hermaphrodite pharynx may be insufficient. sdc-3(y129) is a weak (and therefore viable) allele that partially disrupts dosage compensation and increases expression of X-linked genes in hermaphrodites (DELONG et al. 1993 ). sdc-3(y129) suppresses the hermaphrodite-specific lethality of mlc-2(0) mutants at 25°. We presume that elevated mlc-1(+) expression is the basis of this suppression, but elevated expression of other X-linked genes might contribute to the suppression phenotype.

We find no compelling evidence that mlc-1 or mlc-2 functions in nonmuscle cells. Indeed, all mlc deletion phenotypes can be explained by their muscle defects. Actomyosins are known to be important for C. elegans embryogenesis (STROME and WOOD 1983 ; PRIESS and HIRSH 1986 ; HILL and STROME 1988 ; HIRD and WHITE 1993 ), but none of our mutants exhibit significant embryonic defects. A high proportion of mlc-2(0) and mlc-1(0) mlc-2(0) zygotes hatch as viable, albeit arrested, L1 larvae. The larval arrest of these mutants appears to be due exclusively to their pharyngeal defects. Although mlc-1(0) mlc-2(0) L1 larvae are incompletely elongated, this may be a secondary consequence of their body-wall muscle defects. Embryonic elongation occurs by the coordinated contraction of circumferential bundles of actomyosin filaments in hypodermal cells (PRIESS and HIRSH 1986 ). Such contractions squeeze the ovoid embryo into a long vermiform shape. Although the hypodermis is responsible for elongation, interactions between the hypodermis and the developing body-wall muscle (which underlies the hypodermis) are important for this process (MOERMAN and FIRE 1997 ). Indeed, severe defects of body-wall muscle cause embryonic elongation to arrest at the twofold stage (WATERSTON 1989 ; BARSTEAD and WATERSTON 1991 ; CHEN et al. 1994 ; WILLIAMS and WATERSTON 1994 ). The modest elongation defects of mlc-1(0) mlc-2(0) homozygotes, which are not as severe as those of the twofold arrest muscle mutants, may well be a secondary consequence of their muscle defects.

Numerous indicators of embryonic and postembryonic nonmuscle actomyosin function, such as the multiple cell migrations required for normal gonadal morphogenesis, egg laying, and muscle cell positioning (ANTEBI et al. 1997 ), are normal in mlc-1(0) mutants and in mlc-2(0) escapers. Unfortunately, such phenotypes cannot be scored in mlc-1(0) mlc-2(0) mutants due to their larval lethality. Thus, a role for MLC-1 and MLC-2 in postembryonic nonmuscle cells cannot be excluded. We note, however, that the C. elegans genome sequencing effort (HODGKIN et al. 1995 ) identifies a third probable rMLC (Swiss-Prot accession no. Q09510) with very high amino acid identity (73%) to Drosophila spaghetti-squash, a nonmuscle rMLC gene (KARESS et al. 1991 ). This probable rMLC is 46% identical to mlc-1 or mlc-2 and is, therefore, an excellent candidate for performing nonmuscle rMLC functions in C. elegans. Genetic analysis of this locus will be important for understanding the in vivo functions of regulatory myosin light chains in both muscle and nonmuscle cells.

	FOOTNOTES

¹ Present address: Center for Cancer Research, Department of Biology, Massachusetts Institute of Technology, Cambridge, MA 02139.

	ACKNOWLEDGMENTS

We thank Rolf Samuels, Rock Pulak, Frank Solomon, Barbara Meyer, and Norman Rushforth for their scientific input and for their comments on the manuscript, Bonnie Saari for technical assistance, Jim Kramer and Mike Krause for plasmids, Leon Avery for advice on scoring Eat phenotypes, and the University of Wisconsin Integrated Microscopy Resource for assistance with confocal microscopy. This work was supported by an individual research grant from the National Institutes of Health (GM-30132) to P.A. and by a National Institutes of Health Training Grant in Genetics awarded to the University of Wisconsin.

Manuscript received May 22, 1998; Accepted for publication August 7, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALBERTSON, D. G., R. M. FISHPOOL and P. S. BIRCHALL, 1995 Fluorescence in situ hybridization for the detection of DNA and RNA, pp. 339–364 in Caenorhabditis elegans: Modern Biological Analysis of an Organism, edited by H. F. EPSTEIN and D. C. SHAKES. Academic Press, San Diego.

ANTEBI, A., C. R. NORRIS, E. M. HEDGECOCK and G. GARRIGA, 1997 Cell and growth cone migrations, pp. 583–609 in C. elegans II, edited by D. L. RIDDLE, T. BLUMENTHAL, B. MEYER and J. R. PRIESS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

AVERY, L., 1993  The genetics of feeding in Caenorhabditis elegans.. Genetics 133:897-917[Abstract].

BARSTEAD, R. J. and R. H. WATERSTON, 1991  Vinculin is essential for muscle function in the nematode. J. Cell Biol. 114:715-724[Abstract/Free Full Text].

BLUMENTHAL, T., and K. STEWARD, 1997 RNA processing and gene structure, pp. 117–145 in C. elegans II, edited by D. L. RIDDLE, T. BLUMENTHAL, B. J. MEYER 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].

BYERLY, L., R. CASSADA, and R. RUSSELL, 1976  The life cycle of the nematode C. elegans I. Wild type growth and reproduction. Dev. Biol. 51:23-33[Medline].

CHEN, L., M. KRAUSE, M. SEPANSKI, and A. FIRE, 1994  The Caenorhabditis elegans MYOD homologue HLH-1 is essential for proper muscle function and complete morphogenesis. Development 120:1631-1641[Abstract].

CHEN, P., B. D. OSTROW, S. R. TAFURI, and R. L. CHISHOLM, 1994  Targeted disruption of the Dictyostelium RMLC gene produces cells defective in cytokinesis and development. J. Cell Biol. 127:1933-1944[Abstract/Free Full Text].

COLLINS, J., B. SAARI, and P. ANDERSON, 1987  Activation of a transposable element in the germ line but not the soma of Caenorhabditis elegans.. Nature 328:726-728[Medline].

CRAIG, R., R. SMITH, and J. KENDRICK-JONES, 1983  Light-chain phosphorylation controls the conformation of vertebrate non-muscle and smooth muscle myosin molecules. Nature 302:436-439[Medline].

CUMMINS, C. and P. ANDERSON, 1988  Regulatory myosin light chain genes of Caenorhabditis elegans.. Mol. Cell. Biol. 8:5339-5349[Abstract/Free Full Text].

DELONG, L., J. D. PLENEFISCH, R. D. KLEIN, and B. J. MEYER, 1993  Feedback control of sex determination by dosage compensation revealed through Caenorhabditis elegans sdc-3 mutations. Genetics 133:875-896[Abstract].

FINNEY, M., 1987 The genetics and molecular biology of unc-86, a C. elegans cell lineage gene. Ph.D. Thesis, Massachusetts Institute of Technology, Boston.

GALLAGHER, P. J., B. P. HERRING, and J. T. STULL, 1997  Myosin light chain kinases. J. Muscle Res. Cell Motil. 18:1-16[Medline].

GREENWALD, I. and H. HORVITZ, 1980  unc-93(e1500): a behavior mutant of Caenorhabditis elegans that defines a gene with a wild-type null phenotype. Genetics 96:147-164[Abstract/Free Full Text].

GULICK, A. M. and I. RAYMENT, 1997  Structural studies on myosin II: communication between distant protein domains. Bioessays 19:561-569[Medline].

HARRIS, H. and H. EPSTEIN, 1977  Myosin and paramyosin of Caenorhabditis elegans: biochemical and structural properties of wild-type and mutant proteins. Cell 10:709-719[Medline].

HERMAN, R. K., D. G. ALBERTSON, and S. BRENNER, 1976  Chromosome rearrangements in Caenorhabditis elegans.. Genetics 83:91-105[Abstract/Free Full Text].

HILL, D. P. and S. STROME, 1988  An analysis of the role of microfilaments in the establishment and maintenance of asymmetry in Caenorhabditis elegans zygotes. Dev. Biol. 125:75-84[Medline].

HIRD, S. N. and J. G. WHITE, 1993  Cortical and cytoplasmic flow polarity in early embryonic cells of Caenorhabditis elegans.. J. Cell Biol. 121:1343-1355[Abstract/Free Full Text].

HODGKIN, J., H. HORVITZ, and S. BRENNER, 1979  Nondisjunction mutants of the nematode Caenorhabditis elegans.. Genetics 91:67-94[Abstract/Free Full Text].

HODGKIN, J., R. H. PLASTERK, and R. H. WATERSTON, 1995  The nematode Caenorhabditis elegans and its genome. Science 270:410-414[Abstract/Free Full Text].

HOLMES, K. C., 1995  The actomyosin interaction and its control by tropomyosin. Biophys. J. 68:2-5.

HOLMES, K. C., 1997  The swinging lever-arm hypothesis of muscle contraction. Curr. Biol. 7:112-118[Medline].

HOROWITZ, A., K. M. TRYBUS, D. S. BOWMAN, and F. S. FAY, 1994  Antibodies probe for folded monomeric myosin in relaxed and contracted smooth muscle. J. Cell Biol. 126:1195-1200[Abstract/Free Full Text].

JORDAN, P. and R. KARESS, 1997  Myosin light chain-activating phosphorylation sites are required for oogenesis in Drosophila. J. Cell Biol. 139:1805-1819[Abstract/Free Full Text].

KARESS, R. E., X. J. CHANG, K. A. EDWARDS, S. KULKARNI, and I. AGUILERA et al., 1991  The regulatory light chain of nonmuscle myosin is encoded by spaghetti-squash, a gene required for cytokinesis in Drosophila.. Cell 65:1177-1189[Medline].

KENDRICK-JONES, J., E. SZENTKIRALYI, and A. SZENT-GYORGYI, 1976  Regulatory light chains in myosins. J. Mol. Biol. 104:747-775[Medline].

KRAMER, J. M., R. P. FRENCH, E. PARK, and J. J. JOHNSON, 1990  The Caenorhabditis elegans rol-6 gene, which interacts with the sqt-1 collagen gene to determine organismal morphology, encode a collagen. Mol. Biol. Cell 10:2081-2089.

KRAUSE, M., M. WILD, B. ROSENZWEIG, and D. HIRSH, 1989  Wild-type and mutant actin genes in Caenorhabditis elegans.. J. Mol. Biol. 208:381-392[Medline].

LEHMAN, W. and A. SZENT-GYORGYI, 1975  Regulation of muscular contraction: distribution of actin control and myosin control in the animal kingdom. J. Gen. Physiol. 66:1-30[Abstract/Free Full Text].

LOWEY, S., G. S. WALLER, and K. M. TRYBUS, 1993  Skeletal muscle myosin light chains are essential for physiological speeds of shortening. Nature 365:454-456[Medline].

MELLO, C., and A. FIRE, 1995 DNA transformation, pp. 451–482 in Caenorhabditis elegans: Modern Biological Analysis of an Organism, edited by H. F. Epstein and D. C. SHAKES. Academic Press, San Diego.

MEYER, B. J., 1997 Sex determination and X chromosome dosage compensation, pp. 209–240 in C. elegans 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., and A. FIRE, 1997 Muscle: structure, function, and development, pp. 417–470 in C. elegans II, edited by D. L. RIDDLE, T. BLUMENTHAL, B. J. MEYER and J. R. PRIESS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

OSTROW, B. D., P. CHEN, and R. L. CHISHOLM, 1994  Expression of a myosin regulatory light chain phosphorylation site mutant complements the cytokinesis and developmental defects of Dictyostelium RMLC null cells. J. Cell Biol. 127:1945-1955[Abstract/Free Full Text].

PLASTERK, R. H. A., and H. G. A. M. VAN LUENEN, 1997 Transposons, pp. 97–116 in C. elegans II, edited by D. L. RIDDLE, T. BLUMENTHAL, B. J. MEYER and J. R. PRIESS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

PRIESS, J. R. and D. I. HIRSH, 1986  Caenorhabditis elegans morphogenesis: the role of the cytoskeleton in elongation of the embryo. Dev. Biol. 117:156-173[Medline].

RAYMENT, I., W. R. RYPNIEWSKI, K. SCHMIDT-BASE, R. SMITH, and D. R. TOMCHICK et al., 1993  Three-dimensional structure of myosin subfragment-1: a molecular motor. Science 261:50-58[Abstract/Free Full Text].

RUSHFORTH, A. M., B. SAARI, and P. ANDERSON, 1993  Site-selected insertion of the transposon Tc1 into a Caenorhabditis elegans myosin light chain gene. Mol. Cell. Biol. 13:902-910[Abstract/Free Full Text].

SATTERWHITE, L. L. and T. D. POLLARD, 1992  Cytokinesis. Curr. Opin. Cell Biol. 4:43-52[Medline].

SELLERS, J. R., 1991  Regulation of cytoplasmic and smooth muscle myosin. Curr. Opin. Cell Biol. 3:98-104[Medline].

SEYDOUX, G., and A. FIRE, 1995 Whole-mount in situ hybridization for the detection of RNA in Caenorhabditis elegans embryos, pp. 323–337 in Caenorhabditis elegans: Modern Biological Analysis of an Organism, edited by H. F. EPSTEIN and D. C. SHAKES. Academic Press, San Diego.

SIVARAMAKRISHNAN, M. and M. BURKE, 1982  The free heavy chain of vertebrate skeletal myosin subfragment 1 shows full enzymatic activity. J. Biol. Chem. 257:1102-1105[Abstract/Free Full Text].

SOMLYO, A. V., T. M. BUTLER, M. BOND, and A. P. SOMLYO, 1981  Myosin filaments have non-phosphorylated light chains in relaxed smooth muscle. Nature 294:567-569[Medline].

STINCHCOMB, D., J. SHAW, S. CARR, and D. HIRSH, 1985  Extrachromosomal DNA transformation of C. elegans.. Mol. Cell. Biol. 5:3484-3496[Abstract/Free Full Text].

STROME, S. and W. WOOD, 1983  Generation of asymmetry and segregation of germ-line granules in early C. elegans embryos. Cell 35:15-25[Medline].

SWEENEY, H. L., B. F. BOWMAN, and J. T. STULL, 1993  Myosin light chain phosphorylation in vertebrate striated muscle: regulation and function. Am. J. Physiol. 264:1085-1095.

SZENT-GYORGYI, A. G., E. M. SZENTKIRALYI, and J. KENDRICK-JONES, 1973  The light chains of scallop as regulatory subunits. J. Mol. Biol. 74:179-203[Medline].

TAN, J. L., S. RAVID, and J. A. SPUDICH, 1992  Control of nonmuscle myosins by phosphorylation. Annu. Rev. Biochem. 61:721-759[Medline].

TOHTONG, R., H. YAMASHITA, M. GRAHAM, J. HAEBERLE, and A. SIMCOX et al., 1995  Impairment of muscle function caused by mutations of the phosphorylation sites in myosin regulatory light chain. Nature 374:650-653[Medline].

TRYBUS, K. M., 1991  Assembly of cytoplasmic and smooth muscle myosins. Curr. Opin. Cell Biol. 3:105-111[Medline].

WAGNER, P. and E. GINIGER, 1981  Hydrolysis of ATP and reversible binding to F-actin by myosin heavy chains free of all light chains. Nature 292:560-562[Medline].

WARMKE, J., M. YAMAKAWA, J. MOLLOY, S. FALKENTHAL, and D. MAUGHAN, 1992  Myosin light chain-2 mutation affects flight, wing beat frequency, and indirect flight muscle contraction kinetics in Drosophila.. J. Cell Biol. 119:1523-1539[Abstract/Free Full Text].

WATERSTON, R. H., 1989  The minor myosin heavy chain, mhcA, of Caenorhabditis elegans is necessary for the initiation of thick filament assembly. EMBO J. 8:3429-3436[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].

ZWAAL, R. R., A. BROEKS, J. VAN MEURS, J. T. GROENEN, and R. H. PLASTERK, 1993  Target-selected gene inactivation in Caenorhabditis elegans by using a frozen transposon insertion mutant bank. Proc. Natl. Acad. Sci. USA 90:7431-7435[Abstract/Free Full Text].

This article has been cited by other articles:

				S. L. Hooper and J. B. Thuma Invertebrate Muscles: Muscle Specific Genes and Proteins Physiol Rev, July 1, 2005; 85(3): 1001 - 1060. [Abstract] [Full Text] [PDF]

				D. GuhaThakurta, L. A. Schriefer, R. H. Waterston, and G. D. Stormo Novel transcription regulatory elements in Caenorhabditis elegans muscle genes Genome Res., December 1, 2004; 14(12): 2457 - 2468. [Abstract] [Full Text] [PDF]