unc-94 Encodes a Tropomodulin in Caenorhabditis elegans

Abstract

unc-94 is one of about 40 genes in Caenorhabditis elegans that, when mutant, displays an abnormal muscle phenotype. Two mutant alleles of unc-94, su177 and sf20, show reduced motility and brood size and disorganization of muscle structure. In unc-94 mutants, immunofluorescence microscopy shows that a number of known sarcomeric proteins are abnormal, but the most dramatic effect is in the localization of F-actin, with some abnormally accumulated near muscle cell-to-cell boundaries. Electron microscopy shows that unc-94(sf20) mutants have large accumulations of thin filaments near the boundaries of adjacent muscle cells. Multiple lines of evidence prove that unc-94 encodes a tropomodulin, a conserved protein known from other systems to bind to both actin and tropomyosin at the pointed ends of actin thin filaments. su177 is a splice site mutation in intron 1, which is specific to one of the two unc-94 isoforms, isoform a; sf20 has a stop codon in exon 5, which is shared by both isoform a and isoform b. The use of promoter-green fluorescent protein constructs in transgenic animals revealed that unc-94a is expressed in body wall, vulval and uterine muscles, whereas unc-94b is expressed in pharyngeal, anal depressor, vulval and uterine muscles and in spermatheca and intestinal epithelial cells. By Western blot, anti-UNC-94 antibodies detect polypeptides of expected size from wild type, wild-type-sized proteins of reduced abundance from unc-94(su177), and no detectable unc-94 products from unc-94(sf20). Using these same antibodies, UNC-94 localizes as two closely spaced parallel lines flanking the M-lines, consistent with localization to the pointed ends of thin filaments. In addition, UNC-94 is localized near muscle cell-to-cell boundaries.

Introduction

Sarcomeres are specialized actin cytoskeletal structures that perform the work of muscle contraction. Sarcomeres are molecular or “nano”-machines consisting of a highly ordered assemblage of many proteins. Despite ever-increasing knowledge of the components and functions of sarcomeric proteins, we still do not have a clear picture about how sarcomeres are assembled and how sarcomeres are maintained in the face of repeated muscle activity. The ability to analyze mutants in the nematode Caenorhabditis elegans is being exploited to obtain insights into these questions.1., 2., [3] Molecular genetic experiments using C. elegans complement the typical biochemical analyses that are carried out in mammalian systems. The major striated muscle in the worm lies in the body wall and is used for locomotion. In adults, there are 95 spindle-shaped cells that are divided among four quadrants, which lie just underneath a basement membrane, hypodermis and cuticle. Due to the optical transparency of the worm, the myofibrils can be viewed by polarized light which reveals obvious striations. Bright A-bands alternate with dark I-bands, and each I-band contains a row of dense bodies, which are the analogs of Z-disks of vertebrate striated muscle. Because the striations lie at a slightly oblique angle with respect to the long axis of the worm, this muscle is “obliquely striated.” Rather than filling the entire cell, as is the case for vertebrate striated muscle, in C. elegans body wall muscle, the myofibrils are restricted to a narrow zone of ∼ 1.5 μm along one side of the cell. All the M-lines and Z-disks are attached to the muscle cell membrane, making these structures good models for studying muscle costameres⁴ and focal adhesions of non-muscle cells.³

C. elegans has been used profitably to obtain mutants defective in the formation, function and/or structure of muscle. There are two major classes of muscle-affecting mutations. In one class, the uncoordinated or “Unc” class, the worms develop into adults but are slow moving or paralyzed.5., 6., [7] The second class, the paralyzed arrested at twofold (“Pat”) class of mutants, displays a characteristic embryonic lethality in which embryos do not move within the eggshell and stop development at the twofold stage.⁸ A few genes have both hypomorphic Unc and null Pat phenotypes; examples include unc-112⁹ and unc-45.¹⁰ To date, nearly all the muscle Unc and many of the Pat genes have been cloned and studied at the molecular level. The encoded proteins include both previously known and novel components of the thick and thin filaments and their organizing and membrane attachment structures (M-lines and dense bodies). The cloning of M-line and dense body components has revealed not only a number of familiar components of focal adhesions (perlecan, integrins, vinculin, integrin-linked kinase, PINCH), but also new components of these structures (UNC-112, UNC-98, UNC-96 and UNC-89).[3], [11] This analysis is consistent with a model in which myofibril assembly is directed by signals first laid down in the extracellular matrix and the muscle cell membrane. In addition, most of the components of dense bodies and M-lines are shared, except for the proteins involved in the later stages of assembly, for example, for the dense bodies, vinculin and α-actinin, and for the M-lines, UNC-89.

For thick filaments, the expected genes for the myosins and paramyosin were among the first worm muscle genes identified. New and evolutionarily conserved components of thick filaments were first revealed by this genetic analysis. One example is twitchin, the founding member of the giant kinases that include mammalian titin and insect projectin.12., [13], 14. Another example is UNC-45, which is a conserved chaperone for myosin head folding and for the assembly of myosin into thick filaments.10., 15. For thin filaments, genes encoding the actins and the troponin–tropomyosin complex have been found. In addition, novel components have been identified (e.g., UNC-87),¹⁶ and roles in myofibril assembly were first revealed for several previously known proteins. There are many actin-binding proteins, many of which regulate actin filament dynamics in many eukaryotic cells. This includes proteins that promote actin polymerization (e.g., profilin), severing (e.g., gelsolins, ADF/cofilin), stability (e.g., tropomyosin), depolymerization (e.g., ADF/cofilin), barbed end capping (e.g., capZ) or pointed end capping (e.g., tropomodulin). Molecular genetic analysis of UNC-60B (an ADF/cofilin protein),17., [18] tropomyosin¹⁹ and UNC-78,20., 21. has clearly demonstrated the requirement for regulating actin filament dynamics to ensure proper assembly and maintenance of muscle thin filaments.

In 1980, Zengel and Epstein reported results of a screen for mutants with altered body wall muscle structure.⁷ Their screen involved enrichment for slow-moving worms, followed by assessment by polarized light. Mutants represented new alleles of 10 previously identified genes5., 6. and 7 new genes. Among them was a single mutant allele for a new gene, unc-94. unc-94 (su177) was described as slow moving and by polarized light microscopy to have “irregular birefringent areas.” Electron microscopy (EM) showed large collections of thin filaments interspersed with possible intermediate filaments and patches of thick filaments. Here, we report that unc-94 encodes a tropomodulin, an F-actin pointed-end capping protein. Our results show the expected localization of a tropomodulin in the sarcomere of another animal and its in vivo importance, and point to a new role for tropomodulin in regulating F-actin at muscle cell–to-cell boundaries.