We have isolated a cDNA clone from a human fibroblast cDNA library that contains the entire protein-coding region of a 1.1-kilobase mRNA. This mRNA encodes a 284-amino acid tropomyosin, the primary structure of which most closely resembles smooth muscle tropomyosin. Thus, the expression of both 284-amino acid muscle-type and 247-amino acid non-muscle-type tropomyosins appears to be a normal feature of human non-muscle cells. We also present evidence to suggest that this cytoskeletal tropomyosin and a human skeletal muscle P-tropomyosin are derived from a common structural gene by an alternative RNA splicing mechanism.Tropomyosins are proteins that were first isolated from skeletal muscle (1) but which, like actin, are also found in other types of muscle and most non-muscle tissues (2). In skeletal muscle, tropomyosin serves to mediate the effect of Ca2+ on the actin-myosin interaction. It does so, not by binding Ca2' directly but, through interaction with the troponins, one of which (troponin T) binds to tropomyosin at a specific site in the carboxyl-terminal region of the skeletal muscle tropomyosin molecule (3, 4). In smooth muscle and non-muscle tissues, which lack troponins, tropomyosin has a characteristic and different carboxyl-terminal primary structure (5, 6).

No abstract

We have identified a strain of Caenorhabditis elegans in which the transposable element Tcl is genetically active. Most spontaneous mutations affecting the unc-54 myo-sin heavy chain gene of C. ekgans variety Bergerac are due to insertions of Tcl within unc-54. The Bergerac genome contains an unusually high number of Tcl elements, but this is not responsible for transpositional activity. Another variety of C. ekgans, strain DH424, contains an equally high number of Tcl elements, but transpositions are not detected. Tcl insertion mutations are genetically unstable. They revert to unc-54+ in both germ-line and somatic cells. Germ-line revertants are wild type and contain precise or nearly precise excisions of Tcl. Somatic revertants are genetic mosaics; they contain small patches of revertant muscle tissue in otherwise mutant animals. The pattern of mosaicism often allows us to know when and where during muscle development the excisions occur. Somatic reversion can be over 1000-fold more frequent than germ-line reversion. Most, if not all, organisms contain within their genomes multiple copies of discrete DNA sequences that are capable of transposition to many different chromosomal sites (for review, see ref. 1). These DNA sequences usually range in size from several hundred to several thousand base pairs and collectively are termed transposable genetic elements, or transposons. Transposons are responsible for a variety of genetic phenomena in these organisms, but an understanding of the biological roles of transposons in nature is only beginning to emerge. One important feature of transposable elements is their ability to induce mutations. Insertion of a transposon into (or nearby) a structural gene can disrupt that gene and lead to a mutant phenotype. Mutations of this type were first recognized in maize and systematically described by McClintock (ref. 2; for review, see ref. 3). They since have been documented in many different organisms. Spontaneous mutations (those that arise in the absence of any known mutagenic treatment) are often caused by transposon insertion. This is true for organisms as diverse as E. coli, maize, yeast, Drosophila, and mice (4-14). We have described the gene structures of 65 spontaneous mutations affecting the unc-54 gene of Caenorhabditis elegans, variety Bristol (15). unc-54 encodes one of two myosin heavy chain genes expressed in body-wall muscle cells (16). We were surprised to find that none of these spontaneous mutations is due to transposon insertion within unc-54. Yet, the nematode strain used for these studies (C. elegans variety Bristol) contains within its genome-25 dispersed copies of a DNA sequence, designated Tcl, whose structure is typical for that of a transposable element (17, 18). Two other nematode strains (C. elegans varieties Bergerac and DH424) contain >250 dispersed copies of Tcl in their genome (17, 18). We thought that Tcl might be transpositionally active in these strains because of its high copy number. The Bergerac strain contains a mutator system that exhi...

We have isolated a cDNA clone from a human skeletal muscle library which contains the complete protein-coding sequence of a skeletal muscle a-tropomyosin. This cDNA sequence defines a fourth human tropomyosin gene, the hTMa gene, which is distinct from the hTMnm gene encoding a closely related isoform of skeletal muscle a-tropomyosin. In cultured human fibroblasts, the hTMa gene encodes both skeletal-muscleand smooth-muscle-type a-tropomyosins by using an alternative mRNA-splicing mechanism.

We have determined the sequence of a 2.5 kb mRNA in human fibroblasts encoding a 248 amino acid cytoskeletal tropomyosin. The protein product of this mRNA is TM30nm, one of five tropomyosin-like proteins in human fibroblasts. The structural gene encoding this mRNA can also produce a 1.3 kb mRNA encoding a 285 amino acid skeletal muscle alpha-tropomyosin by tissue-specific alternative mRNA splicing. However, the multiple RNA-copy pseudogenes of this gene family are derived largely if not exclusively from transcripts processed according to the pattern observed in non-muscle cells.

The trk oncogene is a human transforming gene generated by the fusion of tropomyosin gene sequence to a truncated tyrosine kinase receptor gene. We have now characterized the normal tropomyosin gene from which the trk oncogene is derived. At least two different transcripts are expressed by this gene using a tissue-specific alternative messenger RNA splicing mechanism: a 2.5-kilobase (kb) mRNA encoding a 248-amino-acid tropomyosin in human fibroblasts and a 1.3-kb mRNA encoding a 285-amino-acid tropomyosin in human skeletal muscle. The rearrangement which generates the trk oncogene preserves most of the tropomyosin-coding sequences of the normal gene, including exons alternatively spliced in muscle and non-muscle tissue. We therefore expect the trk oncogene to show a tissue-specific pattern of transforming activity. Correct expression of the trk oncogene can occur only in non-muscle tissues. In muscle tissue the oncogene would almost certainly be inactive, as splicing according to the alternative muscle pattern aborts synthesis of the tyrosine kinase domain.

No abstract

No abstract