An HLH-l reversion screen Lihsia Chenl 3, Michael Krause2, and Andrew Firel I Carnegie Institute of Washinton, Baltimore, MD 21210.2NIH-Laboratory of Molecular Biology, NIDDK, Bethesda, MD 20890. 3The Johns Hopkins University, Dept of Biology, Baltimore, MD 21218. The MyoD family of transcription factors has been implicated in the determination and differentiation of vertebrate skeletal muscles. A single homologue of the MyoD family, designated hlh-l, was identified in C. elegans; this gene is expressed in body wall muscle cells and their clonal precursors(l). Animals homozygous for an hlh-l null mutation (
cc450) die as larvae or young adults, suffering muscle contraction and morphogenesis defects. Despite these defects, the hlh-l (
cc450) mutants appear to contain normal numbers of body wall muscles with surprisingly well-organised myofilament structure(2). Elucidating the role of HLH-l will entail the identification of its regulatory targets and of the factors that promote myogenesis in its absence. We have taken both biochemical (cDNA subtraction screening) and genetic approaches to this question. This article describes a genetic reversion screen undertaken to identify compensating regulatory mutations, affecting either a definitive HLH- 1 target or a parallel myogenic pathway. The starting strain used in the screen was homozygous for hlh-l (
cc450), and was rescued by an extrachromosomal array carrying a handicapped hlh-l gene with the last two exons deleted (the array also carries the dominant visible marker
rol-6(sul O06)). While the array allows viability and fertility in the
cc450 background, the rescue is incomplete, with resulting animals exhibiting poor movement and slow growth. Potential revertants can be easily identified in this screen as animals that exhibit improved movement and growth. There are five types of mutations that might give the above phenotype: 1) intragenic reversion at the hlh-l locus; 2) extragenic suppressors which bypass the requirement for hlh-l; 3) amber suppressors (
cc450 is an amber mutation) and potentially other informational suppressors, which might allow HLH-l protein expression by the
cc450 mutant allele; 4) mutations in the transgene array which allow improved expression or function of the handicapped hlh-l gene; and 5) chromosomal mutations which improve the rescued phenotype of array containing animals. We were most interested in whether mutations of class 2 could be isolated (it should also be noted that class 5 might also include loci with direct roles in myogenic regulation). To bias ourselves against mutations in the transgene array, we chose animals that no longer express a strong Rol-6 phenotype. Following EMS mutagenesis, we isolated 10 strains that displayed improved movement (from 30,660 chromosomes). Three of these strains proved to be amber suppressors; this gives an indication of the completeness of the screen, since these are likely to represent anticodon transitions in one of the four known amber suppressors previously shown to suppress
cc450. Of the remaining seven strains identified in our screen, four still contained the extrachromosomal array, suggesting that the array was still required for viability of the strains. In one other strain, the array has integrated into a chromosome. We don't know for these five strains whether the array or an endogenous locus has mutated to allow the improved rescue. Two revertants which do not require the transgene DNA were recovered. One contained a semi-dominant suppressing mutation that is tightly linked to hlh-l; this strain may be a second site intragenic revertant (the
cc450 mutation is still present). While rescue appears to be complete in this strain, the animals have longer bodies than wild type. The second strain contains an unlinked semi-dominant suppressor. Rescue in this strain is only partial. Both strains stain strongly with antibodies against the HLH-l protein. We are currently trying to determine in these two strains if the HLH-l antibody recognises HLH-1 protein or a (as yet hypothetical) cross-reacting HLH-1 homologue which has been overexpressed. The screen described above was an extensive search for mutations that can produce healthy animals capable of fairly good movement in the
cc450 background. The strain used, however, did not allow for the identification of mutations that might less completely suppress defects in
cc450 mutants. We are currently carrying out additional genetic screens that should enable us to identify such mutations. 1. Krause et al. (1990). Cell 63, 907-919. 2. Chen et al. (1994). Development 120, 1631-1641.