Programmed cell death is an essential process that has been conserved through evolution and is required both for normal development and for the maintenance of cellular homeostasis in all animals. Genetic analyses of C. elegans led to the isolation of mutants defective in various aspects of programmed cell death and allowed the establishment of a genetic pathway composed of four distinct, sequential stages: the specification of which cells live and which die, the actual killing process, engulfment, and the degradation of cellular remains (1). Three genes,
ces-1,
ces-2 (ces, cell-death specification), and
egl-1 (egl, egg-laying abnormal), determine whether a particular cell will live or undergo programmed cell death and for this reason have been postulated to be involved in cell-death specification (2, 3). Mutations in these genes affect the cell-death fates of only a subset of cells. By contrast, mutations in the genes
ced-3,
ced-4, and
ced-9 (ced, cell death abnormal), which are involved in the actual killing process, affect most if not all cell deaths that occur during C. elegans development (4, 5). Loss-of-function (lf) mutations in
ced-3 or
ced-4 or gain-of-function (gf) mutations in
ced-9 block most programmed cell deaths and result in the survival of cells normally destined to die. By contrast, loss-of-function mutations in
ced-9 cause most cells to die, including the ones that are normally destined to survive, resulting in embryonic lethality as a consequence of ectopic cell death. This lethality is suppressed by loss-of-function mutations in
ced-3 or
ced-4, indicating that
ced-3 and
ced-4 act downstream of or in parallel to
ced-9 (5). Of the cell-death-specification genes,
ces-1 and
ces-2 have been cloned. Both are predicted to encode transcription factors (Metzstein et al; 10th International C. elegans Meeting 1995, pg. 376). We have begun analyzing the third presumptive cell-death-specification gene,
egl-1. Four dominant and three semi-dominant alleles of
egl-1 were isolated previously, and analyses with deficiencies spanning the
egl-1 locus suggested that these alleles are gain-of-function mutations. Two observations suggested that
egl-1 functions upstream of the actual cell-death machinery during cell-death specification. First,
egl-1(gf) mutations affect only a specific class of neurons, the hermaphrodite-specific neurons (HSNs), which are required for egg-laying. In
egl-1(gf) animals the HSNs undergo programmed cell death, resulting in an egg-laying (Egl) defect (3, 6). Second, the deaths of the HSNs in
egl-1(gf) animals are suppressed by loss-of-function mutations in
ced-3 and
ced-4 and by the gain-of-function mutation in
ced-9 (4, 5). As reported previously, to determine the Egl-l (lf) phenotype and to facilitate the cloning of the gene by transformation rescue, we performed an F1 screen for revertants of the Egl-1(gf) phenotype and isolated a group of seven allelic revertants that are closely linked to the
egl-1 locus (B. Conradt and B. Horvitz; 10th International C. elegans Meeting 1995, pg. 162). In addition to being non-Egl as a result of HSN survival, these revertants have a general block in programmed cell death (a Ced phenotype), as do
ced-3(lf),
ced-4(lf), and
ced-9(gf) mutations. This Ced phenotype is recessive and has been used to map one of the suppressor mutations (
n3082) to LGV between
him-5 and
unc-112, the previously determined site of
egl-1 (3, 6, M. Hengartner, personal communication). To determine where in the cell-death pathway the gene defined by
n3082 acts, we have analyzed whether this allele can suppress the
ced-9(lf) phenotype. In contrast to
ced-3(lf) and
ced-4(lf) mutations,
n3082 does not suppress the lethality caused by a
ced-9(lf) mutation, indicating that like
egl-1, the
n3082 gene acts upstream of or in parallel to
ced-9. Our results from genetic mapping and epistasis analysis suggest that the gene defined by
n3082 and the six additional mutations isolated in our
egl-1(gf) reversion screen are intragenic suppressors within the
egl-1 gene. We propose that the Egl-1(lf) phenotype is a block in cell death and that
egl-1 is a new component of the general cell-death machinery required for most if not all cell deaths that occur during C. elegans development. We are currently searching the region between
him-5 and
unc-112 for RFLPs to map
n3082 with respect to the physical map. (1) Horvitz, H. R., Shaham, S., and M. O. Hengartner. (1994). CSH Symp. Quant. Biol. Vol. LIX, 377-385. (2) Ellis, R. E. and H. R. Horvitz. (1991). Development 112, 591-603. (3) Trent, C., Tsung, N., and H. R. Horvitz. (1983). Genetics 104, 619-647. (4) Ellis, H. and H. R. Horvitz. (1986). Cell 44, 817-829. (5) Hengartner, M., Ellis, R., and H. R. Horvitz. (1992). Nature 356, 494-499. (6) Desai, C. and H.R. Horvitz. (1989). Genetics 121, 703-721.