cell-death signaling. Similarly, it has not been analyzed whether the worm homolog
of mammalian AIF (apoptosis-inducing factor) is implicated in worm programmed
cell death. C.elegans also encodes two homologs of the IAP (inhibitor of apoptosis)
protein family (Fraser et al., 1999; Speliotes et al., 2000). These proteins, which were
initially found as bacolovirus proteins able to prevent cell death of host cells, are also
implicated as antiapoptotic proteins in Drosophila. However, the two C. elegans IAP
proteins do not seem to play a role in the regulation of programmed cell death. Indeed,
C. elegans BIR-1 is involved in the regulation of cytokinesis and chromosome
segregation similarly to the budding and fission yeast counterparts of this protein
family (Fraser et al., 1999; Speliotes et al., 2000). In addition, the C. elegans genome
does not encode any of the known components required for mammalian receptor-
mediated, cell-death signaling. In summary, based on the finding that C. elegans has
only one component of the core apoptotic pathway, C. elegans might contain the
evolutionarily most simple cell-death pathway. Genome-sequencing and extensive
EST projects with Protozoa and sponges will reveal whether these organisms also
contain those four components of the core apoptotic pathway. It is also possible that
there exist other, still unidentified components of the core programmed cell-death
pathway in the worm. These putative genes might have eluded discovery because they
are essential or because they act in redundant pathways.
8.
EGL-1, CED-3, CED-4, and CED-9 are part of a molecular
framework regulating programmed cell deathBiochemical studies indicate that the EGL-1, CED-9, CED-4, and CED-3 are part
of a molecular pathway that regulates programmed cell death. The pattern of
interactions between those proteins suggests a relatively simple molecular model
(Figure 4 ). In cells that are supposed to survive, the proapoptotic activity of CED-4
and CED-3 is kept at bay by the binding of CED-4 to CED-9. CED-9 is localized
at the outer mitochondrial membrane (Chen, F. et al., 2000). At present, it is not
clear whether CED-9 localization at the mitochondrial membrane is indeed necessary
for antiapoptotic CED-9 activity. Likewise, it is unclear whether permeabilization of
the outer mitochondrial membrane is necessary for the induction of programmed cell
death, as is the case in mammalian cells (Hengartner, 2000). According to the model,
apoptosis is initiated, at least during the execution of programmed cell death during
development, by the transcriptional induction of the BH3-only-domain protein
EGL-1 (Conradt and Horvitz, 1998; Conradt and Horvitz, 1999). Upon
transcriptional induction, EGL-1 is supposed to bind to CED-9, thereby releasing
CED-9-bound CED-4 (Conradt and Horvitz, 1998; del Peso et al., 2000; Parrish et
al., 2000). The above model is supported by the in vitro interaction patterns of those
proteins, as well as by the finding that egl-1 is transcriptionally induced in many cells
that are destined to die (Conradt and Horvitz, 1999). It should be noted that models
describing C. elegans and mammalian cell-death activation differ with respect to
CED-9 CED-4 interaction, as there is mounting evidence that the homologs of the
PROGRAMMED CELL DEATH IN C.ELEGANS 169