programmed cell death (Shaham and Horvitz, 1996). In addition, these studies
indicated that ced-3 is likely to function downstream of ced-4 because the induction
of apoptosis by the overexpression of ced-3 does not require ced-4, whereas, in the
converse experiment, the induction of programmed cell death by ced-4 overexpression
requires the presence of ced-3 (Shaham and Horvitz, 1996) (Figure 2).
The following statements of our general understanding of programmed cell death
could be based on the studies described above. (i) The execution of programmed cell
death during development, which, formally, may also be considered as a somewhat
unusual terminal differentiation program, is subject to precise genetic control. (ii)
The execution of the apoptotic fate is an active program that can be best described
as cellular suicide; meaning a cell that senses that it is doomed to die participates
actively in this process by inducing (or helping to induce) its own demise. (iii) Finally,
all single cells may have the potential to undergo programmed cell death.
7.
egl-1, ced-3, ced-4, and ced-9 are functionally conserved
throughout evolutionPositional cloning of the above-mentioned genes revealed that all these nematode cell-
death genes have homologs in mammals that perform similar functions in the control
of apoptosis. ced-9 encodes a protein sharing 24% overall sequence identity with the
mammalian Bcl-2 oncogene, which, like ced-9, negatively regulates cell death
(Hengartner and Horvitz, 1994). As with ced-9, overexpression of bcl-2 protects cells
from dying, whereas bcl-2 (If) mutations make cells hypersensitive to death-inducing
signals (for reviews, see Huang, 2000; Ferri and Kroemer, 2001). ced-3 encodes a
protein with similarity to a family of death-inducing proteases called caspases (Yuan
et al., 1993). CED-4 is related to mammalian Apaf-1, which in mammals acts together
with caspase-9, caspase-3, and cytochrome c in the core apoptotic program (Yuan
and Horvitz, 1992; Li et al., 1997; Zou et al., 1997). Finally, EGL-1 is related to the
mammalian BH3-only-domain proteins, such as Bid, Bad, Bim, or Puma, all of which
are implicated in proapoptotic signaling (Yuan and Horvitz, 1992; Yang et al., 1995;
Wang, K. et al., 1996; Conradt and Horvitz, 1998; O’Connor et al., 1998; Nakano
and Vousden, 2001; Yu et al., 2001; Wu and Deng, 2002).
Interestingly, of all the mentioned C. elegans cell-death genes, only one functional
family member is encoded within the worm genome. C. elegans contains only one
CED-9/bcl-2-like protein and only one CED-4/apaf-1 look-alike. In contrast, C.
elegans contains, besides CED-3, two additional caspases that do not seem to function
in cell-death regulation (1998) (Shaham, 1998). Similarly, besides EGL-1, there is
one additional protein encoded in the worm genome, containing a BH3 sequence
motif (1998). The function of this protein is not known. It is currently still unknown
whether cytochrome c and its release from mitochondria play a role in C. elegans cell-
death regulation. The fact that there are two cytochrome c copies encoded in the
worm genome and the fact that cytochrome c is an essential gene hindered genetic
approaches to determine a potential proapoptotic function of cytochrome c in worm
168 GENETICS OF APOPTOSIS