transmembrane protein, and worm ced-8 does indeed localize to membranes
(Stanfield and Horvitz, 2000). However, it is unclear how ced-8 impinges on the
regulation of the core apoptotic pathway. Another component that modulates
programmed cell death was discovered via its interaction with ced-4. Upon
overexpression of mac-l, which encodes an AAA type ATPase, the level of
programmed cell death is reduced (Wu et al., 1999b). Unfortunately, it could not be
determined whether mac-1 disruption leads to elevated levels of programmed cell
death, as mac-1 is essential for development. nuc-1 encodes a worm DNAse II ortholog
that was implicated in efficient DNA fragmentation during programmed cell death
(Lyon et al., 2000). nuc-1 is also needed for the degradation of DNA in the digestive
tract of C. elegans (Lyon et al., 2000; Wu et al., 2000). However, it seems that several
other Dnases must also be involved in DNA degradation (see below). With respect
to DNA fragmentation, corpses from nuc-1 animals proceed to an intermediate stage
that, in contrast to wild-type animals, allows the detection of apoptotic corpses due
to accumulation of DNA 3’-hydroxyl ends via the Tunnel method. nuc-1-mediated
DNA fragmentation is partially affected by corpse engulfment genes (see below), as
Tunnel staining is reduced in ced-1 and ced-7 corpse engulfment-defective mutants.
In contrast to the situation in mammals, no CPAN/CAP caspase-activated nuclease/
caspase-activated DNAse nuclease activity has been defined in C. elegans.
For identification of further targets of ced-3 caspase, a very elegant genetic screen
was initiated. Assuming there might be redundancy in the cell-death pathway after
the caspase step, Parrish et al. (2001) devised a sensitized genetic system for the
efficient detection of mutations that enhance weak cell-death defects. As part of the
experimental strategy, an activated ced-3 caspase, as well as a GFP reporter, is expressed
in six nonessential mechanosensory neurons. Enhancement of cell death can be easily
detected by a reduced number of surviving mechanosensory cells (reduced number
of GFP dots). One mutant (cps-6) identified is likely to encode the worm ortholog
of endonuclease G (Parrish et al., 2001). As with ced-8 endonuclease G, inactivation
enhances the weak cell-death defects of weak ced-3 and ced-4 mutations, suggesting
that endonuclease G might act downstream of the ced-3 caspase (Parrish et al., 2001).
It is not clear whether endonuclease G is cleaved by ced-3. Interestingly, worm
endonuclease G localizes to mitochondria, like its mammalian otholog, which also
has been implicated in programmed cell death (Li et al., 2001; Parrish et al., 2001).
It will be interesting to see whether this genetic screen will uncover additional
downstream components of the cell-death pathway.
10.
The regulation of programmed cell death in specific cell typesIn contrast to our advanced understanding of the molecular nature of the core
apoptotic pathway, very little is known about the genes that commit specific subsets
of cells to apoptotic death (recent developments concerning the regulation of
programmed cell death in the germ line will be discussed below). In principle, there
are two different classes of mutants that might affect cell type-specific deaths. Mutants
PROGRAMMED CELL DEATH IN C.ELEGANS 171