as a GEF (Gumienny et al., 2001). In vitro assays with purified components did not
reveal an associated GEF activity. Furthermore there is no Rac-GEF domain in
ELMO, DOCK180, and CrkII. However, coimmunoprecipitation experiments with
overexpressed human ELMO1and DOCK180 revealed a Rac GEF activity
(Gumienny et al., 2001). It is therefore likely that apoptotic signals induce the
formation of an active trimeric ELMO/DOCK180/CrkII complex, which, in turn,
becomes associated with a yet to be identified Rac-GEF activity, leading to Rac
activation (Gumienny et al., 2001).
13.
Potential functions of programmed cell death in C. elegansGiven the fact that C. elegans mutants defective in apoptosis remain viable and do
not seem to have any gross deleterious effects in development, fertility, and behavior,
we must ask what the biologic function is of programmed cell death in C. elegans
(Ellis and Horvitz, 1986). If we consider developmental programmed cell death only,
careful analysis reveals that mutant animals develop more slowly, have a slightly
reduced fertility, and are affected in complex behavioral tasks such as chemotaxis.
Although this lack of an overall phenotype provides advantages for the propagation
of mutant animals, it leaves us with the question of the function of cell death during
the development of C. elegans. The lack of gross alterations in the behavior of animals
containing about 12% more cells due to the absence of apoptotic death seems
especially surprising, as many of the ‘undead cells,’ although they never further divide,
terminally differentiate (Ellis and Horvitz, 1986). Some of them, at least partially,
take over the fate of cells closely related to them by lineage An illustrative example of
such a process is provided by the ‘undead’ sister of the M4 neuron (Avery and Horvitz,
1987). The M4 neuron is needed for the proper operation of the pharynx in young
animals, and its killing (as in laser microsurgery) leads to the starvation of operated
animals. However, if the M4 cell is ablated in animals defective for apoptosis, its
undead sister cell can take over its function and rescue the animals from death by
starvation (Avery and Horvitz, 1987). As a matter of fact, the best case for the
importance of programmed cell death during worm development might be made by
evolutionary considerations. Morphogenetic modeling, as accomplished by dooming
specific cells to undergo apoptosis, might be an efficient evolutionary strategy to
remodel morphologic structures. Indeed, the death of a single cell can account for
major anatomical changes. For example, the gross difference between the one-armed
gonad of Panagrellus revidus and the two-armed gonad of C. elegans can be explained
by the death of the posterior distal tip cell in P. revidus (DTCs direct the formation
of gonadal arms) (Sternberg and Horvitz, 1981).
In view of the importance of programmed cell death occurring in the germ line,
the various germ-line intrinsic cell-death pathways have to be considered separately.
Spontaneous germ-cell death was supposed to occur because dying germ cells might
function as nurse cells for oocytes, a hypothesis supported by the following
observations (Gumienny et al., 1999). Within the germ line, mitotic and meiotic
PROGRAMMED CELL DEATH IN C.ELEGANS 181