that leads to mitochondrial disruption and release of a cocktail of prodeath cofactors
(such as cytochrome c and Smac/DIABLO) into the cytoplasm (Li, H. et al., 1998;
Gross et al., 1999b). The interaction of the released cytochrome c with Apaf-1 results
in a nucleotide-dependent conformational change that allows binding of procaspase-9
through N-terminal caspase recruitment domains (CARD) present on both molecules
(Vaux, 1997). The formation of the procaspase-9/Apaf-1/cytochrome c complex (also
called the ‘apoptosome’) promotes the transcatalytic cleavage and scaffold-mediated
transactivation of caspase-9 (Srinivasula et al., 1998; Stennicke et al., 1999).
Caspase-9 activates further downstream caspases such as caspase-3 and caspase-7,
thereby amplifying the caspase cascade and promoting apoptosis (Deveraux et al.,
1998). Therefore, BID represents a mechanistic link between death receptor-induced
activation of caspase-8 (the ‘extrinsic’ pathway) and the mitochondrial activation of
caspases-9 and -3 (the ‘intrinsic’ or ‘mitochondrial’ pathway) (Roy and Nicholson,
2000).
Caspase-8-induced cleavage of BID is instrumental for death receptor-induced
apoptosis in certain cell types (termed ‘type II’) which show weak DISC formation
and therefore depend upon mitochondrial activation of caspase-9 to amplify the
caspase cascade (Scaffidi et al., 1998). Studies with BID-deficient mice indicate that
BID is required for CD95-induced apoptosis in hepatocytes, but not in thymocytes
or fibroblasts (Yin et al., 1999). Studies of BAX-deficient, BAK-deficient, or BAX/
BAK-deficient mice suggest that BAX and BAK play essential, yet mutually
redundant, roles in death receptor-mediated apoptosis in hepatocytes, but are not
required for CD95-induced death of thymocytes or fibroblasts (Lindsten et al., 2000;
Wei et al., 2001). While BAX-/-/BAK-/- hepatocytes resist CD95-induced apoptosis,
hepatocytes from either BAX-/- or BAK-/- mice remain sensitive to death receptor-
induced apoptosis. In contrast, recent studies using colon carcinoma lines that have
wild-type BAX and their isogenic BAX-deficient sister clones have demonstrated that
BAX may be required for death receptor-induced apoptosis in cancer cells (Burns and
el Deiry, 2001; Deng et al., 2002; LeBlanc et al., 2002; Ravi and Bedi, 2002). These
reports indicate that although BAX is dispensable for apical death receptor signals,
i nc l u d i ng a c t i v a ti o n o f c as p a s e - 8 a nd c l e av a g e o f B I D , i t i s n e c e s s a ry f o r mi to c h o nd ri al
activation of caspase-9 and induction of apoptosis in response to Apo2L/TRAIL,
CD95/Fas, or TNF-α (LeBlanc et al., 2002; Ravi and Bedi, 2002). These data suggest
that the basal expression of BAK in these cells cannot substitute for BAX in mediating
death receptor-induced apoptosis of tumor cells. However, upregulation of BAK
expression by exposure to the chemotherapeutic agents, etoposide or irinotecan, is
associated with sensitization of BAX-/-cancer cells to death receptor-induced apoptosis
(LeBlanc et al., 2002). Therefore, tBID may employ BAX or BAK for mitochondrial
activation of apoptosis in a cell-type- and death signal-specific manner.
Since loss of BAX and BAK confer long-term resistance to death receptor-induced
apoptosis, mitochondrial disruption appears to be critical for induction of apoptosis
in type II cells. However, activation of the caspase-9/Apaf1 complex and caspase-3
is not the only mechanism by which mitochondrial disruption results in apoptosis
(Cheng et al., 2001). Absence of these downstream effectors provides only transient
DEATH RECEPTORS IN APOPTOSIS 7