Programmed Cell Death Mechanisms & Implications

The Bredesen laboratory focuses on the biochemical mechanisms that control programs of cell death, with special interest in the roles that these processes may play in neurodegeneration, cancer, and aging. We have identified a novel class of cell death receptors dubbed "dependence receptors", since these create states of cellular dependence or addiction: when expressed, these receptors induce programmed cell death unless bound by ligand, which leads to a block of the death induction (Rabizadeh et al., 1993; Bredesen et al., 1998). We have found that the mechanism by which these receptors create cellular states of dependence on their respective ligands is by forming complexes that include proteases (typically caspases, the pro-apoptotic cysteine aspartyl proteases), then being cleaved by the proteases, resulting in the generation of cytotoxic peptides (Mehlen et al., 1998; Bredesen et al., 1998; Forcet et al., 2001). These receptors may play roles in a number of disease states, such as neurodegenerative diseases and cancer. Furthermore, the cytotoxic peptides generated have served as a model to generate novel anti-cancer peptides that are targeted to angiogenic vasculature (Ellerby et al., 1999).

Examples of dependence receptors include the androgen receptor (Ellerby et al., 1999), the common neurotrophin receptor p75NTR (Rabizadeh et al., 1993), the netrin-1 receptor DCC (Mehlen et al., 1998), the glial-derived neurotrophic factor (GDNF) receptor RET (Bordeaux et al., 2000), the unc5 (uncoordinated gene 5 from C. elegans) homologue UNC5H2, as well as others. Mutations in RET are associated with both neoplasia and Hirschprung's disease, the latter of which results from loss of enteric ganglion cells, resulting in lack of peristalsis. In accord with the theory of cellular dependence, the mutants of RET associated with neoplasia have a markedly reduced pro-apoptotic effect, whereas those associated with Hirschprung's disease retain the pro-apoptotic effect but lack the apoptosis inhibition normally occurring with ligand binding (Bordeaux et al., 2000).

During our studies of cell death-inducing receptors, we noted that one of these # the insulin-like growth factor I receptor (IGF1R) # induces a novel form of programmed cell death that does not display any of the criteria of apoptosis (Fig. 1, Sperandio et al., 2000). We dubbed this form of programmed cell death paraptosis, from para (related to) and apoptosis (literally, "falling away," a form of programmed cell death), since we believe that it is a form of cell death that is related to, but clearly distinct from, apoptosis. We subsequently identified a ligand-receptor pair that induces paraptosis (Castro-Obregon et al., 2002), providing further support for the notion that paraptosis is a form of programmed cell death. Interestingly, paraptosis may be a forerunner of apoptosis, since morphologically similar forms of cell death occur in organisms such as Dictyostelium that evolved prior to the appearance of caspases, which drive the apoptotic program (Sperandio et al., 2000). Furthermore, work from several laboratories has shown recently that the neuronal cell death associated with degenerative diseases such as Huntington's disease and amyotrophic lateral sclerosis does not display the morphological characteristics of apoptosis, and thus alternative programs may be involved. Since paraptosis is not blocked by inhibitors of apoptosis (such as caspase inhibitors), we believe that it will be important to identify inhibitors of non-apoptotic cell death programs to complement currently existing inhibitors of apoptosis.

Our studies of paraptosis induced by IGF1R have identified a novel cell death pathway that is capable of initiation by IGF1R as well as by a variety of insults. Since the IGF1R signaling pathway has been implicated in aging based on mutational analysis in C. elegans, we are interested in comparing the signal transduction pathway required for programmed cell death (paraptosis) induction by IGF1R to that required for "programmed organismal death" (aging?) induction by IGF1R.

Another cell surface molecule that requires cleavage for cell death induction is the Alzheimer's disease-associated gene product, amyloid precursor protein (APP). In collaborative studies with the laboratory of E.H. Koo, we found that APP is not only cleaved to produce -amyloid peptide, but is also cleaved to produce a second cytotoxic peptide, C31 (Lu et al., 2000; Galvan et al., 2002). Interestingly, C31 appears to be part of an Alzheimer's cell death complex (somewhat analogous to the Fas death-inducing signaling complex), the other members of which we are currently attempting to identify. One potential cell death pathway we have identified in Alzheimer's disease involves the activation of PKR, a stress-associated kinase (Fig. 2 in Peel et al., 2001; Peel et al., submitted for publication). Furthermore, there is growing evidence that misfolded proteins play a role in the neurodegenerative process, and therefore we have investigated the mechanisms by which misfolded proteins trigger cell death programs. Results from these studies have defined a new intrinsic pathway for apoptosis (Fig. 3 in Rao et al., 2001; Rao et al., 2002).

Thus the role(s) of cell death programs in neurodegeneration, and of putative organismal death programs in the aging process, is incompletely defined, and the ongoing studies are designed to address these emerging relationships. To this end, we have developed a new approach to the analysis of gene expression microarray data, which identifies crucial nodes controlling given biochemical processes (del Rio et al., 2001), and a novel approach to identifying protein-protein interactions (Fig. 4 in Kurakin and Bredesen, 2002) called TAIS (http://www.buckinstitute.org/TAIS/index.html).