Mismatch Repair

Loss of DNA mismatch repair (MMR) leads to mutator phenotypic cells (Jiricny, 2006), that lack G2 arrest responses (Davis et al, 1998) and are resistant to alkylating agents in terms of cell death and long-term survival. Conversely, MMR-proficient cells respond to MNNG-treatment by inducing G2 cell cycle arrest responses and cell death, typically by inducing apoptosis (Meyers et al, 2001; Meyers et al, 2004). Thus, along with correcting DNA mismatches and 1-2 base pair loops, MMR also acts to eliminate damaged cells by signaling apoptotic responses. The signaling responses leading from MMR-dependent detection of DNA damage to alterations in mitochondrial functions that cause intrinsic cell death pathways remain ill-defined. Our MMR research group are studying the exact retrograde (nucleus to cytoplasm) signaling mechanisms responsible for these MMR-dependent cell death pathways.

Role of c-Abl in MMR-Dependent G2 responses

(A) Knockdown of c-Abl increased the survival of MMNG-treated colon-cancer cells. (B) Model of the role of c-Abl-p73-alpha and CADD45-alpha in G2 arrest and cell death (apoptosis).

In response to N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) alkylation, DNA mismatch repair (MMR) mediates an active G2 cell cycle checkpoint arrest. MMR-deficient cells fail to detect DNA lesions, lack G2arrest responses and consequently have high mutation rates. To elucidate MMR-dependent signal transduction responses involved in G2 arrest responses, isogenic MMR-proficient and -deficient RKO cells were analyzed. Contrary to prior reports, ATR/Chk1 signaling was not specifically triggered by MMR, since identical Chk1 phosphorylation patterns in response to MNNG were noted regardless of MMR status. Furthermore, ATR inhibition by conditional expression of dominant-negative ATR in MMR-proficient cells did not alter G2 arrested responses, while ultraviolet light-induced, ATR Chk1 phosphorylation was prevented.

In contrast, disrupting c-Abl activity using STI571 (GleevecTM, a c-ABL inhibitor) or specific knock-down using small hairpin RNA specific for c-Abl (shRNA-c-Abl) abolished MNNG-induced, MMR-dependent G2 arrest responses, p73-alpha stabilization, and Gadd45-alpha protein expression. Importantly, c-Abl inhibition greatly increased the survival of MNNG-exposed, MMR-proficient cells to a level comparable to MMR-deficient cells, demonstrating that initiation of G2arrest responses, apoptosis and long-term survival triggered by MMR signaling required the function of this proto-oncogene. 

Conversely, inhibiting c-Abl activity did not affect the more subtle MMR-independent, ATR/Chk1-dependent G2 arrest in response to MNNG. These data strongly suggest that MMR-dependent G2 arrest responses triggered after MNNG exposures are dependent on c-Abl activity. Caution should, therefore, be taken in using therapies targeting c-Abl, since increased mutator phenotypes may be an unwanted consequence.

Role of c-Abl in MMR-Dependent Cell Death Responses

Figure 2
MMR-mediated apoptosis in cells exposed to MMNG. (I) MMR-dependent apoptotic cells are visible. (II) Re-expression of the human mutL protein stabilizes hPMS2 and corrects MMR deficiency. Corrected MMR cells elicit an apoptotic cell death that correlates with long-term survival loss.

In response to MNNG alkylation, cells with functional DNA mismatch repair (MMR) stimulate G2 cell cycle checkpoint arrest and induce intrinsic apoptotic responses. In contrast, MMR-deficient cells fail to detect MNNG-induced damage, lack these responses, and have elevated mutation rates. The exact retrograde (nuclear to cytoplasm) signaling mechanisms responsible for these MMR-dependent cell death pathways remain undefined. Since elevated p53 phosphorylation and stabilization were noted specifically in MNNG-treated MMR-proficient and not in MMR-deficient cells, we investigated the role(s) of this tumor suppressor protein in MMR-induced apoptosis. Loss of p53 function by expression of E6, dominant-negative p53, or stable p53 knockdown using specific small hairpin RNA (shRNA-p53) failed to prevent MNNG-induced, MMR-dependent apoptosis (Figure 2), demonstrating that p53 was not an effector of

MMR-dependent cell death. Dramatic MMR-dependent increases in p73-alpha levels after MNNG exposure prompted us to examine the role of this p53 homolog in apoptosis. Stable shRNA knockdown of p73-alpha blocked MNNG-mediated, MMR-dependent apoptosis and increased cell survival after MNNG. Since c-Abl can regulate p73-alpha levels, STI571 prevented MMR-dependent apoptosis. Furthermore, shRNA knockdown of c-Abl prevented MNNG-induced p73-alpha stabilization, c-Abl kinase activity, and abrogated MMR-dependent apoptosis. Thus, MMR-dependent intrinsic apoptosis in response to MNNG occurred independent of p53, but was dependent onhMLH1-c-Abl-p73-alpha retrograde signaling. Thus, caution should be taken when blocking c-Abl activation during the treatment of cancers with monofunctional alkylating agents, such as Temolozomide.

This work was funded by grant DE-FG02-06ER64186 from the Department of Energy and a grant from the National Cancer Institutes.