Acute myeloid leukemia (AML) is a heterogeneous disease presenting with various clinical subtypes. A broad spectrum of cytogenetic abnormalities and molecular mutations has been documented, including mutations in genes that are involved in epigenetic regulation. The recent advances in whole genome and exome sequencing have led to the identification of several novel defects, including misssense mutations that substitute arginine residues in isocitrate dehydrogenases (IDH) 1 and 2, in approximately 15 percent of patients with AML. The mutant enzymes exhibit a neomorphic gain of function effect that produces an oncometabolite, 2-hydroxyglutarate (2-HG), which inhibits several enzymes that regulate the epigenetic signature of the genome.
The precise mechanisms linking these IDH1/2 mutations to AML pathogenesis have not been fully elucidated, but they are acquired very early in the multistep transformation process, which makes these enzymes attractive drug targets. Small-molecule inhibitors of the mutant enzymes have been developed, but Dr. Steven Chan from the laboratory of Dr. Ravi Majeti with colleagues from Stanford University, explored an alternative approach of synthetic lethality to target mutant cells. This concept exploits the fact that malignant cells rely on a non-oncogenic pathway for their survival. They used AML cell lines that had been engineered to express either the wild type IDH1 or the mutant IDH1R132H together with a green fluorescent protein marker under the control of a doxycycline-inducible promoter. They performed a large-scale RNA interference study by transducing the cells with a lentiviral shRNA library and identified two synthetic lethal genes, BCL-2 and BCL-W, which are members of the antiapoptotic BCL-2 family. Knockdown of BCL-2 decreased the viability of IDH1R132H cells compared with wild type cells, and the level of knockdown correlated with the degree of apoptosis, consistent with an on-target effect. Further experiments showed that increasing the intracellular levels of 2-HG by treatment with octyl-R-2-HG, sensitized cells to apoptosis. Subsequent studies were facilitated by the availability of a highly specific, clinically available BCL-2 inhibitor, ABT-199, which was used to recapitulate the synthetic lethal phenotype. In vitro studies on the engineered AML cell lines, as well as ex vivo experiments on blasts from AML patients, showed that cells expressing mutant IDH were more sensitive to the effects of ABT-199. Transplant and xenotransplant experiments in mice demonstrated further that the drug targeted IDH1R132H AML cells, as well as leukemic stem cells in vivo.
These results prompted the researchers to investigate the mechanism of the synthetic lethal effect. The antiapoptotic BCL-2 protein binds to the BH3 domain of proapoptotic proteins BAX and BAK, which prevents them from initiating apoptosis by permeabilizing the outer mitochondrial membrane and triggering the release of cytochrome c into the cytosol. ABT-199 is a BH3 mimetic, which could disrupt the BCL-2/BAX/BAK interaction and induce apoptosis. Treatment of mutant IDH cells or cells with experimentally elevated levels of 2-HG exhibited increased levels of cytochrome c and enhanced mitochondrial membrane depolarization, compared to wild type and untreated cells, confirming their hypothesis. The researchers provided evidence that the synthetic lethal effect was not due to an imbalance between pro-and anti-apoptotic BCL proteins, nor was it related to excessive oxidative stress, and it did not require epigenetic changes. They investigated the mitochondrial electron transport chain and found that 2-HG inhibited cytochrome c oxidase (COX), which correlated with a decreased activity of the enzyme in mutant IDH cells. Treatment of cells with other COX inhibitors confirmed that COX inhibition sensitized cells to ABT-199 treatment.
The researchers propose a model whereby IDH1/2 mutations cause an accumulation of 2-HG, which inhibits COX and increases the dependency on BCL-2 to bind and inactivate BAX/BAK, thereby preventing mitochondrial depolarization and apoptosis. Treatment of mutant IDH AML cells with ABT-199 disrupts the complex and triggers apoptosis.
In Brief
These findings are important for several reasons. First, they demonstrate that IDH1/2 mutations may contribute to leukemia pathogenesis not only by modulating the epigenome, but also by dysregulating mitochondria. Second, the data provide evidence for the mechanism of a synthetic lethal approach of treating AML by inhibiting BCL-2, which has important therapeutic implications. Treatment of AML patients with BCL-2 inhibitors has produced encouraging results, but the response has varied widely between patients. Third, the current study indicates that patients with IDH1/2 mutations should respond favorably to ABT-199 treatment, and thus, it provides an important marker for a personalized approach to cancer chemotherapy. Fourth, these data raise the possibility that a combination of ABT-199 with inhibitors that target COX or other enzymes of the electron transport chain may offer a novel strategy for treating AML patients who have normal IDH1/2.
Competing Interests
Dr. Coetzer indicated no relevant conflicts of interest.
