Grantee: Duke University, Durham, NC, USA
Researcher: Hai Yan, M.D., Ph.D.
Grant Title: Role of IDH1/2 mutations in cell metabolism, differentiation and gliomagenesis
Program Area: Researching Brain Cancer
Grant Type: Research Award
Amount: $450,000
Year Awarded: 2011
Duration: 3 years
Gliomas are the most common primary brain tumors. Included are astrocytomas and oligodendrogliomas, which are classified by the WHO as Grade I to Grade IV [1]. Grade IV glioblastoma (GBM), the highest grade astrocytoma and the one with the worst prognosis, can be further separated into primary and secondary GBM. Primary GBM appears de novo, with no prior history of a lower grade tumor, whereas secondary GBM develops following a prior diagnosis of a lower grade astrocytoma. Combined, Grade II, Grade III, and secondary Grade IV gliomas are collectively referred to as progressive gliomas. Recently, a large-scale genome-wide analysis was completed in GBM tumor samples revealing the presence of mutations in IDH1 at the codon for amino acid residue 132 in 12% of GBMs studied [2]. Further studies revealed heterozygous mutations of either isocitrate dehydrogenase 1 (IDH1) at residue 132 or of isocitrate dehydrogenase 2 (IDH2) at the homologous residue 172 in 60-90% of progressive gliomas, including secondary glioblastomas, but not primary glioblastomas [3-9]. The high frequency of these mutations in progressive gliomas is strongly suggestive of a key role in gliomagenesis.
Nonetheless, it remains unclear what the role of these mutations in the formation of gliomas is. Normally, IDH1 and IDH2 (collectively referred to as IDH1/2 in this application) catalyze the conversion of isocitrate to α-ketoglutarate while reducing NADP+ to NADPH. Whereas IDH1 localizes to the cytosol and peroxisomes [10-12], IDH2 localizes to the mitochondria [13]. The potential consequences of IDH1/2 mutations are two-fold. First, mutations in IDH1R132 and IDH2R172 may lead to a decrease in wild-type activity [5, 6, 14]. Secondly, these mutations lead to a novel function- namely, the conversion of α¬ketoglutarate to 2-hydroxyglutarate and the oxidation of NADPH to NADP+ [15]. Nonetheless, the ramifications of the presence of the mutant enzyme or the new metabolite 2-hydroxyglutarate on the cellular phenotype are still being unraveled.
The Warburg effect, in which there is an increased rate of aerobic glycolysis, is recognized as a common feature of cancer cells [16],[17], and is particularly prominent in GBM [18]. Increased glycolysis may provide an advantage to a dividing cancer cell by providing not only energy but also basic metabolites required by a rapidly dividing cell [17]. It is possible that mutations in IDH1/2 may provide a mechanism for initiating this shift in metabolism, by altering the levels of key metabolites needed in a rapidly growing cell [19]. The changes in metabolites and related pathways are still being elucidated.
Studies looking at the effect of mutant IDH1/2 in leukemias, the only other tumor type in which they are present to any large degree, have suggested that the presence of the mutant protein may inhibit hematopoietic differentiation via changes in DNA methylation status [20, 21]. Nonetheless, it is unclear if the same phenotype is observed in gliomas.
One of the paradoxes in the study of glioblastomas is that patients with mutant IDH1/2 tend to have improved survivals compared to patients with wild-type IDH1/2 (citation needed). It has been proposed that this may be due to an increased sensitivity to oxidative stress. Mutant IDH1/2 can potentially lead to decreases in NADPH through its active depletion. NADPH is necessary for the reduction of glutathione, which serves to protect the cell against oxidative stress (reviewed in Hammond et al. [22]).Studies have shown that, in fibroblast, leukemia, and other cell lines, increased IDH1 levels lead to an increase in the ratio of reduced glutathione to total glutathione [23, 24], and appear to provide protection against oxidative damage [25-27]. Interestingly, studies using the addition of 2-hydroxyglutarate in vitro to cerebral cortex lysates from young rats show an increase in oxidative stress and lipid peroxidation [28]. These studies provide two different mechanisms through which mutant IDH1/2 may lead to increased oxidative stress- decrease in native IDH1/2 activity with a concomitant increase in 2-hydroxyglutarate.
Currently available glioma cells lines do not have endogenous IDH1/2 mutations, and generating lines from tumors which have mutant IDH/2 has proven challenging. Studies involving artificial introduction of mutant IDH1/2 to currently available glioma cell lines may not yield useful data, as these tumors did not require IDH1/2 for tumor formation. Even if gliomas lines with mutant IDH1/2 were available, there would be some shortcomings. As IDH1/2 mutation is an early event, its largest effect may be on tumor initiation, something which would be challenging to study in established glioma lines. Thus, it is necessary to develop a good system in which to study the effects of mutant IDH1/2. Because these are among the earliest changes in progressive gliomas, it is possible that these changes occur in a precursor cell that is common to both oligodendrogliomas and progressive astrocytomas. A cell that is in a more undifferentiated state may have a unique response to the changes induced by the presence of mutant IDH1/2 compared to a more differentiated cell.
To study the effect of mutations in IDH1/2 in undifferentiated cells, we have generated mouse ES cell lines that stably express IDH1R132H or IDH2R172K. These lines provide a unique opportunity to assess the effects of mutant IDH1/2 on cells in a less differentiated state with a comparable wild-type line as a control. We will use these lines to address the following two hypotheses: 1) Mutations in IDH1/2 lead to changes in metabolic pathways, leading to increased oxidative stress and a dependence on glutamine for survival. 2) Mutations in IDH1/2 alter the normal DNA methylation patterns, leading to inhibition of normal differentiation. Additionally, we hypothesize that mutation of either IDH1 or IDH2 is sufficient to promoter tumor formation. Using these mouse ES cell lines, we will generate mice to assess for the ability of mutant IDH1/2 to form tumors. These mice will also give us the opportunity to observe early tumor development, and will be the first genetically faithful model system in which to study progressive gliomas.
These studies will provide key insights into the role of mutant IDH1/2 in the development or progression of tumors by identifying key alterations in metabolic pathways and the ramifications on cellular phenotype and response to stressors. This understanding will allow us to begin to target treatments towards those changes which are specific to cells with mutant IDH1/2. Additionally, mice with mutant IDH1/2 provide a unique in vivo system in which to study the formation of progression of these tumors. Progressive gliomas are highly invasive tumors, which likely interact a great deal with the surrounding parenchyma. This in vivo system will allow us to study progressive gliomas in a system which most closely mimics tumors as seen in human patients, and provides a potential system in which to begin to address ways to treat this devastating disease.
REFERENCES