Introduction
Mutations of the enzyme cytosolic isocitrate dehydrogenase (IDH) 1 in gliomas, initially identified in 2008 (1), have drawn particular attention of neuro-oncologists in recent years. It has been demonstrated that mutations of IDH1 occur in ~80% of grade II and III gliomas and secondary glioblastomas in humans (2–4). IDH1 mutations show high specificity in gliomas, but are rarely found in other types of human cancers (5). Thus, mutations of IDH1 have been proposed to have important roles in the tumorigenesis, progression and prognosis of gliomas.
The family of IDHs consists of three catalytic isozymes: IDH1, IDH2 and IDH3 (6). IDH is a critical enzyme in the tricarboxylic acid cycle (TCA) which catalyzes isocitrate (ICT) into α-ketoglutarate (α-KG) in a nicotinamide adenine dinucleotide phosphate (NADP+)-dependent manner (IDH1 and IDH2) or nicotinamide adenine dinucleotide (NAD+)-dependent manner (IDH3). A mutation has been found in the gene, which leads to substitution of the 132nd amino acid residue (an arginine) of IDH1 by a histidine (R132H); this has been found to account for ~90% of all IDH1 mutations, with other mutations including substitution at this codon for cysteine (R132C), glycine (R132G), leucine (R132L) or serine (R132S) (4,7,8). The substituted R132 is essential for enzymatic activity, and is contained within an evolutionary highly conserved region located at the binding site for ICT (1). More recently, mutations of IDH2 were also identified in a group of gliomas at a low frequency (4). It is intriguing that R172 in IDH2, which is the analogous site to R132 in IDH1, is also the major mutation site of IDH2 in gliomas (4). Notably, mutations of IDH1 and IDH2 are mutually exclusive in gliomas, implying that the outcome of the IDH1 or IDH2 mutations may be equivalent (5).
Functional studies of IDH1 mutations have demonstrated that mutant IDH1 blocks the enzymatic activity and results in decreased α-KG levels (9); however, a gain of neomorphic activity at high levels of 2-hydroxyglutarate (2-HG) is observed (10). The high levels of 2-HG have been indicated to have potential value for the clinical diagnosis and treatment of gliomas (11). However, the manner by which IDH1 mutants are involved in tumorigenesis remains to be elucidated. One of the explored mechanisms of mutant IDH1 in the regulation of tumorigenesis is the upregulation of hypoxia-inducible factor 1-α (HIF1-α) (9). HIF1-α has a critical role in cells under hypoxic conditions, which activates numerous genes that are involved in multiple cellular processes, including cell proliferation and survival, metastasis and angiogenesis, predominantly via the vascular endothelial growth factor (VEGF) (9,12,13). Under normal circumstances, HIF1-α was continuously degraded by prolyl hydroxylase (PHD) which is dependent on α-KG. Thus, decreased α-KG levels downregulate the activity of PHD, resulting in the stabilization and accumulation of HIF1-α and the subsequent activation of the HIF signaling pathway. It is noteworthy that gene expression that was expected to be induced by HIF1-α was not observed in patients with acute myeloid leukemia harboring IDH1 mutations (14). It is possible that multiple cellular signaling pathways are involved in cells with IDH1 mutations.
Various studies have demonstrated that cellular hypoxia and HIF signal activation are associated with the nuclear factor-κB (NF-κB) transcription factor activation (15–17). In 1994, hypoxia was found to induce the activation of NF-κB (18); however, the underlying mechanism remains to be identified (19,20). NF-κB, which is tightly controlled, regulates a wide range of cellular processes and is extensively involved in various cancer types, including gliomas (21). Normally, NF-κB binding to inhibitory IκB proteins is inactivated in the cytoplasm (22). Thus, translocation of NF-κB to the nucleus under dysregulation is involved in numerous diseases (23,24). NF-κB has also been demonstrated to be a critical regulator in tumorigenesis and to be involved in cell survival, metastasis and angiogenesis (22). Considering the parallel roles of the two transcription factors in tumorigenesis, it remains to be elucidated whether there is a correlation in tumorigenesis.
The present study investigated the role of IDH1R132H expression on cell proliferation in glioma cells and the potential underlying mechanism.
Materials and methods
Cell lines and cell culture
Human SVGp12 normal astrocyte and U251 glioma cell lines were obtained from American Type Culture Collection (ATCC, Manassas, VA, USA). All the cells were maintained as per standard protocols. Briefly, the cells were cultured in Dulbecco’s modified Eagle medium (DMEM; Invitrogen Life Technologies, Carlsbad, CA, USA) with 10% of fetal bovine serum. Penicillin (100 U/ml), streptomycin (100 μg/ml) and L-glutamate (2 mM) (Sangon, Shanghai, China) were added. All the cells were cultured at 37°C in an incubator (Life Technologies, Baltimore, MD, USA) containing 5% CO2.
Cloning, site-directed mutagenesis and expression vector construction of IDH1
The total RNA was extracted from cultured U251 glioma cells using TRIzol® reagent (Invitrogen Life Technologies) according to the manufacturer’s instructions. Up to 5 μg of total RNA was reverse-transcribed into cDNA using Moloney murine leukemia virus (M-MLV) reverse transcriptase (Clontech, Palo Alto, CA, USA). The cDNAs were used as templates for the amplification of wild-type IDH1 using two primers (forward, 5′-atgtccagaaaaatccatggcggttctg-3′ and reverse, 5′-ttaaagtttggcctgagctagtttggccttc-3′), according to the open reading frame of the IDH1 cDNA sequence (GenBank accession no. NM_031510.1). The specific site-directed mutagenesis primers (forward, 5′-ctatcatcataggtcatcatgcttatggggatc-3′ and reverse, 5′-gatccccataagcatgatgacctatgatgatag-3′) were used to amplify the cDNA sequence of mutated IDH1R132H with the wild-type IDH1 cDNA sequence as templates according to the protocols outlined in the Site-Directed Gene Mutagenesis kit (Beyotime Institute of Biotechnology, Shanghai, China). The obtained sequences were confirmed by DNA sequencing (Sangon, Shanghai, China) and the resultant fragments of wild-type and mutated IDH1 were sub-cloned into pCMV-Sport6 plasmid (Invitrogen, Carlsbad, CA, USA) with restriction sites NotI and XhoI.
Recombinant plasmids and small interfering RNA (siRNA) transfection
The cells were seeded in a six-well culture plate (2×105 cells/well) and incubated at 37°C with 5% CO2 until the cells reached 80% confluence. Cell transfection was performed according to the manufacturer’s instructions. Briefly, plasmid DNA (1 μg) or siRNA (HIF-1α siRNA sc-35561 and VEGF siRNA sc-29520, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) was diluted in 500 μl of DMEM with 5 μl Lipofectamine (Invitrogen), prior to being mixed and incubated at room temperature (RT) for 15 min. The mixtures were then added to the cells resulting in a final volume of 3 ml medium and incubated for the indicated times.
Nuclear protein extraction
The nuclear proteins were extracted using an extraction kit (Sangon, Shanghai, China) according to the manufacturer’s instructions. Briefly, the cells were lysed in cytoplasmic buffer containing protease inhibitors, mixed and incubated for 15 min at 4°C followed by centrifugation at 13,400 × g for 20 min at 4°C. The cell pellets were collected and resuspended in nucleus buffer (Ficoll 400 and protease inhibitor; Sangon, Shanghai, China) for 10 min at 4°C. Next, the sample was centrifuged at 13,400 × g for 10 min at 4°C. The supernatant containing nuclear proteins was collected for analysis.
Western blot analysis
Proteins from cultured cells were collected and a total of 20–30 μg protein was fractionated by 12% SDS-PAGE and transferred to a nitrocellulose membrane (Amersham, Little Chalfont, UK). The membrane was treated using the following procedure by agitating and blocking at RT with 2% skimmed milk in Tris-buffered saline (TBS) for 1 h followed by incubation in primary antibodies [IDH1R132H specific mouse monoclonal antibody IDH1-R132H mutant monoclonal antibody (IMab-1) (Dianova, Hamburg, Germany); mouse monoclonal HIF1-α antibody, mouse monoclonal VEGF antibody and rabbit polyclonal NF-κB p65 antibody (Santa Cruz Biotechnology, Inc.); mouse monoclonal IκBα and pIκBα antibody (Cell Signaling Technology, Boston, MA, USA)]. The solution was diluted in blocking buffer (2% skimmed milk powder dissolved in TBS) at 4°C overnight and washed three times with TBS and Tween-20 (TBST; 10 mM Tris-HCl, pH 7.5, 150 mM NaCl, and 0.05 % Tween-20) for 10 min each at RT. Subsequently, the membrane was incubated in peroxidase-conjugated secondary antibody (Boster Corporation, Wuhan, Hubei, China; diluted at 1:5,000 in blocking buffer) for 1 h. Following washing three times with TBST and once with TBS each for 10 min, 1 ml of 4-chloro-1-naphthol as a horse radish peroxidase substrate with 9 ml of TBS and 6 μl of H2O2 was used for visualizing the target protein in the dark for 5–30 min.
MTT assay
For the MTT assay, cells were planted in 96-well plates and cultured under regular conditions until they reached 80% confluency. The plasmid or siRNA was transfected according to the standard protocols, and were continually incubated with cells at 37°C with 5% CO2 for 48 h. Next, the culture medium was discarded and fresh medium containing MTT (5 mg/ml in PBS, 150 μl/well, Sangon, Shanghai, China) and incubated with cells for an additional 4 h. Next, 150 μl of dimethyl sulfoxide (Sigma-Aldrich, St. Louis, MO, USA) was added per well, agitated gently for 10 min to dissolve the formazan, and the absorbance at 490 nm was determined by an ELISA reader (Bio-tek Instruments, Inc., Winooski, VT, USA). Each cell viability assay was performed in quadruplicate and repeated three times. The data are expressed as the mean ± standard error of the mean and differences were analyzed by Student’s t-test.
Quantitative polymerase chain reaction (qPCR) analysis
The total RNA was extracted from cultured cells using TRIzol reagent (Invitrogen Life Technologies) according to the manufacturer’s instructions. Up to 5 μg of total RNA was reverse-transcribed into cDNA using M-MLV reverse transcriptase (Clontech). The cDNAs were used as templates for qPCR. The primers were as follows: c-myc, forward, 5′-acacatcagcacaactacgc-3′ and reverse, 5′-cctcttgacattctcctcggt-3′; cyclin D1, forward, 5′-gccaacctcctcaacgaccgg-3′ and reverse, 5′-gtccatgttctgctgggcctg-3′; cyclin E, forward, 5′-gtcctggctgaatgatacatgc-3′ and reverse, 5′-ccctattttgttcagacaacatggc-3′; β-actin (forward, 5′-ctccatcctggcctcgctgt-3′ and reverse, 5′-gctgtcaccttcaccgttcc-3′) were used as the control. The qPCR mixture contained 5 μl SsoFast™ EvaGreen Supermix (Bio-Rad, Hercules, CA, USA), 1 μl of cDNA (diluted at 1:50) and 2 μl of each of the forward and reverse primers (1 μM) to a final volume of 10 μl. The PCR procedure was as follows: 94°C for 4 min, 94°C for 20 sec, 55°C for 30 sec and 72°C for 20 sec; 2 sec for plate reading for 35 cycles; and a melting curve from 65 to 95°C. β-actin was used as the control for normalizing gene expression. In total, three independent experiments were performed. The data obtained were calculated by the 2−ΔΔCt method and subjected to statistical analysis as previously described (25), followed by an unpaired sample t-test.
Results
Expression of IDH1<sup>R132H</sup> in glioma cells
To analyze the role of IDH1 mutations in glioma cells, human U251 glioma and SVGp12 normal astrocyte cell lines transfected with vectors expressing the myc-tagged wild-type or R132H mutant of IDH1 were established. The cell lysates were assessed by for western blot analysis of IDH1 protein expression. In order to determine IDH1R132H protein levels specifically, the IDH1R132H-specific monoclonal antibody IMab-1 was used. The results showed that IDH1R132H was only detected in cells transfected with IDH1R132H expression vectors (Fig. 1, upper panel). Furthermore, following incubation with IDH1 antibodies, two bands were detected in overexpressing, vector-transfected cells: The upper band denoted the myc-tagged IDH1 and the lower band denoted endogenous proteins of IDH1 (Fig. 1, middle panel). These results implied that the IDH1R132H mutant was successfully expressed in U251 and SVGp12 cells.
Expression of IDH1<sup>R132H</sup> promotes cell proliferation
To investigate the role of forced expression of IDH1R132H in cell growth, the MTT assay was performed. The results showed that the expression of IDH1R132H significantly increased cell proliferation in U251 cells, while overexpression of IDH1 wild-type had no marked influence on cell proliferation (Fig. 2A). Similar results were also observed in SVGp12 cells (Fig. 2B). The data indicated that IDH1 mutants increased cell proliferation.
Knockdown of VEGF has no significant impact on cell proliferation induced by IDH1<sup>R132H</sup> expression
HIF1-α augmentation in IDHR132H-expressing cells has been shown to be involved in tumor growth, which was associated with VEGF (9). To clarify whether IDH1R132H promotes cell proliferation via HIF1-α-induced VEGF expression, VEGF RNA interference or transfection with IDH1 expressing vector was performed in U251 glioma cells (Fig. 3). Western blot analysis showed that the expression of IDH1R132H significantly increased HIF1-α protein levels in U251 glioma cells. The results also demonstrated that expression of IDH1R132H upregulated VEGF expression, causing a high cell proliferation rate. Notably, cell proliferation was not markedly affected when VEGF was knocked down, which implies that IDH1R132H-induced cell proliferation may be involved in multiple signaling pathways (Fig. 3). Similar results were obtained using SVGp12 cells (data not shown).
Expression of IDH1<sup>R132H</sup> promotes NF-κB nuclear translocation and IκBα phosphorylation
In order to assess whether NF-κB was activated in IDH1R132H expressing cells, the NF-κB signaling pathway was investigated in transfected cells. It is known that NF-κB dimers are inhibited by IκB (α, β or ɛ) in the cytoplasm, while the phosphorylation of IκB results in NF-κB nucleus translocation and activation (22). Thus, the phosphorylation state of IκB in IDH1R132H expressing U251 cells was assessed. The results revealed that the expression of IDH1R132H promoted the phosphorylation of IκBα, implying that IκBα was released from NF-κB (Fig. 4, upper left panels). In addition, in order to define whether NF-κB nuclear translocation is increased upon IDH1R132H expression, the levels of p65 protein, which is one subunit of NF-κB, were detected in the nucleus. As expected, the NF-κB p65 protein levels were increased in nuclear extracts, as was HIF1-α (Fig. 4, lower left panel). The same effects of overexpression of IDH1R132H on NF-κB were observed in SVGp12 cells (Fig. 4, right panels). These results indicated that expression of IDH1R132H promoted NF-κB activation.
Knockdown of HIF1-α blocks NF-κB activation induced by expression of IDH1<sup>R132H</sup>
In order to confirm whether NF-κB activation was dependent on HIF1-α, HIF-1α RNA silencing in combination with transfection with a vector expressing IDH1R132H was performed in cells. At ~72 h post transfection, the cells were harvested and nuclear proteins were extracted and subjected to western blot analysis. In the U251 and SVGp12 cell lines, knockdown of HIF-1α significantly blocked NF-κB p65 nuclear translocation induced by expression of IDH1R132H (Fig. 5), implying that NF-κB activation occurred in a HIF1-α-dependent manner.
IDH1<sup>R132H</sup> promotes cell proliferation through cyclin D1 and E and c-myc activated by NF-κB
Various studies have demonstrated that NF-κB is involved in the regulation of cell proliferation through cyclin D1 and E and c-myc (26–28). In order to further define whether expression of IDH1R132H was involved in NF-κB-mediated cell proliferation, the transcription levels of cyclin D1 and E and c-myc were determined by qPCR. IDH1R132H, as opposed to wild-type IDH1, significantly induced the expression of cyclin D1 and E and c-myc in U251 cells (Fig. 6). Furthermore, increased expression levels of cyclin D1, cyclin E and c-myc were also observed in IDH1R132H expressing SVGp12 cells (data not shown).
Discussion
The IDH1 gene is specifically mutated in gliomas, including astrocytomas, oligodendrogliomas or oligoastrocytomas of grades II and III and secondary glioblastomas, with an occurrence of ~80%; mutations in this gene seldomly occur in primary glioblastomas or other types of human gliomas (29). Mutation of IDH1 has been identified as an early event in tumorigenesis, indicating that it has critical roles in the evolutionary development of gliomas (3,4). However, the molecular mechanism of IDH1 mutants underlying the development and progression of tumors remains to be elucidated.
In 2009, Zhao et al (9) first reported that the IDH1 mutation contributed to tumorigenesis through the induction of HIF1-α. In this study, overexpression of IDH1 mutants in cultured cells suppressed the activity of wild-type IDH1 via the formation of heterodimers, resulting in a decrease of the enzyme product, α-KG. α-KG is essential for PHD, which promotes HIF1-α degradation. Thus, forced expression of the IDH1 mutant activated the HIF1-α signal and subsequently targeted gene expression, including that of glucose transporter 1, VEGF and phosphoglycerate kinase 1. Another study demonstrated that IDH1 mutants acquired neomorphic activity that catalyzed α-KG to 2-HG and resulted in an accumulation of 2-HG that led to tumor formation and malignancy (10). In accordance with a previous study, the present study identified that forced expression of IDH1R132H mutants in cells increased HIF1-α and cell proliferation. Furthermore, it was also shown that IDH1R132H activated the NF-κB signaling pathway, which is frequently dysregulated in cancers (30).
In the present study, it was demonstrated that the knockdown of VEGF, the HIF1-α target gene, notably contributed to angiogenesis in tumorigenesis, and did not affect the cell proliferation evoked by expression of the IDH1 mutant. Thus, it was hypothesized that the IDH1 mutant may activate other signaling pathways dependent on HIF1-α. As expected, it was shown that forced expression of IDH1 mutants activated NF-κB in a HIF1-α dependent manner. The crosstalk between HIF1-α and NF-κB has been previously reported (15). HIF1-α was indicated to promote inflammation of the local stromal microenvironment via NF-κB, which was in favor of tumor cell growth and metastasis (31). NF-κB was demonstrated to be a downstream effector of the HIF1-α-dependent response, which was involved in the regulation of neutrophil survival (16). Expression of IDH1R132H promoted the phosphorylation of IκBα and resulted in the release of NF-κB, which subsequently accumulated and translocated into the nucleus. However, there were limitations to the present study: The manner by which HIF1-α regulates NF-κB activation was not defined. Whether molecular events between HIF1-α and NF-κB activation exist requires further investigation. Previously, it has been proposed that the oxygen-sensing hydroxylases which regulate the HIF pathway may also be involved in the regulation of significant components of the NF-κB pathway (32,33).
It has been reported that the activation of NF-κB is tightly controlled. Normally, NF-κB is inhibited by IκB, which is able to be phosphorylated by the IκB kinase complex, resulting in NF-κB liberation and activation (34,35). NF-κB is activated in a wide range of cancers and activates various genes involved in tumorigenesis processes, including cell survival, apoptosis, angiogenesis and metastasis (21). One of the significant features of NF-κB with regard to gliomas is the regulation of cellular proliferation. Oncogenes including cyclin D1 and E, and c-myc have been reported to be activated by NF-κB. High levels of cyclin D1 and E were found in the majority of glioblastomas, and cyclin D1 was closely correlated to cell proliferation and to the progression of tumors (36–38). c-myc is a critical transcription factor, which induces numerous cyclins and CDK proteins and is highly elevated in tumors (39,40). In the present study, it was demonstrated that expression of IDH1 mutants activated NF-κB and induced high expression levels of cyclin D1 and E, and c-myc in cultured cells. Thus, it was deduced that IDH1R132H promoted cell proliferation not only via HIF1-α-induced VEGF activation, but also by HIF1-α-induced NF-κB activation.
It is of note that patients with IDH1 mutations have an improved prognosis compared to those harboring wild-type IDH1 (1,41). In addition, the IDH1 mutants showed high sensitivity to chemotherapeutics (42,43). It is indicated that the IDH1 mutations not only reduce the activity of NADPH generation, but also consume NADPH, thus leading to an increase in oxidative stress and oxidative DNA damage (10,44). This may explain why glioma cells harboring IDH1 mutations have a high sensitivity to chemotherapy and patients with IDH1 mutations have longer survival times compared with patients harboring wild-type IDH1.
Collectively, the mutant IDH1 is involved in tumorigenesis through the induction of HIF1-α. On the other hand, IDH1 sensitizes tumor cells to chemotherapy via an increase in cellular oxidative stress. In the present study, it was demonstrated that the forced expression of IDH1 mutants promoted the proliferation of glioma cells via the activation of NF-κB in a HIF1-α dependent manner. The present study provided new insights into the IDH1 mutation in the regulation of tumorigenesis. Since the results of the present study showed that IDH1 mutants are oncogenic, it requires to be further determined whether targeting IDH1 mutations is helpful for glioma therapy. More recently, a selective inhibitor (AGI-5198) of DHI1R132H has been developed by Rohle et al (45), which suppresses the production of 2-HG as well as colony formation. In particular, this inhibitor had no effect on wild-type IDH1. This inhibitor also decreased the proliferation of IDH1 glioma xenografts, while not affecting apoptosis, implying that targeting IDH1 mutations may reverse the dysregulation of cell growth and may be beneficial for gliomas. However, the precise molecular pathogenesis of IDH1 mutants in gliomas requires further investigation.