IDH1R132H Mutation Inhibits the Proliferation and Glycolysis of Glioma Cells by Regulating the HIF-1α/LDHA Pathway

Background: This study aims to explore the role and underlying mechanism of the IDH1 R132H in the growth, migration, and glycolysis of glioma cells. Methods: The alternation of IDH1, HIF-1α, and LDHA genes in 283 LGG sample (TCGA LGG database) was analyzed on cBioportal. The expression of these three genes in glioma tissues with IDH1 R132H mutation or IDH1 wild type (IDH1-WT) and normal brain tissues was also assessed using immunohistochemistry assay. In addition, U521 glioma cells were transfected with IDH1-WT or IDH1 R132H to explore the role of IDH1 in the proliferation and migration of glioma cells in vitro. Cell growth curve, Transwell mitigation assay, and assessment of glucose consumption and lactate production were conducted to evaluate the proliferation, migration, and glycolysis of glioma cells. Results: The expression of HIF-1α and LDHA in IDH1 R132H mutant was signicantly lower than that in glioma cells with wild type IDH1 (P<0.05). IDH1 R132H inhibited the proliferation and glycolysis of U521 glioma cells. Conclusion: The IDH1 mutation IDH1 R132H plays an important role in the occurrence and development of glioma through inhibiting the expression of HIF-1α and glycolysis.


Introduction
Due to aggressive in ltration and metastasis, glioma is a major cause of cancer deaths with a high rate of recurrence, and poor prognosis and clinical outcomes (1). There are two major types of glioma, astrocytoma and oligodendroglioma, which originates from astrocytes and oligodendrocytes, respectively (2). With intensive healthcare efforts, the ve-year survival rate of glioma in the United States was only 5% (3). Therefore, understanding the molecular mechanisms underlying the development of glioma becomes a critical challenge to effective treatment of glioma worldwide (4).
A number of molecular markers, which are associated with the diagnosis, prognosis, and therapy of glioma have been identi ed and characterized. For example, mutations of the gene encoding isocitrate dehydrogenase 1 (IDH1) mutation have been detected in 70% of low-grade gliomas (WHO grading II-III) (4). In addition, approximately 90% of IDH1 mutations in glioma patients leads to the replacement of arginine by histidine at the 132nd amino acid (IDH1 R132H mutation) (5). It has been reported that most gliomas patients with the IDH1 R132H mutation exhibited signi cantly better response to treatment and longer survival compared with glioma patients with wild-type IDH1 (6). However, the mechanism underlying the association between IDH1 R132H mutation and improved survival rate has not been fully understood.
It is widely recognized that energy and nutrients are necessary for the proliferation of cells. Uncontrolled proliferation of solid tumor cells may cause severe hypoxia, especially in the center of tumors. In order to survive under low oxygen conditions, tumor cells adapt metabolic patterns to the harsh microenvironment, which is known as metabolic reprogramming of tumor cells (7). Tumor metabolism has become a vital issue in tumor studies (8). Initially, Warburg et al. observed that rapidly proliferating cancer cells utilize glucose to produce lactic acid under aerobic conditions (9). While this metabolic pathway exhibits low e ciency in the production of adenosine triphosphate, it's important for cancer cells to synthesize macromolecules and generate su cient energy for growth. This special metabolism of tumor cells is known as the Warburg effect (10,11). Hypoxia-inducible factor 1α (HIF-1α) is an important regulator involved in glucose metabolism (12). In order to stimulate uncontrolled proliferation of cancer cells, HIF-1α reprograms the metabolic pathways of cancer cells in a different way compared with normal cells. HIF-1α altered the metabolism of amino acids and lipids through increasing the uptake of glucose and glutamine and the production of lactic acid (13). Therefore, HIF-1α is a potential target for cancer treatment, which may regulate cell proliferation, metabolism and carcinogenic stress. Biologically, HIF-1α is a transcription factor, which regulates chromatin structure to control several transcriptional processes.
The association between IDH1 R132H mutation and HIF-1α in glioma has not been well understood (14)(15)(16). While HIF-1α-mediated anaerobic glycolysis plays an important role in the occurrence and development of a variety of tumors, only a few studies have shown that IDH1 mutations inhibit glycolysis. In order to evaluate the effect of IDH1 R132H mutation on glioma growth and its association with glycolysis, we generated glioma U251 cell line with IDH1 R132H overexpression. We also investigated the downstream molecules of the IDH1 pathway in U251 cell line overexpressing IDH1 R132H . Our study will provide new clues for the development of targeted therapy for glioma.

Materials And Methods
2.1 Alterations of IDH1, HIF-1α, and LDHA genes in glioma tissues from cBioPortal The alterations including ampli cation, deep deletion, and mutations of IDH1, HIF-1α, and Lactate dehydrogenase A (LDHA) genes in glioma were analyzed on cBioPortal (http://www.cbioportal.org). OncoPrint was constructed in cBioPortal (TCGA provisional) to directly characterize the gene ampli cation, deep deletion, and mutation of IDH1, HIF-1α, and LDHA genes in glioma tissues. To comprehensively identify the alterations of IDH1, HIF-1α, and LDHA genes, we checked two data sets, the Brain Lower Grade Glioma (TCGA, Provisional) of 530 samples and the Merged Cohort of low-grade glioma(LGG) and glioblastoma(GBM) (TCGA, Cell 2016) with a total of 1102 samples. In addition, the Kaplan Meier survival curve of cBioportal was used to evaluate the association between IDH1, HIF-1α and LDHA gene alterations and overall disease-free survival of glioma patients.

Glioma specimens
Glioma tissues were obtained from patients undergone surgical tumor removal in the Department of Neurosurgery of Sixth Medical Center, Chinese PLA General Hospital. All the patients who participated in this study have signed informed consent, and the protocol involving human specimens has been approved by the hospital Ethics Committee. The glioma specimens were rapidly frozen in liquid nitrogen after removed from patients and stored at − 80 ℃ for subsequent analyses.

Cell lines and cell culture
Human glioma cell lines U251 was purchased from the American type culture collection (ATCC, USA) and have been checked to ensure they are free of contamination. U251 cells were cultured in Dulbecco's modi ed eagle medium/F12 mixed medium supplemented with 10% FBS at 37°C in a humidi ed incubator containing 5% of CO 2 .

RNA isolation and real-time quantitative polymerase chain reaction (RT-qPCR)
Total RNA was isolated from U253 cells using the TRIzol reagent (Invitrogen, USA), and reversely transcribed into cDNA using the Reverse Transcription cDNA Kit (Osaka, Japan) according to the manufacture's protocols. RT-qPCR was conducted on the ABI Prism 7500 real time PCR platform (Applied Biosystems, Life Technologies, Carlsbad, CA) using the SYBR Premix EX Taq (Takara Dalian, Dalian, China). Human GAPDH (glyceraldehyde-3-phosphate dehydrogenase) was used as an internal control in RT-PCR assay. Primers used for stem-loop reverse-transcription PCR of IDH1: IDH1-human-F: GGTGACATACCTGGTACATAACTTTG, IDH1-human-R: GTGTGCAAAATCTTCAATTGACT, GAPDH-F: TCATTGACCTCAACTACATGG, GAPDH-R: TCGCTCCTGGAAGATGGTG. After RT-PCR reactions, the PCR products were collected and analyzed by 1% agarose gel electrophoresis.
2.6 Cell transfection with the IDH1 mutant (IDH1 R132H -mut) and wild-type IDH1 (IDH1-wt) IDH1 R132H mutant and IDH1-wt lentivirus were purchased from the Gene Pharma company. A total of 10,000 cells were seeded into 12-well plates and incubated at 37°C for 2 h prior to transfection with lentivirus in the medium containing 5 µg/mL polybrene. Lentivirus vector-treated or not treated cells were used as controls.

Cell proliferation assay
Primary or treated cells were harvested and mechanically separated by pipetting. Then the cells were seeded into 24-well plates (5 x 10 4 cells/well), and incubated for 7 days. The cells were counted every day using a hemocytometer. The experiment was performed in triplicate.

Transwell invasion assay
Transwell invasion assay was performed using 24-well Transwell inserts (diameter 8 mm, Corning, NY, USA) precoated with Matrigel (60-80 µL, Corning, NY, USA) on the top surface of the polycarbonic membrane (pore size 8 µm). Brie y, the dissociated cells were seeded into the upper chambers of Transwell inserts (5 x 10 4 cells / insert) in 200 µL medium. The lower compartments were lled with 500 µL 10% of FBSD/F-12 medium according to the protocol. After incubation at 37°Cfor 48 h, the cells, which migrated into lower chambers, were harvested and counted under a microscope (Nikon, Tokyo, Japan). 2.10 Glucose uptake and lactate production assay U251cells were cultured in sugar-free DMEM (dulbecco's modi ed eagle medium) for 16 h, and then cultured in high-sugar DMEM (4.5g/L glucose) for 24 h under non-ionic conditions. The medium was then removed and the glucose level in U251 cells was determined using the uorescence-based Glucose Assay kit (BioVision, Milpitas, California, USA) according to the manufacturer's instructions. Lactic acid levels were measured using the colorimetric method (Beotim, Wuxi, China) based on lactate oxidase reading at 540nm according to the manufacturer's instructions, and standardized to cell numbers.

Statistical analysis
All experiments were conducted in triplicate and experimental data were presented as mean ± SD. Independent 2-tailed student's t-test, or two-way analysis of variance (ANOVA) was used to identify statistical signi cance (*: P < 0.05 and **: P < 0.01).

Results
3.1. The alternations of IDH1, HIF-1α, and LDHA genes in glioma tissues Alterations of IDH1, HIF-1α, and LDHA genes were identi ed in 77%, 0.4%, and 1.4% of the sequenced cases, respectively, in the LGG data obtained from the OncoPrint schematic of cBioPortal (Fig. 1A), and 51%, 0.1%, and 0.6% of the sequenced cases, respectively, in the LGG and GBM data. In addition, 90% of IDH1 mutation led to the replacement of arginine at position 132 by histidine (IDH1 R132H mutation). The cancer types were listed in the chart (Fig. 1B, 1C, 1D). Regarding the LGG dataset, only one patient harbored deep deletion of the HIF-1α gene, and 4 patients had alterations of the LDHA gene, including two cases of gene ampli cation and two cases of deep deletion. Correlation analyses between gene alterations and patient survival showed that the overall survival rate (OS) and progression-free survival rate (PFS) of LGG patients were associated with IDH1 mutation (Fig. 2A, 2B) (P < 0.05). The correlation of the gene alternations and the clinical features of glioma were shown in Table 1 and Table 2.

Low-expression of HIF-1a and LDHA in glioma tissues with IDH1 R132H -mut
We rst identi ed the expression of HIF-1α and LDHA in 10 IDH1 R132H and 10 IDH1-wt glioma specimens using immunohistochemical staining. The results showed signi cantly reduced expression of HIF-1α and LDHA in glioma tissues with IDH1 R132H mutation compared with that in glioma samples with IDH1-wt ( Fig. 3, P < 0.001). The clinical characteristics of the patients who donated the tumor specimens were list in Table 3.

Discussion
Glioma is the most common malignant tumors in central nervous system (CNS), however, the pathogenesis of glioma remains unclear. Currently, most clinical data demonstrated that glioma patients with the IDH1 R132H mutation had a preferable outcome compared with glioma patients with IDH1 wild-type. Paradoxically, IDH mutation was proven to be a triggering event in gliomagenesis. For example, the 2-hydroxybutyrate (2-HG), which is the vice-product of IDH1 mutant, can promote glioma growth through its capacity to competitively inhibit a-KG-dependen enzymes. Interestingly, we identi ed that there is an alterations of IDH1, HIF-1α, and LDHA genes in the glioma data obtained from the OncoPrint schematic of cBioPortal as showing in Fig. 1A. Therefore, we investigated the role of IDH1 R132H mutation in the development of glioma in this study.
Notably, the expression of HIF-1α was signi cantly decreased in U521 glioma cells tranfected with IDH1 R132H mutation in our study. As we know, HIF-1α was an important transcription factor, and could promote tumor growth under hypoxia condition (17). The stability of HIF-1α is regulated by alpha ketoglutaric acid (α-KG), an important enzyme product of IDH1 (18). This decreased trendency of HIF-1α was consistent with better clinical prognosis in gliomas, and alterations of HIF-1α in the glioma database obtained from the OncoPrint schematic of cBioPorta. However, Zhao et al. (19) demonstrated that IDH1 R132H over-expression increased the protein level of HIF 1α in U87 glioma cells in the previous study. This difference may attribute to application of two different cell lines in these two experiments.
Our experiment also showed that the IDH1 R132H mutation inhibited the proliferation and migration of glioma U251 cells in vitro. However, whether this result was correlted with IDH1 R132H mutation, and the underlying mechanism was the next step that we wanted to explore. Generally, HIF-1α played important roles in glycolysis. In order to reach proliferate rapidly, cancer cells need more energy through glycolysis than normal cells. Meanwhile, Lactate dehydrogenase A (LDHA) plays a principal role in cancer metabolism (7,20,21). It has been shown that LDHA correlated with a number of clinicopathological features and the survival outcomes of a variety of tumors (22,23). And many previous studies demonstrated that inactivation of HIF-1α/LDHA axis in cancer could inhibit the Warburg effect and tumor progression (23)(24)(25).
Therefore, to explore the role of IDH1 R132H mutation in HIF-1α-mediated glycolysis in glioma, we examined the expression of LDHA levels in IDH1 R132H mutated glioma U251 cells. As seen in Fig. 4C and Table 4, IDH1 R132H -mut inhibited glucose consumption and lactate production in U251 glioma cells (P < 0.001). However, glucose consumption and lactate production were increased in the U251-vector and U251-IDH1-wt groups (P < 0.001). Meanwhile, IDH1 R132H mutation decreased the expression of HIF-1α and LDHA at both mRNA and protein levels (Fig. 4A, B). These down-regulation of LDHA in transfected U251-IDH1 R132H -mut cell lines indicated that IDH1 R132H mutation inhibited the glycolysis in U251 glioma cells through regulating the HIF-1α/LDHA pathway. This result is in line with the other studies. Charles et al. (26) rstly demonstrated the downregulation of LDHA in IDH(mt) derived BTSCs, and concluded that silencing of LDHA was associated with increased methylation of the LDHA promoter. In recent research, Victor et al. (27) disclosed that the aggressive glioma models had lost DNA mythylation in the promoters of glycolytic enzymes, especially LDHA, and have increased mRNA metabolite levels compared with the indolent model. Our results suggest that the IDH1 R132H mutation reduced the expression of HIF-1α, which then inhibited the glycolysis and proliferation of glioma cells probably through interaction with LDHA. However, the underlying mechanism needs further investigation.

Conclusions
This study investigated the role of the IDH1 R132H mutation in the proliferation and migration of glioma cells and potential underlying mechanisms. Our research indicated that IDH1 R132H mutated gliomas have decreased glycolytic capacity, which may lead to their slow growth pattern and better clinical prognosis compared with IDH1-wt gliomas. Therefore, targeting the HIF-1α/LDHA pathway may be a potential therapeutic approach for the treatment of gliomas.