0495

Metabolic reprogramming associated with IDH1-targeted treatments in low-grade glioma cell models: a ¹H and ¹³C MRS study

Abigail R Molloy¹, Chloé Najac¹, Aliya Lakhani¹, Elavarasan Subramani¹, Georgios Batsios¹, Anne Marie Gillespie¹, Russell O Pieper^2,3, Pavithra Viswanath¹, and Sabrina M Ronen^1,3
¹Radiology and Biomedical Imaging, University of California, San Francisco, San Francisco, CA, United States, ²Department of Neurological Surgery, University of California, San Francisco, San Francisco, CA, United States, ³Brain Tumor Research Center, University of California, San Francisco, San Francisco, CA, United States

Synopsis

Mutant IDH1 (IDH1mut) drives glioma development, and targeted IDH1mut inhibitors show promising results in clinical trials. However, treatment is not associated with tumor shrinkage, and there is an urgent need for early imaging biomarkers of response. Our studies in IDH1mut-expressing glioma cells indicate that treatment with IDH1mut inhibitors leads to ¹H and ¹³C-MRS-detectable metabolic changes. Specifically, we show a decrease in 2-HG and increase in glutamate, as well as an increase in metabolic flux from glutamine to glutamate. Furthermore, hyperpolarized [1-¹³C] α-ketoglutarate can probe these alterations in metabolism. This identifies potential non-invasive biomarkers of response to IDH1mut inhibition in glioma.

Introduction

Mutations in the cytosolic enzyme isocitrate dehydrogenase 1 (IDH1) are reported in 70-90% of low-grade gliomas and secondary glioblastomas^{1, 2}. Mutant IDH1 (IDH1mut) induces MRS-detectable metabolic reprogramming and catalyzes the reduction of α-ketoglutarate (α-KG) to 2-hydroxyglutarate (2-HG), an oncometabolite that drives tumorigenesis^{1, 2}. Inhibition of IDH1mut is therefore an attractive therapeutic approach and targeted inhibitors of IDH1 and pan-IDH1/2 such as AG-120 and AG-881, respectively, have shown promising results in clinical trials for gliomas^{7, 8}. However, response is typically associated with inhibition in tumor growth rather than tumor shrinkage. Therefore, there is an urgent need to identify non-invasive methods of imaging response to IDH1mut inhibitor therapy. Prior work from our laboratory has revealed the role of 2-HG in inducing a broad magnetic resonance spectroscopy (MRS)-detectable metabolic reprogramming in IDH1mut glioma cells^3-6. Therefore, the goal of this study was to examine the utility of ¹H and ¹³C magnetic resonance spectroscopy (MRS) to non-invasively image response to IDH1mut inhibition in low-grade gliomas.

Methods

Two genetically engineered cell lines were transduced to express IDH1mut (U87IDH1mut and NHAIDH1mut) and maintained in culture as described³. Cells were labeled with [1-¹³C] glucose or [3-¹³C] glutamine⁴ and treated every 24 hours with 1µM AG-120, 1µM AG-881, or DMSO (vehicle, >0.1%) for 72 hours. Metabolites were extracted using the dual-phase extraction method as described⁹ and ¹H-MRS (90˚ FA, 3s TR, 384 acquisitions) and ¹³C-MRS (30˚ FA, 3s TR, 2048 acquisitions) spectra were acquired using a 500MHz Bruker Avance spectrometer. Detectable metabolites were quantified by peak integration using Mnova (Mestrelab), corrected for saturation effects and NOE and normalized to cell number and external reference (43mM TSP)³. Statistical significance of differences was determined using an unpaired Student’s t-test, with p<0.05 considered significant. For hyperpolarized ¹³C-MRS experiments, [1-¹³C] α-ketoglutarate was prepared as described⁶, polarized (Oxford HyperSense DNP), dissolved in isotonic buffer, and injected into cell suspensions or lysates in a 5mm NMR tube. ¹³C-MRS spectra were acquired every 3 seconds for 300 seconds using a 1.5T pulsar spectrometer or 500MHz Varian spectrometer. Spectrophotometric enzyme assays were performed on an Infinite m200 spectrophotometer (Tecan Systems) using appropriate activity assay kits (Biovision).

Results and Discussion

Prior to investigating the metabolic effects of AG-120 and AG-881, we confirmed that treatment results in a significant drop (30-70%) in IDH1mut activity in both cell models (NHAIDH1mut and U87IDH1mut) compared with DMSO-treated control cells (Fig. 1A). Additionally, IDH1mut inhibition was associated with a significant drop (40-50%) in the number of colonies produced by both cell models (Fig. 1B).

As expected, our ¹H-MRS data indicated that treatment of both cell models with AG-120 or AG-881 induced a significant decrease in steady-state 2-HG compared with control cells. Interestingly however, we also observed a significant increase in steady-state glutamate and phosphocholine, pointing to broader changes in the metabolism of treated cells and a unique MRS-detectable signature (Fig. 2A-2B, data shown only for NHAIDH1mut model; similar data observed for U87IDH1mut).

To further evaluate whether changes in steady-state levels of glutamate and 2-HG were associated with altered metabolic fluxes, we probed glutamate and 2-HG synthesis from [1-¹³C] glucose or [3-¹³C] glutamine, the main metabolic precursors of these metabolites³. ¹³C-MRS data showed a significant increase in the flux of [3-¹³C] glutamine to glutamate following AG-120 or AG-881 treatment in both cell models (Fig. 3A&3C). The flux of [1-¹³C] glucose to glutamate remained unchanged in the NHAIDH1mut model (Fig. 3B&3C) and was slightly increased in the U87IDH1mut model. Fluxes from both [1-¹³C] glucose and [3-¹³C] glutamine to 2-HG were significantly decreased following treatment in both cell models (Fig. 3D).

Based on these results, we dynamically probed the metabolism of glutamate and 2-HG from hyperpolarized [1-¹³C] α-ketoglutarate. We found an increase in metabolism to glutamate (Fig. 4A-4B) and a decrease in metabolism to 2-HG (Fig. 5A-5B) from [1-¹³C] α-ketoglutarate in treated cells, consistent with the above-mentioned labeling and steady-state results.

Taken together, our studies indicate that IDH1mut inhibition induces unique MRS-detectable metabolic alterations that can potentially be exploited for early non-invasive, clinically translatable detection of response to emerging IDH1mut inhibitors.

Conclusion

In this study, we have identified potential ¹H and ¹³C-MRS detectable biomarkers of response to two IDH1mut inhibitors in glioma. Although further studies are needed to evaluate the utility of these biomarkers in vivo, we expect that hyperpolarized [1-¹³C] α-ketoglutarate could serve as a metabolic imaging probe for response to treatment.

Acknowledgements

This work is supported by NIH R01CA197254, NIH R01CA172845, American Brain Tumor Association grant BRF1700006, UCSF Brain Tumor Loglio Collective

References

1. Yang, H., et al., IDH1 and IDH2 mutations in tumorigenesis: mechanistic insights and clinical perspectives. Clin Cancer Res, 2012. 18(20): p. 5562-71.

2. Dang, L., et al., Cancer-associated IDH1 mutations produce 2-hydroxyglutarate. Nature, 2010. 465(7300): p. 966.

3. Izquierdo-Garcia, J.L., et al., Metabolic reprogramming in mutant IDH1 glioma cells. PLoS One, 2015. 10(2): p. e0118781.

4. Izquierdo-Garcia, J.L., et al., IDH1 Mutation Induces Reprogramming of Pyruvate Metabolism. Cancer Res, 2015. 75(15): p. 2999-3009.

5. Chaumeil, M.M., et al., Hyperpolarized [1-13C] glutamate: a metabolic imaging biomarker of IDH1 mutational status in glioma. Cancer Res, 2014. 74(16): p. 4247-57.

6. Chaumeil, M.M., et al., Non-invasive in vivo assessment of IDH1 mutational status in glioma. Nat Commun, 2013. 4: p. 2429.

7. Yen, K., et al., Abstract B126: AG-881, a brain penetrant, potent, pan-mutant IDH (mIDH) inhibitor for use in mIDH solid and hematologic malignancies. Mol Cancer Ther.

8. Mellinghoff, I.K., et al., ACTR-46. AG120, a first-in-class mutant IDH1 inhibitor in patients with recurrent or progressive IDH1 mutant glioma: results from the phase 1 glioma expansion cohorts. Neuro Oncol. 18(suppl_6): p. vi12.

9. Ronen, S.M., et al., Magnetic resonance detects changes in phosphocholine associated with Ras activation and inhibition in NIH 3T3 cells. Br J Cancer, 2001. 84(5): p. 691-6.

Figures

Figure 1. AG-120 and AG-881 affect IDH1mut enzyme activity and clonogenicity. (A) Spectrophotometric enzyme assays show a significant decrease in IDH1mut activity following treatment with AG-120 or AG-881 in both NHAIDH1mut and U87IDH1mut cells. (B) Clonogenicity assays show a significant decrease in the number of colonies formed following IDH1mut inhibition in both cell models.

Figure 2. ¹H-MRS spectra show AG-120 and AG-881 affect steady-state metabolite levels. (A) Representative ¹H-MRS spectra of control and treated NHAIDH1mut cells. (B) Quantification of steady-state 2-HG, glutamate, and phosphocholine shows significant metabolic differences between control and treated cells.

Figure 3. ¹³C labeling shows that AG-120 and AG-881 alter metabolic fluxes to glutamate and 2-HG. Representative ¹³C-MRS spectra of control and treated NHAIDH1mut cells labeled with (A) [3-¹³C] glutamine and (B) [1-¹³C] glucose (region of labeled glutamate and 2-HG peaks expanded). (C) Quantification of ¹³C glutamate peaks suggests increased metabolic flux from [3-¹³C] glutamine to glutamate. (D) Quantification of ¹³C 2-HG peaks shows decreased metabolic flux from both [1-¹³C] glucose and [3-¹³C] glutamine to 2-HG.

Figure 4. Hyperpolarized ¹³C-MRS shows that AG-881 affects [1-¹³C] glutamate production. (A) Representative ¹³C-MRS spectral array of [1-¹³C] glutamate production from hyperpolarized [1-¹³C] α-ketoglutarate in NHAIDH1mut cells. (B) Quantification of ¹³C-MRS spectra shows increased [1-¹³C] glutamate production following treatment.

Figure 5. Hyperpolarized ¹³C-MRS shows that AG-881 affects [1-¹³C] 2-HG production. (A) Representative ¹³C-MRS spectral array of [1-¹³C] 2-HG production from hyperpolarized [1-¹³C] α-ketoglutarate in NHAIDH1mut cells. (B) Quantification of ¹³C-MRS spectra shows decreased [1-¹³C] 2-HG production following treatment.

Proc. Intl. Soc. Mag. Reson. Med. 28 (2020)

0495