In vivo characterization of brain metabolism by ¹H MRS, ¹³C MRS and ¹⁸FDG PET reveals significant glucose oxidation of invasively growing glioma cells

Abstract

Glioblastoma are notorious for their highly invasive growth, diffusely infiltrating adjacent brain structures that precludes complete resection, and is a major obstacle for cure. To characterize this "invisible" tumor part, we designed a high resolution multimodal imaging approach assessing in vivo the metabolism of invasively growing glioma xenografts in the mouse brain. Animals were subjected longitudinally to magnetic resonance imaging (MRI) and H spectroscopy (MRS) at ultra high field (14.1 Tesla) that allowed the measurement of 16 metabolic biomarkers to characterize the metabolic profiles. As expected, the neuronal functionality was progressively compromised as indicated by decreasing N-acetyl aspartate, glutamate and gamma-aminobutyric acid and reduced neuronal TCA cycle (-58%) and neurotransmission (-50%). The dynamic metabolic changes observed, captured differences in invasive growth that was modulated by re-expression of the tumor suppressor gene WNT inhibitory factor 1 (WIF1) in the orthotopic xenografts that attenuates invasion. At late stage mice were subjected to C MRS with infusion of [1,6- C]glucose and FDG positron emission tomography (PET) to quantify cell-specific metabolic fluxes involved in glucose metabolism. Most interestingly, this provided the first in vivo evidence for significant glucose oxidation in glioma cells. This suggests that the infiltrative front of glioma does not undergo the glycolytic switch per se, but that environmental triggers may induce metabolic reprograming of tumor cells.

In vivo 13C and 1H‐[13C] MRS studies of neuroenergetics and neurotransmitter cycling, applications to neurological and psychiatric disease and brain cancer

“…Given the much higher efficiency of ATP production from substrate oxidation versus glycolysis (on the order of 15 to 16 times greater ATP production per carbon) the results indicate that at least for a subset of human brain tumors the major role of the Warburg effect is probably not ATP production. A similar conclusion was reached in an in vivo 13 C MRS study of an animal model of infiltrative GBM .…”

“…The ability to hyperpolarize 13 C labeled substrates such as pyruvate to levels 10,000x or more than the natural polarization, has generated great excitement in the MR field. While the applications in brain have been mostly focused on animal models, recent hyperpolarized data have been acquired in human brain 15. A primary limitation for studying the brain, outside of brain cancer, is due to the blood brain barrier, consisting of the capillary endothelia, which restricts the rate of [1‐ 13 C] pyruvate entry into the brain relative to other organs and makes it rate limiting .…”

2-Deoxy-d-Glucose and Its Analogs: From Diagnostic to Therapeutic Agents

“…Due to the greater electronic similarity of fluorine to a hydroxyl group (versus the hydrogen atom in 2-DG), 18 F-DG is more reminiscent of glucose [41]. PET scanning after intravenous injection of 18 F-DG illustrates the uptake and distribution of the radiopharmaceutical, reflecting glucose uptake and metabolism throughout the body [42]. 18 F-DG is converted to 18 F-DG-6-phosphate and accumulates in tumors or organs that intensively metabolize or excrete glucose, such as the brain, heart, kidneys, ureters, and bladder [43].…”

Metabolic and transcriptomic profiles of glioblastoma invasion revealed by comparisons between patients and corresponding orthotopic xenografts in mice

“…This illustrates the differences in individual metabolite changes, induced by tumor growth in the injected side of the brain and invasion of the contralateral side, between patients, and the similarity among mice injected with tumor cells of the same patients. Absence or minimal increase of lactate (Lac) was observed for xenografts derived of most patients' tumors, in line with the invasive nature and absence of a necrotic tumor core in the mouse xenografts [21]. The growth pattern of the tumors was patient-dependent, and ranged from diffuse growth with migration across the corpus callosum and invasion of the non-injected side to well delineated tumors with only local invasion (Fig.…”

“…1 H-MRS experiments were carried out in a 14.1 Tesla animal scanner with a 28-cm horizontal bore (Agilent Technologies, Palo Alto, CA, USA) using the SPECIAL sequence (TR = 4s, TE = 2.8ms, 160 or 240 scans) [18] as previously described [21]. An axial T2-weighted image (fast spin-echo sequence, TR/TE = 5000/13 ms, FOV =18x18 mm, slice thickness 0.6 mm, 6 averages) was acquired before the 1 H-MRS for voxel positioning (VOI, 2x2x2 mm 3 ), centered in the striatum of the injected, and symmetrically, in the contralateral hemisphere.…”