2954

Quantification of Cerebral Metabolic Rates of 17O-Labeled Glucose in Mouse Brain with Dynamic 17O-MRS
Robert Borowiak1,2,3, Wilfried Reichardt1,2,3, Dmitry Kurzhunov1, Christian Schuch4, Benjamin Görling5, Dieter Leibfritz6, Jochen Leupold1, Thomas Lange1, Helge Haas7, Jens Timmer7, and Michael Bock1

1Dept. of Radiology, Medical Physics, Medical Center-University of Freiburg, Germany, Freiburg, Germany, 2German Cancer Consortium (DKTK), Heidelberg, Germany, Heidelberg, Germany, 3German Cancer Research Center (DKFZ), Heidelberg, Germany, Heidelberg, Germany, 4NUKEM Isotopes Imaging GmbH, 5Bruker BioSpin GmbH, 6Faculty of Medicine, University of Tübingen, Germany, 7Institute of Physics, University of Freiburg, Germany

Synopsis

We studied the chemical exchange kinetics of 17O-labeled glucose at the C1 and the C6 position with dynamic 17O-MRS. A profile likelihood analysis is performed to determine identifiability and confidence intervals of the metabolic rate CMRGlc. The exchange experiments confirm that the C6-17OH label is transferred via glycolysis exclusively by the enzyme enolase into the metabolic end product H217O, while C1-17OH ends up in water via direct hydrolysis as well as via glycolysis. From H217O-concentration time-courses cerebral metabolic rates of CMRGlc = 0.05‑0.08 µmol/g/min are obtained which are in of the same order of magnitude as 18F-FDG PET.

Introduction

Malignant tumors predominately gain energy by anaerobic glycolysis 1. Currently, the clinical gold standard to assess glucose metabolism is positron emission tomo­gra­phy (PET) which uses the radioactively labeled [18F]-fluor­deoxy­glucose (FDG). Recently, we have performed dynamic 17O-MRS of 17O-labeled glucose for the first time to follow up glycolysis in mouse brain 2. The purpose of this study was to further investigate the dynamics of glucose labeled with 17O at the C1 (Glc-1, 68 % labeled) and the C6 position (Glc-6, 43 % labeled) in vivo using dynamic 17O-MRS at ultra-high field. Two representative in vivo Glc-6 data sets were acquired as described in 2, and profile likelihood analysis (PL) was performed 3–6 to reliably determine metabolic rates of glucose consumption (CMRGlc) from the recorded time dependent course of the H217O resonances using a pharmacokinetic model.

Material and Methods

The 1-OH group at the anomeric C1 carbon of glucose (Glc-1) undergoes a known temperature and pH-dependent and concentration-independent chemical exchange with unlabeled water in aqueous solution7,8. Under physiological conditions the OH-group at the C6 position (Glc-6) cannot be replaced via chemical exchange in aqueous solutions, and no enzyme-catalyzed reaction is reported in the literature to substitute the C6-OH group in mammalians. However, recently we could show2 that the C6-OH label is transferred in the glycolytic downstream by the enzyme enolase into the metabolic end product H217O (Figure 1).

Exchange Measurements

To corroborate exchange dynamics, in a phantom experiment two 55 mM aqueous solutions of Glc-1 and Glc‑6 dissolved in phosphate-buffered saline PBS (pH = 7.4, Sigma Aldrich) were prepared. Dynamic 17O-MRS was performed of the solutions with a 500 MHz spectrometer (Avance III 500, Bruker Biospin) over up to 200 min. Each spectrum was measured with an FID sequence at a body temperature (37°C) with the following parameters: 90°-pulse dura­tion Tpulse = 21.5 µs, acquisition delay 10 µs, TE = 21 µs, TR = 50 ms, spectral band width BW = 31.25 kHz (461 ppm). Within the acquisition time of Tacq = 33 ms each FID was sampled with 2048 points and a dwell time of 16 µs. In total, 1024 FID signals were averaged per spectrum resulting in a measurement time of 1 min.

Model Fit and Profile Likelihood Analysis

With the exchange rates of the phantom experiments it was investigated whether CMRGlc can be reliably determined from dynamic 17O MRS data. For this, a pharmacokinetic model was used 9 that requires an input function with model parameters α and ρ. These parameters were estimated from glucose tolerance tests 10 in mice after intravenous injection of unlabeled glucose. A profile likelihood analysis was then performed to assess whether CMRGlc can be determined reliably from the time course of the H217O-resonances; for this, it was considered that 1mol Glc-6 is converted into 1mol H217O during glucose metabolism.

Results and Discussion

In the dynamic Glc-6 experiment (Figure 2a) neither a signal increase of the H217O-resonance nor a decrease of the 6-OH resonance is observed which proves that exchange of hydroxyl groups at C-6 is kinetically inhibited, whereas in the Glc-1 exchange experiment (Figure 2b,c) a signal increase of 3.5 % is observed within a measurement time of 200 min. Thus, the C6-OH label will show up in water in vivo via glycolysis exclusively, while C1-OH ends up in H217O via either direct hydrolysis or glycolysis. Moreover this result indicates that the in vivo conversion rate of Glc-1 into H217O due to chemical exchange with water in blood is expected to be less than the metabolic rate 2. As described in 11, α was estimated to 0.32 from the Glc-6 enrichment k = 43 %, the baseline and maximum concentration of the blood sugar measurements. The exponential fit to the blood sugar measurement yielded ρ-values of 0.033 /0.031 min-1 (Figure 3). In the PL analysis CMRGlc rates in the range of 0.05‑0.08 µmol/g/min were obtained (Fig. 5a,b) from the H217O-concentration-time courses (Figure 4). Note that similar metabolic rates of CMRGlc = 0.06 µmol/g/min are obtained using a simplified model as proposed in 12. The deviations from the literature value 13 0.26 ± 0.10 µmol/g/min (18F-FDG PET, mouse, 1.0 % iso­flurane anaesthesia) might be due to imperfections of the pharmacokinetic model and uncertainties of α and ρ-values.

Outlook

Although 17O-labeled glucose is currently less cost-effective than enriched 13C-glucose, oxygen-17 is a promising tracer to investigate novel metabolic pathways, which might provide enhanced sensitivity compared to established 13C-MRS methods 14. In a future step dynamic 17O-MRS will be applied in a mouse model to monitor the glucose turnover in tumors.

Acknowledgements

Support from NUKEM Isotopes Imaging GmbH is gratefully acknowledged

References

1. O. Warburg, On the origin of cancer cells, Science. 123 (1956) 309–314.

2. R. Borowiak, R. Wilfried, D. Kurzhunov, C. Schuch, J. Leupold, T. Lange, M. Reisert, A. Krafft, E. Fischer, M. Bock, Initial investigation of glucose metabolism in mouse brain using 17O-glucose and dynamic 17O-MRS, In proceedings of 24rd Annunal Meeting of ISMRM Singapore 2015, p. 3964.

3. D. Kurzhunov, R. Borowiak, H. Hass, P. Wagner, A.J. Krafft, J. Timmer, M. Bock, Quantification of oxygen metabolic rates in Human brain with dynamic 17O MRI: Profile likelihood analysis, Magn. Reson. Med. 2017.

4. A. Raue, C. Kreutz, T. Maiwald, J. Bachmann, M. Schilling, U. Klingmüller, J. Timmer, Structural and practical identifiability analysis of partially observed dynamical models by exploiting the profile likelihood, Bioinformatics. 25 (2009) 1923–1929.

5. A. Raue, B. Steiert, M. Schelker, C. Kreutz, T. Maiwald, H. Hass, J. Vanlier, C. Tönsing, L. Adlung, R. Engesser, W. Mader, T. Heinemann, J. Hasenauer, M. Schilling, T. Höfer, E. Klipp, F. Theis, U. Klingmüller, B. Schöberl, J. Timmer, Data2Dynamics: a modeling environment tailored to parameter estimation in dynamical systems, Bioinformatics. 31 (2015) 3558–3560.

6. A.C. Hindmarsh, P.N. Brown, K.E. Grant, S.L. Lee, R. Serban, D.E. Shumaker, C.S. Woodward, SUNDIALS: Suite of nonlinear and differential/algebraic equation solvers, ACM Trans Math Softw. 31 (2005) 363–396.

7. J.M. Risley, R.L. Van Etten, Kinetics of oxygen exchange at the anomeric carbon atom of D-glucose and D-erythrose using the oxygen-18 isotope effect in carbon-13 nuclear magnetic resonance spectroscopy, Biochemistry (Mosc.). 21 (1982) 6360–6365.

8. T.L. Mega, S. Cortes, R.L. Van Etten, The oxygen-18 isotope shift in carbon-13 nuclear magnetic resonance spectroscopy. 13. oxygen exchange at the anomeric carbon of D-glucose, D-mannose, and D-fructose, J. Org. Chem. 55 (1990) 522–528.

9. I.C. Atkinson, K.R. Thulborn, Feasibility of mapping the tissue mass corrected bioscale of cerebral metabolic rate of oxygen consumption using 17-oxygen and 23-sodium MR imaging in a human brain at 9.4 T, NeuroImage. 51 (2010) 723–733.

10. S. Andrikopoulos, A.R. Blair, N. Deluca, B.C. Fam, J. Proietto, Evaluating the glucose tolerance test in mice, Am. J. Physiol. - Endocrinol. Metab. 295 (2008) E1323–E1332.

11. S.M. Fitzpatrick, H.P. Hetherington, K.L. Behar, R.G. Shulman, The flux from glucose to glutamate in the rat brain in vivo as determined by 1H-observed, 13C-edited NMR spectroscopy, J. Cereb. Blood Flow Metab. 10 (1990) 170–179.

12. N. Zhang, X.-H. Zhu, H. Lei, K. Ugurbil, W. Chen, Simplified Methods for Calculating Cerebral Metabolic Rate of Oxygen Based on 17O Magnetic Resonance Spectroscopic Imaging Measurement during a Short 17O2 Inhalation, J. Cereb. Blood Flow Metab. 24 (2004) 840–848.

13. H. Toyama, M. Ichise, J.-S. Liow, K.J. Modell, D.C. Vines, T. Esaki, M. Cook, J. Seidel, L. Sokoloff, M.V. Green, R.B. Innis, Absolute quantification of regional cerebral glucose utilization in mice by 18F-FDG small animal PET scanning and 2-14C-DG autoradiography, J. Nucl. Med. 45 (2004) 1398–1405.

14. R.A. de Graaf, P.B. Brown, D.L. Rothman, K.L. Behar, Natural abundance 17O NMR spectroscopy of rat brain in vivo, J. Magn. Reson. 193 (2008) 63–67.


Figures

Figure 1: Degradation of 17O-labeled glucose at the C-1 (green) and C-6 positions (red) via glycolysis in ten steps to the final product pyruvate is shown. Chemical exchange of the C1-OH label with water in blood can take place before glycolysis. In the end of the TIM reaction two 17O-labeled GAP molecules (blue) are formed from one FPB molecule. Note that both glucose isotopologues (Glc-1 and Glc-6) lead to a labeled and an unlabeled GAP. Furthermore, each GAP is converted into 2-phosphoglycerate (2PG). Finally, H217O is cleaved off from each 2PG molecule by the enzyme enolase to form phosphoenolpyruvate (PEP).

Figure 2: Two representative 17O MRS (f0 = 67.8 MHz) spectra are shown from the start (tstart = 0 min, blue) and end (tend = 200 min, red) of the (a) Glc-6 (b) Glc-1 chemical exchange experiment performed with a temporal resolution of 1 min. The α (36 ± 1 ppm) and β (47 ± 1 ppm) forms of the anomeric hydroxyl oxygen can be detected at physiological temperature (37°C). c) Normalized signal dynamics (peak height) of the H217O (set to 0 ppm, line width FWHM = 1 ppm) resonance are shown over the time course of 200 min.

Figure 3: Time courses of two venous blood glucose concentration experiments (red, blue) after administration of 80 mg of unlabeled glucose in 200 µl (0.9 % NaCl) and fits (solid line) with an exponential decay. In both blood sugar measurements the glucose level increases in­stantaneously to its maximum 42/ 36 mM value and then returns exponentially to the mean baseline concentration of 11/ 9 mM. Increase of glucose concentration level from baseline is indicated as a dashed line.

Figure 4: Two representative H217O concentration time-courses for the Glc-6 experiments Exp1 and Exp2 and pharmacokinetic model fits in red respectively blue and simplified model fits are shown. Glucose bolus (80 mg Glc-6 dissolved in 200 µl 0.9 % NaCl) was given at t = 27 min. Note that presented data was acquired using the same experimental parameters and setup as described in 2. Both concentration-time curves show very similar dynamics and initial slopes.

Figure 5: a) Profile likelihood analysis of the parameters CMRGlc, KL and KG of the pharmacokinetic model. Confidence intervals are indicated by the red dashed lines. b) The identifiability of the rates CMRGlc, KL and KG is proved by finite confidence intervals. The rates CMRGlc, KL and KG have units of µmol/g/min.

Proc. Intl. Soc. Mag. Reson. Med. 25 (2017)
2954