Glucose-phosphates (Glc-1P and Glc-6P) are known to be important intermediate metabolites for storage and transfer of energy, being part of the glycogen synthesis.

High resolution magic angle spinning (HR-MAS) NMR is increasingly being used to metabolically characterize tissue biopsies [1]. ¹H HR-MAS NMR allows to assess these phosphate sugars qualitatively and quantitatively as a minimally invasive analytical tool without the need of extraction or separation steps, conserving the biopsy integrity. It has been shown that ¹H HR-MAS NMR can be used for studying phosphate sugar metabolic pathways, with unambiguous assignment of α-Glc-1P and Glc-6P in skeletal and cardiac muscle biopsies [2].

The aim of this 1H HR-MAS study was to investigate biopsies of skeletal muscles comparing athletes, sedentary slim and sedentary obese subjects. The results demonstrate a potential application for specifically investigating Glc-1P and Glc-6P.

Study Population

This study is part of an on-going clinical trial involving healthy sedentary and active older adults (60-80 years old) [3]. 31 skeletal muscle biopsies were obtained from the vastus lateralis by the Bergstrom technique. Out of those 31 older adults, 15 were athletes (more than 3 structured aerobic exercise sessions per week, BMI=21.6±1.4), 7 were slim sedentary (less than 1 structured aerobic exercise session per week, BMI=24.5±1.5) and 9 were obese sedentary (BMI=32.3±1.5).

HR-MAS NMR Spectroscopy

Biopsies were washed and placed together with PBS in a 4 mm rotor using a 12 μl insert. The ¹H HR-MAS NMR experiments were performed on a Bruker Avance II spectrometer operating at 500.13 MHz at 277 K at a spinning speed of 5 kHz. A CPMG pulse sequence with water presaturation (cpmgpr1d) was used for the acquisition of 1D ¹H NMR spectra.

Data Processing

1D CPMG spectra were phase- and baseline corrected using Topspin. For chemometric analysis, a total of 66 buckets (between 0.95 and 8.62 ppm) were selected, with a variable size according to the peak width. Spectral regions comprising only noise were excluded as well as pure lipid regions, in order to investigate small molecules only for this sub-study. The buckets were normalized by probabilistic quotient normalization (PQN) [4] and scaled with Pareto scaling [5]. Matlab, PLS-Toolbox and Excel were used for statistical analysis, including partial least squares discriminant analysis (PLS-DA).

Chemometric analysis of the biopsy spectra employing PLS-DA comparing the three groups, i.e. athletes, sedentary obese and sedentary slim subjects, achieved near complete separation between athletes and sedentary subjects, while obese and slim sedentary subjects overlap (Fig.1). The loading plot demonstrates that the main discriminators were creatine, glutamine and glutamate, lactate (probably with contributions from remaining lipids), and strongly also glucose and glucose-phosphates. A specific individual spectra analysis revealed that clearly detectable Glc-1P contributions (i.e. clearly above the noise level) were present in 3 of 15 spectra from athletes, in 3 of 7 spectra from slim sedentary subjects and in 6 of 9 obese sedentary subjects. Averaged spectra from all three groups (scaled to the creatine methyl-peak) are shown in Fig. 2. The insert shows the characteristic doublet of doublet resonance at 5.45ppm of Glc-1P. The Glc-1P content is slightly higher in obese than in slim sedentary subjects and the Glc-1P content is clearly higher in both sedentary groups than in athletes.¹H HR-MAS NMR allows the direct assessment of glucose-phosphates contained in skeletal muscle biopsies, as also previously shown [2]. The current study is the first example for a potential application, demonstrating differences in Glucose-phosphates between muscle tissues from athletes and sedentary subjects. The results suggest that quantitative assessment by ¹H HR-MAS NMR of Glc-1P and Glc-6P being key players in energy metabolism may prove important for metabolic studies in biopsies. The spectral quality and achievable SNR suggest that these glucose-phosphates might also be assessable in vivo by MRS methods, when nearby resonating lipid signals are sufficiently suppressed.