Mapping hypoxia-induced bioenergetic rearrangements and metabolic signaling by 18O-assisted 31P NMR and 1H NMR spectroscopy

Abstract

Brief hypoxia or ischemia perturbs energy metabolism inducing paradoxically a stress-tolerant state, yet metabolic signals that trigger cytoprotection remain poorly understood. To evaluate bioenergetic rearrangements, control and hypoxic hearts were analyzed with 18O-assisted 31P NMR and 1H NMR spectroscopy. The 18O-induced isotope shift in the 31P NMR spectrum of CrP, betaADP and betaATP was used to quantify phosphotransfer fluxes through creatine kinase and adenylate kinase. This analysis was supplemented with determination of energetically relevant metabolites in the phosphomonoester (PME) region of 31P NMR spectra, and in both aromatic and aliphatic regions of 1H NMR spectra. In control conditions, creatine kinase was the major phosphotransfer pathway processing high-energy phosphoryls between sites of ATP consumption and ATP production. In hypoxia, creatine kinase flux was dramatically reduced with a compensatory increase in adenylate kinase flux, which supported heart energetics by regenerating and transferring beta- and gamma-phosphoryls of ATP. Activation of adenylate kinase led to a build-up of AMP, IMP and adenosine, molecules involved in cardioprotective signaling. 31P and 1H NMR spectral analysis further revealed NADH and H+ scavenging by alpha-glycerophosphate dehydrogenase (alphaGPDH) and lactate dehydrogenase contributing to maintained glycolysis under hypoxia. Hypoxia-induced accumulation of alpha-glycerophosphate and nucleoside 5'-monophosphates, through alphaGPDH and adenylate kinase reactions, respectively, was mapped within the increased PME signal in the 31P NMR spectrum. Thus, 18O-assisted 31P NMR combined with 1H NMR provide a powerful approach in capturing rearrangements in cardiac bioenergetics, and associated metabolic signaling that underlie the cardiac adaptive response to stress.

Fig. 1

Sequences in labeling of cellular phosphates by ¹⁸O. On perfusion of isolated hearts with media enriched in ¹⁸O water, ¹⁸O first incorporates into inorganic phosphate (P_i) by cellular ATPases. Then, ATP synthesis by oxidative phosphorylation and glycolysis introduce ¹⁸ O-labeled P_i into -phosphate of ATP. Creatine kinase transfers ¹⁸O-labeled phosphoryl from -ATP to CrP, while adenylate kinase transfers ¹⁸O-labeled phosphoryl to βADP and βATP. In this way, the ¹⁸O-phosphoryl labeling procedure detects only newly generated molecules containing ¹⁸O-labeled phosphoryls reflecting cellular ATP turnover and net fluxes through individual phosphotransfer pathways.

Fig. 2

Reduced creatine kinase and increased adenylate kinase phosphotransfer under hypoxia. ³¹P NMR spectra, from 5 to −21 ppm, are presented in control and hypoxic hearts. Regions of spectra that correspond to CrP, βADP and βATP peaks are magnified to illustrate the ¹⁸O-induced shift following incorporation of ¹⁸O atoms. Under hypoxia, ¹⁸O incorporation into CrP, a consequence of creatine kinase phosphotransfer, was reduced, while ¹⁸O incorporation into βADP and βATP resulting from adenylate kinase phosphotransfer was increased. Incorporation of each ¹⁸O atom induces an isotope shift of 0.0250 in ³¹P NMR spectra of CrP and 0.0228 ppm in βADP, respectively. The isotope shift of βATP was different for bridging and non-bridging ¹⁸O oxygens, 0.0170 and 0.0287 ppm, respectively. ¹⁶O, ¹⁸ O₁, ¹⁸O₂and ¹⁸O₃ indicate species of phosphoryls containing 0, 1, 2 and 3 of atoms of ¹⁸O. Up to three ¹⁸O atoms, can be incorporated in phosphoryls of CrP, βADP and βATP. Peaks corresponding to bridging and non-bridging ¹⁸O oxygens of βATP are marked as one peak. PME – phosphomonoester region; Pi – inorganic phosphate; PDE – phosphodiester region.

Fig. 3

Redistribution of phosphotransfer flux between creatine kinase and adenylate kinase under hypoxic stress. Average phosphotransfer fluxes were determined from ¹⁸O-labeled phosphoryl appearance in CrP, βADP and βATP, respectively. *Indicates statistically significant differences at p < 0.01.

Fig. 4

¹H NMR-based analysis of metabolic signals following activation of adenylate kinase phosphotransfer in hypoxia. (A) Aromatic portions (ppm 8.62–8.45 and 8.37–8.18) of ¹H NMR spectra with decrease of ATP and increase of ADP, AMP, adenosine and IMP in hypoxia. (B) Average metabolite levels determined from ¹H NMR spectra. *Indicates statistically significant differences at p < 0.01.

Fig. 5

α-Glycerophosphate is a principal component of the phosphomono-ester region in ³¹P NMR spectra under hypoxia. Phosphomonoester (PME) regions of ³¹P NMR spectra in control and hypoxic hearts. Arrows indicate increase of α-glycerophosphate (αGP) and nucleotide monophosphates (NMP) in hypoxia. G6P – Glucoso-6-phosphate; PE – phoshoethanolamine; X – unknown resonance; PC – phosphocholine.

Fig. 6

Metabolic signaling events triggered by stress-induced phosphotransfer redistribution between creatine kinase and adenylate kinase. In hypoxia, decrease in CK flux activates AK-phosphotransfer with generation of metabolic signals in the form of AMP, IMP, adenosine and inosine. These metabolites adjust the activities of ATP/ADP-, AMP-, IMP-, adenosine- and inosine-sensitive cellular components, including glycolysis/glycogenolysis, and contribute to the generation of a stress-tolerant state. AMPK – AMP-activated protein kinase; K_ATP – ATP-sensitive K⁺ channel.