Rearrangement of energetic and substrate utilization networks compensate for chronic myocardial creatine kinase deficiency

Abstract

Plasticity of the cellular bioenergetic system is fundamental to every organ function, stress adaptation and disease tolerance. Here, remodelling of phosphotransfer and substrate utilization networks in response to chronic creatine kinase (CK) deficiency, a hallmark of cardiovascular disease, has been revealed in transgenic mouse models lacking either cytosolic M-CK (M-CK(-/-)) or both M-CK and sarcomeric mitochondrial CK (M-CK/ScCKmit(-/-)) isoforms. The dynamic metabolomic signatures of these adaptations have also been defined. Tracking perturbations in metabolic dynamics with (18)O and (13)C isotopes and (31)P NMR and mass spectrometry demonstrate that hearts lacking M-CK have lower phosphocreatine (PCr) turnover but increased glucose-6-phosphate (G-6-P) turnover, glucose utilization and inorganic phosphate compartmentation with normal ATP γ-phosphoryl dynamics. Hearts lacking both M-CK and sarcomeric mitochondrial CK have diminished PCr turnover, total phosphotransfer capacity and intracellular energetic communication but increased dynamics of β-phosphoryls of ADP/ATP, G-6-P and γ-/β-phosphoryls of GTP, indicating redistribution of flux through adenylate kinase (AK), glycolytic and guanine nucleotide phosphotransfer circuits. Higher glycolytic and mitochondrial capacities and increased glucose tolerance contributed to metabolic resilience of M-CK/ScCKmit(-/-) mice. Multivariate analysis revealed unique metabolomic signatures for M-CK(-/-) and M-CK/ScCKmit(-/-) hearts suggesting that rearrangements in phosphotransfer and substrate utilization networks provide compensation for genetic CK deficiency. This new information highlights the significance of integrated CK-, AK-, guanine nucleotide- and glycolytic enzyme-catalysed phosphotransfer networks in supporting the adaptivity and robustness of the cellular energetic system.

Figure 1. Altered kinetics of high-energy phosphoryl exchange and energetic communication in hearts with CK gene deletions

A, kinetics of ¹⁸O-labelled phosphoryl appearance in PCr representing CK metabolic flux in wild type (WT) versus M-CK^−/− and M-CK/ScCKmit^−/− hearts. B, kinetics of ¹⁸O-labelled phosphoryl appearance in β-phosphoryls of ATP indicating adenylate kinase metabolic flux. C, ¹⁸O-labelling of cellular inorganic phosphate (Pi) representing ATPase velocity and compartmentation of Pi. D, ¹⁸O-labelling of γ-phosphoryls of ATP indicating cellular ATP synthesis rate. E, changes in Pi/γ-ATP ¹⁸O-labelling ratio, an indicator of intracellular energetic communication between ATP utilization and synthesis sites. n = 6 in each group. *P < 0.05.

Figure 2. Total cellular ATP turnover and interchange between CK and AK phosphotransfer fluxes

Estimated rates of total ATP turnover, creatine kinase (CK) and adenylate kinase (AK) phosphoryl fluxes in wild type (WT), M-CK^−/− and M-CK/ScCKmit^−/− hearts. n = 6 in each group. *P < 0.05.

Figure 3. Phosphotransfer redistribution through guanylate and glycolytic pathways and adjustment of overall energetic dynamics in CK-deficient hearts

A, kinetics of ¹⁸O-labelled phosphoryl appearance in γ-GTP representing GTP synthesis rate in succinyl-CoA synthase (Krebs cycle) and nucleoside diphosphate kinase (NDPK) reactions in wild type (WT), M-CK^−/− and M-CK/ScCKmit^−/− hearts. B, kinetics of ¹⁸O-labelled phosphoryl appearance in β-phosphoryls of GTP representing metabolic flux through nucleoside monophosphate kinase (NMPK)-catalysed reactions. C, ¹⁸O-labelling of G-6-P signifying changes in glycolytic phosphotransfer rate.

Figure 4. Adjustments in glucose uptake and substrate utilization in CK-deficient hearts

A and B, basal glucose (2-DG) uptake in wild type (WT), M-CK^−/− and M-CK/ScCKmit^−/− hearts. C and D, insulin-stimulated glucose (2-DG) uptake in WT, M-CK^−/− and M-CK/ScCKmit^−/− hearts. The insulin-stimulated glucose uptake rates have been calculated for each individual heart for the 15 min infusion of insulin (i.e. in the time from 90 to 105 min). The averaged data for each group are presented. E and F, glucose utilization in WT, M-CK^−/− and M-CK/ScCKmit^−/− hearts (n = 5 each). *P < 0.05.

Figure 5. Increased muscle glycolytic and mitochondrial capacities and glucose tolerance in CK-deficient mice

A and B, glycolytic capacities in heart and skeletal muscles measured as maximal glucose conversion to lactate rate; C, mitochondrial capacities in heart and skeletal muscles assessed by citrate synthase activity; D, intraperitoneal glucose tolerance test (IPGTT) in WT and M-CK/ScCKmit^−/− mice. *P < 0.05, n = 5 in each group.

Figure 6. Metabolomic profiles of WT, M-CK^−/− and M-CK/ScCKmit^−/− hearts

A (left panel), PLS-DA score plot shows clear separation among groups based on metabolite levels and turnover/¹⁸O-labelling rates indicating different metabolic profiles of WT, M-CK^−/− and M-CK/ScCKmit^−/− hearts (R² = 0.87, Q² = 0.82); right panel represents plot of variable importance in the projection (VIP) signifying importance of metabolites in discriminating between metabolomic profiles of the groups in the PLS-DA model. B and C (left panels), PLS-DA analysis of WT vs. M-CK^−/− (R² = 0.95, Q² = 0.88) and WT vs. M-CK/ScCKmit^−/− (R² = 0.97, Q² = 0.92) hearts separately; right panels represent regression coefficient plots of metabolic variables in the PLS discriminating model; larger coefficient values (positive or negative) indicate a stronger correlation with group metabolic profile classification, n = 4–6 in each group.

Figure 7. Integrated model of cellular phosphotransfer circuits facilitating high-energy phosphoryl (∼P) export from mitochondria and distribution to remote cellular ATPases

Intramitochondrial, intermembrane and cytosolic phosphotransfer enzymes (CK, AK, glycolytic and NDPK/NMPK) facilitate nucleotide exchange between mitochondria and ATP consumption sites. Deletion of Mi-CK or AK2 compromises nucleotide exchange and AK2 null mutation is embryonically lethal (Spindler et al. 2002; Fujisawa et al. 2009; Zhang et al. 2010) while presence of NDPK is necessary for processing of nucleoside diphosphates (Lacombe et al. 2009). In myofibrils, phosphotransfer enzymes are positioned close to ATP utilization sites by interaction with scaffold protein DRAL/FHL-2 (down-regulated in rhabdomyosarcoma LIM domain protein/four-and-a-half LIM domain protein)attached to titin (Lange et al. 2002; Dzeja et al. 2007a). Reduction of DRAL/FHL-2 level is associated with disruption of the normal subcellular localization of metabolic enzymes in human heart failure (Bovill et al. 2009). Abbreviations: Mi-CK and M-CK, mitochondrial and cytosolic isoforms of CK, respectively; AK1 and AK2, cytosolic and mitochondrial isoforms of AK, respectively; AK3, mitochondrial matrix AK isoform; ANT, adenine nucleotide translocator; Hex, hexokinase; NDPK/NMPK, nucleoside di- and mono-phosphate kinases; DRAL/FHL-2, phosphotransfer enzyme anchor LIM domain protein; i.m. and o.m., inner and outer membranes, respectively.