Search

AMP deaminase (AMPD) plays a crucial role in adenine nucleotide metabolism. Recently we found that upregulated AMPD activity is associated with ATP depletion and contractile dysfunction under the condition of pressure overloading in the heart of a rat model of type 2 diabetes mellitus (T2DM), OLETF. Here we examined the mechanism of AMPD upregulation by T2DM. The protein level of 90-kDa full-length AMPD3 was increased in whole myocardial lysates by 55% in OLETF compared to those in LETO, a non-diabetic control. In contrast, the mRNA levels of AMPD3 in the myocardium were similar in OLETF and LETO. AMPD3 was comparably ubiquitinated in OLETF and LETO, and its degradation ex vivo was more sensitive to MG-132, a proteasome inhibitor, in OLETF than in LETO. MicroRNA array analysis revealed downregulation (>50%) of 57 microRNAs in OLETF compared to those in LETO, among which miR-301b was predicted to interact with the 3'UTR of AMPD3 mRNA. AMPD3 protein level was significantly increased by a miR-301b inhibitor and was decreased by a miR-301b mimetic in H9c2 cells. A luciferase reporter assay confirmed binding of miR-301b to the 3'UTR of AMPD3 mRNA. Transfection of neonatal rat cardiomyocytes with a miR-301b inhibitor increased 90-kDa AMPD3 and reduced ATP level. The results indicate that translational regulation by miR-301b mediates upregulated expression of cardiac AMPD3 protein in OLETF, which potentially reduces the adenine nucleotide pool at the time of increased work load. The miR-301b-AMPD3 axis may be a novel therapeutic target for intervening enegy metabolism in diabetic hearts.

These studies have examined the effect of hypoxia inducible factor 1α (HIF-1α) on nucleotide metabolism in the ischemic heart using a genetic mouse model with heart-specific and regulated expression of a stable form of HIF-1α. We find that AMP deaminase (AMPD), the entry point of the purine nucleotide cycle (PNC), is induced by HIF-1α at the level of mRNA, protein, and activity. AMP that accumulates during ischemia can be metabolized to adenosine by 5′-nucleotidase or to IMP by AMPD. Consistent with the finding of AMPD induction, adenosine accumulation during ischemia was much attenuated in HIF-1α-expressing hearts. Further investigation of nucleotide salvage enzymes found that hypoxanthine phosphoribosyl transferase (HPRT) is also upregulated in HIF-1α-expressing hearts. Treatment of hearts with an inhibitor of the PNC, hadacidin, hastens the fall of the adenylate energy charge during ischemia and the accumulation of AMP. The results provide new insight into the role of the PNC in the heart, especially as it relates to ischemia, and indicate that HIF-1α regulates nucleotide metabolism as a compensatory response to hypoxia.

Type 2 diabetes mellitus (T2DM) is often complicated with diastolic heart failure, which decompensates under increased afterload. Focusing on cardiac metabolomes, we examined mechanisms by which T2DM augments ventricular diastolic stiffness in response to pressure overloading. Pressure-volume relationships (PVRs) and myocardial metabolomes were determined at baseline and during elevation of aortic pressure by phenylephrine infusion in a model of T2DM, OLETF, and its non-diabetic control, LETO. Pressure overloading augmented diastolic stiffness without change in systolic reserve in OLETF as indicated by a left-upward shift of end-diastolic PVR. In contrast, PVRs under cardioplegic arrest in buffer-perfused isolated hearts were similar in OLETF and LETO, indicating that extracellular matrix or titin remodeling does not contribute to the afterload-induced increase in stiffness of the beating ventricle of OLETF. Metabolome analyses revealed impaired glycolysis and facilitation of the pentose phosphate pathway in OLETF. Pressure overloading significantly reduced ATP and total adenine nucleotides by 34% and 40%, respectively, in OLETF but not in LETO, while NADH-to-NAD(+) ratios were similar in the two groups. The decline in ATP by pressure overloading in OLETF was associated with increased inosine 5-monophosphate and decreased adenosine levels, being consistent with the 2.5-times higher activity of cardiac AMP deaminase in OLETF. Tissue ATP level was negatively correlated with tau of LV pressure and LVEDP. These results suggest that ATP depletion due to excessive degradation of adenine nucleotides by up-regulated AMP deaminase underlies ventricular stiffening during acute pressure overloading in T2DM hearts.

Evaluation of enzyme activities involved in nucleotide metabolism and adenosine production within different cell types can provide important information on their contribution to the overall metabolism of the heart. The following enzyme activities were determined: adenosine kinase (AK), adenosine deaminase (ADA), S-adenosylhomocysteine hydrolase (SAHH), purine nucleoside phosphorylase (PNP), AMP deaminase (AMPD), membrane 5′ nucleotidase (M5′N), AMP specific (AC5′N) and IMP specific(IC5′N) cytosolic 5′nucleotidases in (1) rat heart (n=5), (2) rat cardiomyocytes obtained by collagenase digestion (n=5), (3) human heart (n=6) obtained from explants or papillary muscles collected during heart transplantation or mitral valve replacement, and (4) human umbilical cord endothelial cells in primary culture (n=4). In the human heart, activities (μmol/min/g wet wight) were as follows: AK (0.14±0.01), ADA (0.46±0.03), SAHH (0.001±0.0003), PNP (0.43±0.08), AMPD (0.41±0.05), M5′N (1.75±0.12), IC5′N (0.21±0.03) and AC5′N (0.11±0.02). These enzyme activities were lower than those determined in the rat heart with the exception of AC5′N and IC5′N which were equal. The most prominent difference observed was for AMPD and M5′N which were nine and five-fold more active in the rat heart. Rat cardiomyocyte enzyme activities were comparable to those measured in whole rat heart with the exception of ADA (six-fold lower) and PNP (16-fold lower). Endothelial cell activities were notably different from those in the human heart particularly in the case of SAHH (nine-fold higher) and PNP (16-fold higher). These data highlight the fact that myocardial enzymes of purine metabolism are generally less active in the human heart than in rat heart, whereas the crucial pathways of intercellular adenosine production and reincorporation are comparable. The umbilical vein endothelium shows substantial catalytic capacity for adenosine formation both via intracellular and extracellular dephosphorylation and SAHH pathways. Comparison of activities in endothelial cell preparations with those in the whole heart indicate that the enzymes of adenosine breakdown pathway. ADA and PNP, appear to be located predominantly in the endothelium.

The contribution of 5′-nucleotidase and AMP-deaminase to adenine nucleotide degradation in human cardiomyocytes isolated from diseased or normal heart was investigated. The preparation used contained 30 to 50% of viable cells and the nucleotide degradation was stimulated by addition of deoxyglucose and oligomycin. To distinguish pathways of nucleotide degradation, adenosine deaminase was inhibited by erythro-9-(2-hydroxy-3-nonyl) adenine (EHNA). Under these conditions, ATP concentration was decreased by 60% after 45 min of incubation. Simultaneously, increases in intra- and extracellular catabolite concentrations have been observed. Adenosine was the predominant catabolite found in both the cells and in the extracellular medium accounting for more than 70% of all degradation products. Intracellular adenosine concentration rose to 300 times greater than that outside the cell. An increase in intra- and extracellular inosine was also seen. Only a small increase of IMP concentration was observed. No hypoxanthine accumulation was found. No significant change in initial adenine nucleotide concentrations were observed in isolated cells during aerobic incubation without deoxyglucose and oligomycin. In conclusion, a pathway involving adenosine production appears to be the principal route of nucleotide degradation in human cardiomyocytes.

The pathways of adenine nucleotide catabolism were investigated in cultured beating cardiomyocytes. The activity of the enzymes involved in AMP degradation was assayed in cell extracts. Fluxes of label from ATP to the various purine derivatives were measured in intact cells. Under physiological conditions, cells degraded AMP through deamination to IMP. IMP was rapidly degraded to inosine, hypoxanthine, xanthine and uric acid, which were effluxed from the cells. This is in accord with the fact that the activity of AMP deaminase (EC 3.5.4.6) was 7-fold that of AMP 5'-Nucleotidase (EC 3.1.3.5). Mild ATP-degradation, induced by inhibition of glycolysis by iodoacetate, caused no alterations in the degradation pathways (more than 85% through deamination to IMP). However, fast ATP-degradation (83% of adenine nucleotides/10 min), induced by simultaneous inhibition of glycolysis and electron transport (by antimycin A), caused increased dephosphorylation of AMP to adenosine (50% of total AMP-degradation). The cardiomyocyte extracts were found to contain a significant activity of purine nucleoside phosphorylase (EC 2.4.2.1). Despite the presence of hypoxanthine-guanine phosphoribosyltransferase (EC 2.4.2.8), salvage of hypoxanthine to IMP, both at physiological as well as at conditions associated with ATP degradation, was slow. The salvage of adenosine appeared to be efficient at physiological conditions, but not at fast rates of ATP degradation.

Measurements of atrial and ventricular adenine nucleotides, nucleosides and bases in dog heart indicated that normal atrial ATP, ADP and AMP levels are about one-half those of normal ventricular tissue, but atrial and ventricular tissue levels of adenosine, inosine, and hypoxanthine were about equal. With ischemia, production of adenosine and inosine was less in atrial than in ventricular muscle. In the ventricle, 5′-nucleotidase is associated with external membranes, including those of the T-tubular system and the lack of a T-tubular network in atrium suggested that there might be fewer AMP hydrolyzing sites and hence a slower rate of adenosine production in response to ischemia. This possibility was ruled out since measurements of 5′-nucleotidase on crude tissue extracts indicated that atrial tissue has 2 to 3 times the activity of ventricular tissue. Furthermore, measurements of adenosine kinase activity indicate that atrial and ventricular activities of this enzyme are about equal; hence the lower adenosine levels in atria are also not due to a higher rate of adenosine rephosphorylation. Inhibition of 5′-nucleotidase by ATP and ADP can also be eliminated as a cause of the differences in atrial and ventricular adenosine formation since the concentrations of these compounds are lower in atrial than in ventricular muscle. Finally, the lower levels of IMP and inosine in atrial tissue indicate that deamination by adenylate deaminase is not the major pathway for AMP degradation in atrial tissue. It is possible that the lower substrate (AMP) concentration for 5′-nucleotidase is responsible for the lower adenosine production by atrial tissue in response to ischemia. However, to what extent this contributed to the lower atrial adenosine levels cannot be determined from these data.

The loss of the catabolic products of adenosine triphosphate in the form of purine nucleosides and oxypurines during ischemia and subsequent reperfusion may limit adenine nucleotide regeneration. This study compared the effects of infusion of inhibitors of the major reactions involved in the degradation of adenosine triphosphate to inosine on the postischemic recovery of high energy phosphate and myocardial function. Inhibitors of adenylate kinase, 5'nucleotidase, adenosine translocase and adenosine deaminase were studied. Following 30 minutes of ischemia, only hearts infused with alpha, beta, methylene adenosine diphosphate (5' nucleotidase inhibitor) recovered significantly better ventricular function than control (p less than 0.05), but all hearts had increased adenosine triphosphate regeneration (p less than 0.05). The formation and washout of greater than 30% of the total adenine pool metabolites was not prevented by any drug. Nevertheless all manipulations of adenine metabolism resulted in recruitment of high energy phosphate during preischemic infusion.