Calcium-insensitive splice variants of mammalian E1 subunit of 2-oxoglutarate dehydrogenase complex with tissue-specific patterns of expression.

The 2-oxoglutarate dehydrogenase (OGDH) complex is an important control point in vertebrate mitochondrial oxidative metabolism, including in the citrate cycle and catabolism of alternative fuels including glutamine. It is subject to allosteric regulation by NADH and the ATP/ADP ratio, and by Ca(2+) through binding to the E1 subunit. The latter involves a unique Ca(2+)-binding site which includes D(114)ADLD (site 1). Here, we describe three splice variants of E1 in which either the exon expressing this site is replaced with another exon (loss of site 1, LS1) or an additional exon is expressed leading to the insertion of 15 amino acids just downstream of site 1 (Insert), or both changes occur together (LS1/Insert). We show that all three variants are essentially Ca(2+)-insensitive. Comparison of massive parallel sequence (RNA-Seq) databases demonstrates predominant expression of the Ca(2+)-sensitive archetype form in heart and skeletal muscle, but substantial expression of the Ca(2+)-insensitive variants in brain, pancreatic islets and other tissues. Detailed proteomic and activity studies comparing OGDH complexes from rat heart and brain confirmed the substantial difference in expression between these tissues. The evolution of OGDH variants was explored using bioinformatics, and this indicated that Ca(2+)-sensitivity arose with the emergence of chordates. In all species examined, this was associated with the co-emergence of Ca(2+)-insensitive variants suggesting a retained requirement for the latter in some settings. Tissue-specific expression of OGDH splice variants may thus provide a mechanism that tunes the control of the enzyme to the specialized metabolic and signalling needs of individual cell types.

Studies on the regulation of the human E1 subunit of the 2-oxoglutarate dehydrogenase complex, including the identification of a novel calcium-binding site.

The regulation of the 2-oxoglutarate dehydrogenase complex is central to intramitochondrial energy metabolism. In the present study, the active full-length E1 subunit of the human complex has been expressed and shown to be regulated by Ca2+, adenine nucleotides and NADH, with NADH exerting a major influence on the K0.5 value for Ca2+. We investigated two potential Ca2+-binding sites on E1, which we term site 1 (D114ADLD) and site 2 (E139SDLD). Comparison of sequences from vertebrates with those from Ca2+-insensitive non-vertebrate complexes suggest that site 1 may be the more important. Consistent with this view, a mutated form of E1, D114A, shows a 6-fold decrease in sensitivity for Ca2+, whereas variant ∆site1 (in which the sequence of site 1 is replaced by A114AALA) exhibits an almost complete loss of Ca2+ activation. Variant ∆site2 (in which the sequence is replaced with A139SALA) shows no measurable change in Ca2+ sensitivity. We conclude that site 1, but not site 2, forms part of a regulatory Ca2+-binding site, which is distinct from other previously described Ca2+-binding sites.

Regulation of mitochondrial dehydrogenases by calcium ions.

Studies in Bristol in the 1960s and 1970s, led to the recognition that four mitochondrial dehydrogenases are activated by calcium ions. These are FAD-glycerol phosphate dehydrogenase, pyruvate dehydrogenase, NAD-isocitrate dehydrogenase and oxoglutarate dehydrogenase. FAD-glycerol phosphate dehydrogenase is located on the outer surface of the inner mitochondrial membrane and is influenced by changes in cytoplasmic calcium ion concentration. The other three enzymes are located within mitochondria and are regulated by changes in mitochondrial matrix calcium ion concentration. These and subsequent studies on purified enzymes, mitochondria and intact cell preparations have led to the widely accepted view that the activation of these enzymes is important in the stimulation of the respiratory chain and hence ATP supply under conditions of increased ATP demand in many stimulated mammalian cells. The effects of calcium ions on FAD-isocitrate dehydrogenase involve binding to an EF-hand binding motif within this enzyme but the binding sites involved in the effects of calcium ions on the three intramitochondrial dehydrogenases remain to be fully established. It is also emphasised in this article that these three dehydrogenases appear only to be regulated by calcium ions in vertebrates and that this raises some interesting and potentially important developmental issues.

Insulin stimulation of pyruvate dehydrogenase in adipocytes involves two distinct signalling pathways.

In isolated rat adipocytes, the insulin stimulation of pyruvate dehydrogenase can be partially inhibited by inhibitors of PI3K (phosphoinositide 3-kinase) and MEK1/2 (mitogen-activated protein kinase/extracellular signal-regulated kinase kinase). In combination, U0126 and wortmannin completely block the insulin stimulation of pyruvate dehydrogenase. It is concluded that the effect of insulin on pyruvate dehydrogenase in rat adipocytes involves two distinct signalling pathways: one is sensitive to wortmannin and the other to U0126. The synthetic phosphoinositolglycan PIG41 can activate pyruvate dehydrogenase but the activation is only approx. 30% of the maximal effect of insulin. This modest activation can be completely blocked by wortmannin alone, suggesting that PIG41 acts through only one of the pathways leading to the activation of pyruvate dehydrogenase.

Reciprocal feedback regulation of insulin receptor and insulin receptor substrate tyrosine phosphorylation by phosphoinositide 3-kinase in primary adipocytes.

Signalling by the insulin receptor substrate (IRS) proteins is critically dependent on the tyrosine phosphorylation of specific binding sites that recruit Src homology 2 (SH2)-domain-containing proteins, such as the p85 subunit of phosphoinositide 3-kinase (PI 3-kinase), the tyrosine phosphatase SHP-2 and the adapter protein Grb2. Here we show that stimulation by insulin of freshly isolated primary adipocytes resulted in the expected rapid tyrosine phosphorylation of the insulin receptor, IRS-1 and IRS-3. Inhibition of PI 3-kinase enhanced the insulin-stimulated phosphorylation of IRS-1 on (i) Tyr(612) and Tyr(941) (p85 binding sites), concomitant with an increased association of the p85 subunit of PI 3-kinase; (ii) Tyr(896) (a Grb2 binding site); and (iii) Tyr(1229) (an SHP-2 binding site), although little or no binding of SHP-2 to IRS-1 was detectable under any conditions. In contrast, inhibition of PI 3-kinase led to a decrease in insulin-stimulated p85 binding to IRS-3, but had no effect on SHP-2 binding. Furthermore, insulin-induced insulin receptor tyrosine phosphorylation, phosphorylation of Tyr(1158) and insulin receptor tyrosine kinase activity were all reduced by inhibition of PI 3-kinase at later time points (>or=20 min). The results demonstrate that, in primary adipocytes, PI 3-kinase feedback control of signalling by the insulin receptor and IRS proteins is multifaceted and reciprocal, illustrating the complexity of predicting the net flux of the insulin signal(s) through the IRS proteins.