Tracing hepatic gluconeogenesis relative to citric acid cycle activity in vitro and in vivo. Comparisons in the use of [3-13C]lactate, [2-13C]acetate, and alpha-keto[3-13C]isocaproate.

The validity of the use of a carbon tracer for investigating liver intermediary metabolism in vivo requires that the labeling pattern of liver metabolites not be influenced by metabolism of the tracer in other tissues. To identify such specific tracer, livers from 48-h starved rats were perfused with recirculating buffer containing [3-13C]lactate, [2-13C]acetate, or alpha-keto[3-13C]isocaproate. Conscious 48-h starved rats were infused with the same tracers for 5 h. The labeling patterns of liver glutamate and extracellular glucose were assayed by gas chromatography-mass spectrometry. In vivo data were corrected for 13CO2 reincorporation into C-1 of glutamate and C-3 and C-4 of glucose, using data from control rats infused with NaH13CO3. With [3-13C]lactate the labeling pattern of liver glutamate was the same in perfused organs and in vivo. In contrast, with [2-13C]acetate and alpha-keto[3-13C]isocaproate the labeling pattern of liver glutamate in vivo was clearly influenced by the expected labeling pattern of citric acid cycle intermediates formed in non-gluconeogenic organs, presumably glutamine made in muscle. Indeed, the labeling pattern of plasma glutamine and liver glutamate were similar in experiments with [3-13C]lactate but different in experiments with [2-13C]acetate and alpha-keto[3-13C]isocaproate. Similar conclusions were drawn from the labeling patterns of glucose. Therefore, labeled lactate appears as the best tracer for studies of liver intermediary metabolism in vivo. Our data also show that a substantial fraction of alpha-ketoisocaproate metabolism occurs in peripheral tissues.