The following experiments were performed in order to ascertain whether or not glucose-6-3H represents the entire glucose molecule. Permanent indwelling polyethylene cannulae were inserted into the aorta and vena cava of Sprague?Dawley rats using the method of Popovic et al.3. Several days later, when the rats had returned to their pre-operative weights, a mixture containing D-glucose-U-14C and D-glucose-6-3H was administered intra-arterially to the unanaesthetized animals in the post-absorptive state. Serial samples of venous blood were taken at intervals between 30 and 130 min after isotope administration and specific activities of the plasma glucose with respect to both carbon-14 and tritium were determined. As expected, the specific activities of the plasma glucose decreased with time, presumably because of the production of new glucose. However, as can be seen in Table 1 under Exp, 1, the 3H ; carbon-14 isotope ratio did not remain constant but decreased with time, indicating that the tritium was lost more rapidly than the carbon-14. This decreasing ratio provides evidence that the metabolism of the glucose chain cannot be followed quantitatively by a tritium tracer on carbon-6. The mechanisms which could account for this excess loss of tritium include the operation of a tritium isotope effect, cleavage of the carbon chain followed by resynthesis of glucose, or cleavage of the (6)?C?H bond of glucose.Table 1. THE 3H : 14C ISOTOPE RATIOS OF PLASMA GLUCOSE FOLLOWING INTRA-ARTERIAL ADMINISTRATION OF CARBON-H AND TRITIUM GLUCOSE MIXTURES INTO EATSExp, Isotopic glucose administered Time after isotope administration 3H ; 14C Per cent of 3H lost relative to initial 8H ; "C 0 8-3Glucose-U-I4C 30 7-5 9-61 and 60 6-5 12-2 glueose-6-3H 90 5-3 36-1110 4-2 49-4 130 4-0 518 0 13-9Glueose-0-14C 30 12-3 11-52 and 60 11-7 158 gluoose-6-5H 110 94 32-4 130 8-4 39-60 7-5Glueose-6-14C 30 6-6 12-03 and 60 6-2 173 glucose-6"8H 90 5-1 32-0110 4-7 37-3 130 4-0 46-7This represents the 3H:14C ratio of the labelled glucose before administration to the rat at time 0. When tritium is used as a tracer in biological systems, experimental results can be markedly influenced by isotope effects which result in lower reaction rates for tritium transfer as compared to protium. However, our results show that the 3H : carbon-14 ratio of plasma glucose decreases with time, giving precisely the opposite result from that which would be expected from a tritium isotope effect4,5.A change in the 3H : carbon-14 ratio could result from the formation of new glucose from three-carbon fragments derived from the administered doubly-labelled glucose. It is conceivable that tritium, but not carbon-14, could be lost from these fragments, possibly during the enolization step leading to formation of pyruvic acid, However, von Holt6 has demonstrated that in non-fasted rats only 12 per cent of the plasma glucose is derived from glucose degradation products in 120 min. Cleavage of the carbon?tritium bond could occur in an aqueous biological system, either by physical-chemical exchange or by an enzymatic hydrogen transfer system, The former possibility seems unlikely, since we established in the course of this investigation that the tritium label is stable during the synthesis of the glucosatriazole derivative used to isolate the plasma glucose. During preparation of this derivative the tritium-labelled glucose is in an aqueous medium for many hours under varying physical and chemical conditions including high temperatures (110 ?140 ).In order to demonstrate that the excess loss of tritium was due to cleavage of a (6)?C?H bond rather than cleavage of the carbon chain itself, glucose labelled with carbon-14 at the sixth carbon was administered together with D-glucose-6-3H, As can be seen in Table 1, Expts, 2 and 3, the 8H : carbon-14 isotope ratio decreased with time at rates comparable to that of the experiment with D-glucose-U-carbon-14. This strongly suggests that loss of tritium is due to cleavage of the (6)?C?H bond. One known metabolic reaction that involves cleavage of the (6)?C?H bond is the oxidation of glucose to glucu-ronic acid. However, the change in the 8H : 14C ratio indicates that approximately 40?50 per cent of the glucose pool undergoes hydrogen exchange in 130 min. It is highly unlikely that such extensive glucuronic acid formation occurs in this time.The most likely explanation for our results appears to be a hydrogen transfer reaction involving the (6)-C-3H bond of glucose. Speculation can be made as to where in the organism such transfer takes place. The fact that the decrease in the 3H : 140 ratio is found in the plasma glucose itself tends to rule out the intracellular space of extra hepatic tissues as the site of cleavage. It is well established that in the post-absorptive animal the liver is the source of plasma glucose. Once this plasma glucose enters extra-hepatic tissues, it is phosphorylated and cannot be released as glucose because of the absence of glucose-6-phosphatase in these tissues. However, it is possible that hydrogen exchange at carbon six could occur in the liver. The experiments of Bloom and Foster indicate that tritium is not removed from carbon-6 of intracellular glucose formed from glycerol 3-3H, 1,3-14C in rat liver slices and diaphragm in vitro7. In both tissues the 3H : 14C ratio of glucose isolated from glycogen was equal to that expected from the tritium and carbon-14 content of the precursor glycerol. However, the results do not eliminate the possibility that hydrogen exchange occurs in the process of transfer of the glucose in and/or out of the liver cell. Removal of tritium from carbon-6 could possibly occur at the cell surfaces of other tissues as well. In conclusion, these results indicate that glucose-6-3H cannot be used as a quantitative tracer for the glucose carbon chain in the intact rat. However, the loss of tritium observed in our experiments may reflect an important physiological event in glucose metabolism in the intact organism.This investigation was supported in part by U.S. Public Health Service research grant AM 07215?01, from the National Institute of Arthritis and Metabolic Diseases, by a research grant from the Diabetes Association of Southern California, and by an American Cancer Society Institutional grant to the University of Southern California.