Renal Gluconeogenesis Its Importance in Human Glucose Homeostasis
Total Page:16
File Type:pdf, Size:1020Kb
Reviews/Commentaries/Position Statements REVIEW ARTICLE Renal Gluconeogenesis Its importance in human glucose homeostasis JOHN E. GERICH, MD HANS J. WOERLE, MD Various isotopic methods have been CHRISTIAN MEYER, MD MICHAEL STUMVOLL, MD used to assess the proportion of overall glucose release attributable to gluconeoge- nesis in humans. The ingenious approach developed by Landau et al. (4), which uses the ratio of enrichments of the C-2 carbon Studies conducted over the last 60 years in animals and in vitro have provided considerable evi- to the C-5 carbon of plasma glucose after dence that the mammalian kidney can make glucose and release it under various conditions. ingestion of deuterated water, appears to be Until quite recently, however, it was generally believed that the human kidney was not an the most widely accepted. Investigators important source of glucose except during acidosis and after prolonged fasting. This review will summarize early work in animals and humans, discuss methodological problems in assessing using this approach have found that gluco- renal glucose release in vivo, and present results of recent human studies that provide evidence neogenesis accounted for 54 ± 2% of all that the kidney may play a significant role in carbohydrate metabolism under both physio- glucose released into the circulation of logical and pathological conditions. overnight-fasted normal volunteers (5). These results are in excellent agreement Diabetes Care 24:382–391, 2001 with those predicted both from NMR stud- ies of hepatic glycogen depletion (2) and from a stable isotope approach using indi- o function effectively as a source of This lactate and the lactate generated via rect calorimetry (51 ± 5%) (6), but they are fuel in the brain, renal medulla, and glycolysis of glucose from plasma by blood higher than those reported using mass iso- nucleated blood cells and to supple- cells, the renal medulla, and other tissues topomer distribution analysis during infu- T 13 ϳ ment energy provided to other tissues can be absorbed by gluconeogenic organs sions of [2- C]glycerol ( 36%) (3). (e.g., muscle and splanchnic organs) by and re-formed into glucose. Only two organs in the human body— free fatty acids and amino acids, glucose is Recent studies using nuclear magnetic the liver and the kidney—possess suffi- normally released into the circulation of resonance (NMR) spectroscopy of changes cient gluconeogenic enzyme activity and humans who were fasted overnight in hepatic glycogen content (2) indicate glucose-6-phosphatase activity to enable (postabsorption) at a rate of 10–11 µmol и that in overnight-fasted normal volunteers, them to release glucose into the circulation kgϪ1 и minϪ1 (1). This release of glucose is net hepatic glycogenolysis occurred at a as a result of gluconeogenesis. As we will the result of one of two processes: rate of ϳ5.5 µmol и kgϪ1 и minϪ1 and later discussed, a wealth of animal experi- glycogenolysis and gluconeogenesis. accounted for 45 ± 6% of the overall release ments performed over the last 60 years Glycogenolysis involves the breakdown of of glucose into the circulation, which was have provided evidence that both the liver glycogen to glucose-6-phosphate and its measured isotopically. As indicated earlier, and the kidney release glucose into the cir- subsequent hydrolysis by glucose-6-phos- only the liver contains appreciable glycogen culation under a variety of conditions (7). phatase to free glucose. Gluconeogenesis and glucose-6-phosphatase, making it the Nevertheless, until quite recently, it was involves the formation of glucose-6-phos- only organ that can directly release glucose thought that the liver was the sole site of phate from precursors such as lactate, glyc- as a result of glycogen breakdown. Thus, gluconeogenesis in normal postabsorptive erol, and amino acids with its subsequent these data represent total glycogenolysis individuals and that the kidney became an hydrolysis by glucose-6-phosphatase to and indicate that ϳ55% of all glucose important source of glucose only in acidotic free glucose. Liver and skeletal muscle released into the circulation in the postab- conditions or after prolonged fasting (8). In contain most of the body’s glycogen stores. sorptive state is a result of gluconeogenesis. fact, the literature is replete with publica- However, because only the liver contains It should be pointed out that to a certain tions that refer to isotopic measurements of glucose-6-phosphatase, the breakdown of extent, this approach may lead to an over- the overall release of glucose into the cir- hepatic glycogen leads to the release of estimation of gluconeogenesis’ effects culation as hepatic glucose output. glucose, whereas the breakdown of muscle resulting from glycogen cycling and other However, the concept that the liver is glycogen leads to the release of lactate. considerations (3). the sole source of glucose, except in acidotic conditions and after prolonged fasting, has been challenged on several grounds. First, the classic studies of Felig et al. (9), Wahren From the Department of Medicine (J.E.G., C.M., H.J.W.), the University of Rochester, Rochester, New York; and the University of Tubingen (M.S.), Tubingen, Germany. et al. (10), and Ahlborg et al. (11) indicated Address correspondence to John E. Gerich, MD, University of Rochester School of Medicine, 601 Elm- that net splanchnic uptake of gluconeogenic wood Ave., Box MED/CRC, Rochester, NY 14642. E-mail: [email protected]. Address reprint precursors could maximally account for requests to Cadmus Journal Services Reprints, P.O. Box 751903, Charlotte, NC 28275-1903. only 20–25% of glucose release (not Received for publication 21 June 2000 and accepted in revised form 3 October 2000. Abbreviations: NMR, nuclear magnetic resonance. 36–55%), assuming that 100% of the net A table elsewhere in this issue shows conventional and Système International (SI) units and conversion uptake of these precursors were incorpo- factors for many substances. rated into glucose by the liver. Indeed, these 382 DIABETES CARE, VOLUME 24, NUMBER 2, FEBRUARY 2001 Gerich and Associates methods. These investigators injected 14C- labeled glucose into groups of rats that had been either hepatectomized or hepatec- tomized and nephrectomized. In the former group, there was dilution of the plasma glu- cose 14C specific activity as the animals became hypoglycemic, indicating the release of unlabeled (i.e., endogenously produced) glucose into the circulation from some source other than the liver. Dilution of the plasma specific activity of the injected glucose did not occur in hepatectomized animals that had been nephrectomized, providing evidence that the source of the endogenous glucose released into the circu- lation after hepatectomy was the kidney. Four years later, Teng (18) reported that renal cortical slices taken from animals with experimentally induced diabetes released glucose at an increased rate, but that treat- Figure 1—Endogenous glucose release (EGP) before and after removal of the liver in individuals under- ment of the animals with insulin could going liver transplantation. Reproduced from Joseph et al. (13) with permission. reverse this effect. In 1960, using a similar model, Landau (19) demonstrated that glu- coneogenesis from pyruvate was increased values might be overestimations, because Shortly thereafter, Reinecke (16) repro- more than twofold by the diabetic kidney. portal venous lactate, glycerol, and amino duced such results in rats, but also mea- Near that time, Krebs began a series of acid levels are generally equal to or lower sured arteriorenal venous glucose experiments characterizing the substrates than arterial levels (12). Second, in individ- concentrations in the hepatectomized ani- used for renal gluconeogenesis (20), the uals undergoing liver transplantation, mals. It was found that renal vein glucose capacity of the kidney for gluconeogenesis endogenous glucose release does not drop levels exceeded arterial levels as the animals in different species (21), and various to zero after removal of the liver (13,14); became hypoglycemic, thus demonstrat- aspects of the regulation of renal gluco- indeed, Joseph et al. (13) (Fig. 1) reported ing that under these conditions, the kid- neogenesis (22,23), including its stimula- that 1 h after removal of the liver, endoge- neys released glucose into the circulation. tion by free fatty acids (24). Because the nous glucose release decreases by only Several years later, Drury et al. (17) cor- kidney had a greater concentration of glu- ϳ50%. Finally, recent studies using a com- roborated this conclusion using isotopic coneogenic enzymes (in terms of weight) bination of net renal glucose balance and isotopic measurements have demonstrated that the kidney releases significant amounts of glucose in postabsorptive normal volun- teers (7). This article resummarizes and updates current information on human renal glucose metabolism as recently reviewed by Meyer and Gerich (7). EARLY NONHUMAN STUDIES — In 1938, Bergman and Drury (15) presented the first evidence that the kidney might release glucose and be important for maintenance of normal glucose homeostasis. These investigators used the glucose clamp technique to maintain eugly- cemia in two groups of rabbits—one func- tionally hepatectomized and one functionally hepatectomized and nephrectomized. As shown in Fig. 2, functional removal of the kidneys in hepatectomized animals led to an abrupt increase in the amount of glucose