Journal of Applied Physiology Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 98: 930-939, 2005; doi:10.1152/japplphysiol.00687.2004
8750-7587/05 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Christopher, M. J.
Right arrow Articles by Alford, F. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Christopher, M. J.
Right arrow Articles by Alford, F. P.

Prevailing hyperglycemia is critical in the regulation of glucose metabolism during exercise in poorly controlled alloxan-diabetic dogs

Michael J. Christopher,1,2 Christian Rantzau,1,2 Glenn McConell,3 Bruce E. Kemp,4 and Frank P. Alford1,2

1Departments of Endocrinology and Diabetes, St. Vincent's Hospital Melbourne, Fitzroy; Departments of 2Medicine and 3Physiology, University of Melbourne, Parkville; and 4St. Vincent's Institute of Medical Research, Fitzroy, Victoria, Australia

Submitted 2 July 2004 ; accepted in final form 10 October 2004


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 
The separate impacts of the chronic diabetic state and the prevailing hyperglycemia on plasma substrates and hormones, in vivo glucose turnover, and ex vivo skeletal muscle (SkM) during exercise were examined in the same six dogs before alloxan-induced diabetes (prealloxan) and after 4–5 wk of poorly controlled hyperglycemic diabetes (HGD) in the absence and presence of ~300-min phlorizin-induced (glycosuria mediated) normoglycemia (NGD). For each treatment state, the ~15-h-fasted dog underwent a primed continuous 150-min infusion of [3-3H]glucose, followed by a 30-min treadmill exercise test (~65% maximal oxygen capacity), with SkM biopsies taken from the thigh (vastus lateralis) before and after exercise. In the HGD and NGD states, preexercise hepatic glucose production rose by 130 and 160%, and the metabolic clearance rate of glucose (MCRg) fell by 70 and 37%, respectively, compared with the corresponding prealloxan state, but the rates of glucose uptake into peripheral tissues (Rdtissue) and total glycolysis (GF) were unchanged, despite an increased availability of plasma free fatty acid in the NGD state. Exercise-induced increments in hepatic glucose production, Rdtissue, and plasma-derived GF were severely blunted by ~30–50% in the NGD state, but increments in MCRg remained markedly reduced by ~70–75% in both diabetic states. SkM intracellular glucose concentrations were significantly elevated only in the HGD state. Although Rdtissue during exercise in the diabetic states correlated positively with preexercise plasma glucose and insulin and GF and negatively with preexercise plasma free fatty acid, stepwise regression analysis revealed that an individual's preexercise glucose and GF accounted for 88% of Rdtissue during exercise. In conclusion, the prevailing hyperglycemia in poorly controlled diabetes is critical in maintaining a sufficient supply of plasma glucose for SkM glucose uptake during exercise. During phlorizin-induced NGD, increments in both Rdtissue and GF are impaired due to a diminished fuel supply from plasma glucose and a sustained reduction in increments of MCRg.

glucose uptake; plasma-derived glycolysis; skeletal muscle; free fatty acids; metabolic clearance rate of glucose; phlorizin-induced normoglycemia

THE REGULATION OF GLUCOSE METABOLISM in skeletal muscle (SkM) during exercise occurs predominantly via an insulin-independent pathway (13, 16, 22) and involves the recruitment of SkM GLUT4 transporters from a distinct insulin-independent GLUT-4 pool (13). It is known that the prior insulin sensitivity of an individual is important in determining the rate of glucose uptake into SkM (Rdtissue) in response to acute exercise (28). Furthermore, the intensity and duration of the exercise are also critical in determining the relative importance of carbohydrate and fat substrates as prime sources of oxidative energy and whether the fuel substrates are derived from plasma or from within the exercising muscle itself (36, 49). During acute, moderately intense exercise at ~55–65% of maximal oxygen capacity (O2 max) in humans, the total energy expenditure of normal exercising SkM increases ~12-fold, with the rates of carbohydrate and fat oxidation increasing by ~14- and ~10-fold, respectively, compared with the resting state (36, 49). At this exercise intensity, the major fuel substrates carbohydrate (derived from SkM glycogen and plasma glucose) and fat [derived from plasma free fatty acids (FFA) and other fat sources] contribute equally to the total energy expenditure (36, 49). To meet this rapidly increasing demand for oxidative energy (ATP production), there are marked enhancements in the translocation of SkM GLUT4 transporters to the cell surface membrane (16, 22), the transformation of SkM pyruvate dehydrogenase (PDH) complex into its active form (33), and the activation of SkM carnitine palmitoyltransferase-1 (34, 39).

Poorly controlled Type 1 (26, 54), Type 2 (27, 47), and experimentally induced (3, 23) diabetes are characterized by chronic hyperglycemia, absolute and/or relative hypoinsulinemia, raised hepatic glucose production (HGP), elevated circulating FFA and glucagon levels, and severe insulin resistance, particularly in SkM (48, 54). Moreover, in response to acute mild to moderate exercise (~40–60% O2 max), suboptimally controlled hypoinsulinemic diabetic subjects exhibit metabolic abnormalities, including reduced exercise-stimulated Rdtissue [Rdtissue(ex)] and oxidation rates from both plasma glucose and SkM glycogen (29, 34, 57), and increased exercise-stimulated oxidation rates from both plasma FFA and SkM fat (34, 51). In such diabetic individuals, the shift from carbohydrate to fat oxidation during exercise in diabetic subjects has been attributed to a combination of factors, including downregulation of GLUT4 transporters (12, 14), inadequate plasma insulin concentrations (41, 52, 53), decreased PDH activity (18), inadequate glycogenolysis due to a reduced SkM glycogen storage pool (1, 34), enhanced exercise-induced lipolysis (17, 53), and/or a chronically altered oxidative status in SkM due to the diabetic state itself (10).

Hyperglycemia per se has a clearly defined impact on whole body glucose turnover, the partitioning of intracellular (IC) glucose into the total glycolytic (GF) and glucose storage pathways, and SkM glucose metabolism (6, 47, 48, 55). It is also well established that the actions of hyperglycemia in the resting, poorly controlled diabetic state involve both compensatory (48, 55) and glucose toxic (38, 39) effects. However, little is known about the metabolic role of hyperglycemia per se on the regulation of exercise-stimulated whole body and SkM glucose metabolism in poorly controlled diabetes. Previous exercise studies in hypoinsulinemic diabetes, either produced experimentally (14, 41, 52, 53) or in Type 1 diabetic subjects (29, 34, 51, 57), have provided conflicting data. Some studies observed normal Rdtissue(ex) with moderate exercise in diabetic models employing either fixed low (51), normal (14), or several-fold increased supplemental insulin (57). Other investigators found reduced Rdtissue(ex) during moderate exercise in totally insulin-deficient (52, 53), hypoinsulinemic (41), normoinsulinemic (34), or hyperinsulinemic (29) diabetes. Unfortunately, interpretation of the role of hyperglycemia per se on the regulation of exercise-stimulated in vivo whole body and IC glucose metabolism in these diabetic exercise studies is made difficult by the levels of glycemia, which were not well controlled, varying from normoglycemic to markedly hyperglycemic levels and the varying insulinemic states.

In fact, only two groups have specifically examined the impact of hyperglycemia per se (at "basal" insulinemia) on the exercise-induced increment (d) in Rdtissue (dRdtissue) and glucose metabolism. In the first study, nondiabetic dogs infused with somatostatin and constant basal insulin (~14 mU/l) were exercised (~40% O2 max) at various glycemic steady states (56). They showed that dRdtissue, both in the whole body and across the leg, rose in direct proportion to the increasing circulating glucose concentration, thereby resulting in an unchanged exercise-induced increment in the metabolic clearance rate of glucose (dMCRg). These authors concluded that hyperglycemia profoundly enhances exercise-induced glucose uptake and glucose metabolism, which may be critical in diabetes (56). In contrast, Fisher et al. (14) performed exercise studies (also at ~40% O2 max) in alloxan-induced chronic diabetic dogs at their prevailing fasting hypoinsulinemia (~4 mU/l) and hyperglycemia (~22 mM). These workers found that dRdtissue in the untreated alloxan-induced diabetic dogs was normal, but dMCRg was reduced approximately fourfold compared with the nondiabetic dogs. However, when relative normoglycemia (NGD; ~7 mM) was restored in the alloxan-induced hyperglycemic diabetic dogs by an acute infusion of the insulin-independent glycosuric agent, phlorizin, dRdtissue remained normal, and dMCRg was normalized (14). Therefore, Fisher et al. (14) concluded that the markedly reduced MCRg in suboptimally controlled hyperglycemic diabetes (HGD) protects resting and working SkM against the potentially deleterious effects of excessive glucose uptake. Thus two opposing views exist as to the role that hyperglycemia plays in the glucose metabolic response of SkM to exercise in insulin-resistant states at basal insulinemia and that hyperglycemia directly enhances glucose uptake and metabolism during exercise (56) or that hyperglycemia is potentially harmful to working SkM but is protected by the coexisting reduced MCRg (14). Unfortunately, neither of these glucose turnover studies (14, 56) examined ex vivo SkM glucose metabolism at rest and during exercise.

We recently reported that the preexercise activity of SkM AMP-activated protein kinase (AMPK), a key regulator and monitor of SkM IC energy balance, was chronically elevated in the low-dose alloxan-induced diabetic dogs after 4- to 5-wk poorly controlled (low-dose insulin treated) HGD. When the same diabetic dogs underwent acute phlorizin infusion to restore NGD, preexercise AMPK {alpha}1- and {alpha}2-isoform activities remained elevated compared with the prealloxan state (5). However, AMPK isoform activities did not increase further with moderate treadmill exercise (~65% O2 max) in either diabetic state. We also noted a ~50% reduction in dRdtissue in the NGD state (5).

Therefore, the aims of the present study were, first, to examine the separate impacts of the alloxan-induced, poorly controlled chronic diabetic state itself and the levels of glycemia (hyperglycemia vs. phlorizin-induced NGD) on plasma substrates (glucose, FFA, and lactate) and hormones (insulin and glucagon), in vivo HGP, Rdtissue, MCRg, and GF, and ex vivo SkM glucose metabolism [total and IC glucose, glucose 6-phosphate (G-6-P), lactate, and glycogen concentrations]. Second, we wished to determine the impacts of these two diabetic states on the exercise-induced whole body and SkM glucose metabolic responses to 30 min of moderate treadmill exercise (~65% O2 max) and whether hyperglycemia plays a primary role in the supply of fuel during exercise. In particular, we wished to identify which preexercise parameters were important in the regulation of Rdtissue(ex) in the diabetic states compared with the prealloxan state and whether there was a reduction in the utilization of carbohydrates as a major fuel source during exercise in the HGD and/or NGD states.

We hypothesized that, in the chronic HGD state, normal basal and exercise-stimulated Rdtissue and plasma-derived GF are maintained by compensatory alterations in glucose turnover, in particular HGP and MCRg, and SkM IC glucose metabolism. However, when the mass-action effect of the hyperglycemia in diabetes is removed by acute phlorizin infusion, independently of insulin, both exercise-induced Rdtissue and plasma-derived GF are impaired due to an insufficient fuel supply from plasma glucose to meet the increased energy demand and a sustained defect in exercise-induced MCRg.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 
Animals.   Studies were carried out on six male dogs of mixed breed (18–26 kg body wt), with the permission of the Experimental Medical and Surgical Research Ethics Committee at St. Vincent's Hospital, Melbourne, Australia. Surgical preparation of an arteriovenous shunt for arterialized blood sampling and catheterization of a jugular vein, connected by tubing to a subcutaneously fastened stainless-steel port, for infusion of solutions was performed at least 14 days before the first study, as previously described (6, 7). Exercise, training, and health monitoring of dogs was as previously described (6, 7). In brief, the daily diet given to the prealloxan dogs consisted of ~2,800 calories/day (28% energy from protein, 30% from fat, 42% from carbohydrate). Dogs were acclimatized to the treadmill stepwise exercise program for at least 2 wk before the first study and underwent a standard 3-h intravenous (iv) glucose tolerance test (IVGTT), commencing with an iv bolus of 50% glucose (0.3 g/kg) and 50 µCi highly purified tritiated water (3H2O) (NEN Life Science Products, Boston, MA) given over 30 s, with frequent blood samples taken over the next 3 h for measurement of plasma glucose and total insulin, and 3H2O-specific activity. These data were used to calculate glucose tolerance, the acute incremental insulin response to the glucose bolus from 0 to 10 min and from 0 to 40 min (24), and total body water content (11). After this study, a recovery period of at least 9 days was allowed before the prealloxan exercise study was performed.

Dogs were rendered diabetic (after a 24-h fast) by iv injection of low-dose alloxan monohydrate (35 mg/kg) (Sigma, St. Louis, MO), as previously described (5). Within 2–3 days, the diabetic dogs began receiving twice daily subcutaneous injections of low-dose, long-acting (duration: 15–20 h) insulin (human Monotard, Novo Nordisk, Sydney, Australia) with their food for 4–5 wk, aimed at producing chronic hyperglycemia [premeal blood glucose concentrations were 14.6 ± 0.4 mM (mean ± SE) and 18.3 ± 0.3 mM in the morning and afternoon, respectively]. The diabetic dogs were given a weight-maintaining [diabetic: 21.3 ± 1.2 vs. prealloxan: 22.1 ± 1.2 kg; P = not significant (NS)] diet, which consisted of an ~40% increase in dog chow per day to compensate for the urinary loss of calories due to glycosuria (5). The diabetic dogs required 10 ± 1 units of Monotard per day. The HGD dogs underwent a 3-h IVGTT at their prevailing fasting hyperglycemia, with a constant "basal" iv infusion rate (9.3 ± 0.9 mU·kg–1·h–1) of insulin (human Actrapid, Novo Nordisk; in isotonic saline containing 10% Haemaccel for plasma substitution) commenced 35 min before and continued throughout the IVGTT. The infusion rate of insulin given to each HGD dog was chosen to match the plasma insulin level seen in the same fasting prealloxan dog.

Experiments.   Dogs were fasted for at least 15 h before each study and had free access to water at all times. Before fasting, the diabetic dogs received a 20–30% lower dose of insulin (that is, the dogs received an average of 3–4 U Monotard) with their usual food ration to reduce the risk of insulin-induced nocturnal fasting hypoglycemia. On the morning of each study, an 18-gauge blood sampling catheter was inserted in the dog's foreleg arteriovenous fistula, and a 19-gauge infusion set was inserted in the subcutaneous venous access port connected to the jugular vein (5–7). After a rest period of at least 30 min, fasting blood samples were taken for measurement of plasma glucose, total insulin, FFA, lactate, glucagon, tritiated glucose ([3-3H]glucose), and 3H2O.

The prealloxan exercise study involved a 150-min fasting preexercise equilibration period, employing a primed (20 µCi)-continuous infusion (10 µCi/h) of highly purified [3-3H]glucose (NEM Life Science Products), as previously described (5–7), with the 30-min exercise test performed from 150 to 180 min. The paired diabetic exercise studies were randomized. In one exercise study, the HGD dog underwent a 150-min fasting preexercise equilibration period, commencing with a primed (30–50 µCi) iv bolus of [3-3H]glucose [adjusted in proportion to the prevailing fasting hyperglycemia to adequately label the glucose pool (6, 20)] followed by a continuous infusion (10 µCi/h) of [3-3H]glucose, with the same exercise test performed from 150 to 180 min. In the other exercise study, the same fasting diabetic dog underwent an ~330-min constant iv infusion (50 µg·kg–1·min–1) of phlorizin (phloretin-2,1-{beta}-D-glucoside; Sigma) to induce a normoglycemic state (NGD). Phlorizin inhibits sodium-dependent glucose cotransport in the renal tubules, leading to inhibition of renal glucose reabsorption and profound glycosuric-induced NGD independent of insulin (14, 39). After phlorizin had been infused for ~150 min, a 150-min primed (25 µCi)-continuous infusion (20 µCi/h) of [3-3H]glucose was commenced, with the same exercise test performed from 300 to 330 min.

The 30-min treadmill exercise test protocol has been described previously (5) and represents an exercise load of ~65% O2 max for dogs (9). During each exercise test, additional saline (3–4 ml/min) was infused to ensure adequate hydration of the dogs (14), and the [3-3H]glucose infusion rate was increased stepwise by 3.0-, 2.0-, and 1.5-fold during exercise in the prealloxan, HGD, and NGD studies, respectively, to minimize the change in specific activity obtained due to exercise-induced increments in SkM Rdtissue (14, 36, 49).

Collection of blood, urine, and SkM biopsy samples.   For the exercise studies, regular blood samples were collected 10–15 min apart during the preexercise periods and every 5 min during the exercise test, placed into tubes containing appropriate anticoagulants and preservatives, and centrifuged at 4°C within 2 h, and the separated plasma was stored frozen at –20°C until assayed for plasma glucose, total insulin, FFA, lactate, glucagon, [3-3H]glucose, and 3H2O, as previously described (7). In the diabetic studies, the bladder was emptied (using a urinary catheter coated with 1% lignocaine gel) immediately before administering the [3-3H]glucose bolus, just before the commencement of exercise, and at the completion of exercise. The total volume of urine produced throughout each period was measured, and the urine analyzed for both glucose concentration and specific activity of [3-3H]glucose to precisely quantify urinary glucose and [3-3H]glucose loss (UrGloss) throughout the preexercise and exercise periods (14). Preparation, collection, and storage of SkM biopsy samples from the thigh (vastus lateralis) at the completion of the basal (preexercise) and exercise periods (on separate days) were performed as previously described (5). Contracting SkM biopsy samples were taken under anesthesia within 2–4 min of the completion of exercise (5). SkM biopsy samples were analyzed for total glucose and IC glucose, G-6-P, lactate, and glycogen concentrations (8, 19, 47).

Laboratory analyses and calculations.   Plasma levels of glucose, total insulin, FFA, lactate, and glucagon, and specific activities of [3-3H]glucose in plasma and urine samples were measured as previously described (6, 7). The rates of total glucose appearance (Ratotal) and total glucose disposal (Rdtotal) during the 150-min preexercise periods were determined from plasma [3-3H]glucose-specific activities at steady state over the last 30 min, as previously defined (6, 20), or if not at steady state by employing Steele's non-steady-state equations (45), assuming a glucose pool fraction of 0.65 and a glucose volume of distribution of 200 ml/kg (7). During the exercise period, Ratotal and Rdtotal were calculated, employing Steele's non-steady-state equations and the same assumptions detailed above (45). Any errors in the calculation of Ratotal and Rdtotal using these assumptions are minimized by the regularity of sampling (5 min) and the lack of exercise-induced changes in plasma glucose levels and [3-3H]glucose-specific activities (due to the increasing [3-3H]glucose infusion rates). The rate of UrGloss was subtracted from Rdtotal in the diabetic studies to reflect the actual Rdtissue (5–7). The measurement of the rate of in vivo GF during the preexercise period from extracellularly derived glucose due to the generation of plasma 3H2O from [3-3H]glucose infusion was estimated from the formula described previously (6, 37). The accumulation of plasma 3H2O was linear [r2 (adjusted) = 94.3 ± 0.7%] from 30 to 150 min in all preexercise periods. During the exercise period, a combination of the increased muscle capillary blood flow (31) and non-steady-state conditions does not make it possible to precisely measure GF using this method. However, given that published data have shown that virtually all plasma glucose that enters exercising muscle during moderate exercise is oxidized (36, 49), we can assume that the exercise-induced increment in plasma-derived GF (dGF) is the difference between Rdtissue(ex) and preexercise GF. Preexercise MCRg was calculated by averaging the four values of Rdtissue determined over the last 30 min of the preexercise period and dividing this number by the average plasma glucose value obtained at the corresponding time points. For the exercise period, MCRg was calculated by averaging the two values of Rdtissue determined over the last 5 min of exercise, and dividing this number by the average plasma glucose value obtained at the corresponding time points.

The concentrations of SkM G-6-P and IC glucose were determined in an ~15- to 20-mg wet weight muscle sample using a fluorometric-coupled enzymatic assay, as previously outlined from our laboratory (8, 19), as modified from the method of Schalin-Jantti et al. (40). For the remaining assays, ~20–30 mg wet weight of muscle was freeze dried, reweighed, crushed into a powder, and separated from any connective tissue. Approximately 1.5 mg of this freeze-dried muscle was hydrolyzed in 375 µl of 2 M HCl at 95–100°C (with frequent agitation) for 2 h, neutralized with 1,125 µl of 0.667 M NaOH, and stored at –70°C until analysis (25). SkM glycogen content was determined from 10 µl of neutralized extract measured in triplicate using a fluorometric enzymatic assay, as previously described by our laboratory (5, 8, 19). Of the remaining free-dried muscle, ~2.0 mg was homogenized in 250 µl precooled 0.5 M perchloric acid/1 mM EDTA, frequently vortexed for 10 min, and centrifuged at 14,000 rpm (2°C) for 2 min. Then, 50 µl of precooled 2.1 M KHCO3 was added to exactly 200 µl of the supernatant, vortexed, allowed to stand for 5 min, vortexed again, and centrifuged at 14,000 rpm (2°C) for 2 min. The remaining ~180-µl supernatant was then transferred to a fresh cryule and stored at –70°C until analysis (25). SkM lactate concentration was determined from 20 µl of neutralized extract measured in triplicate using a fluorometric enzymatic assay (1.0 M hydrazine/1.0 M glycine buffer system) (25). All SkM substrates were measured at room temperature on a 650-10S Fluorometer (Hitachi, Tokyo, Japan) (excitation wavelength = 340 nm; emission wavelength = 455 nm) in a total volume of 1.0 ml (10-mm light path). For every assay session, the absorbancies of a freshly thawed set of NADH standards were measured on a spectrophotometer (Hitachi model U-2000, Tokyo, Japan) at 340 nm to determine the exact concentration of NADH. Concentrations of SkM total glucose and G-6-P are expressed as millimoles per kilogram dry weight. Glycogen content is expressed as millimole glucose residues per kilogram dry weight. Concentrations of SkM IC glucose and lactate are expressed as millimole per liter IC water, assuming an extracellular water content in the biopsies of 0.3 l/kg dry wt and an IC water content in the biopsies of 2.8 l/kg dry wt, as discussed previously (43, 44, 48).

Statistical analysis.   Data are presented as means ± SE. One-way analysis of variance was used for repeated measures, with differences within and/or between groups determined by using Wilcoxon's matched-pairs signed-rank test. Correlation analyses were performed by using the Spearman rank correlation coefficient (r values). Stepwise (default), best-subsets, and multiple-regression analyses were performed using the statistical analysis program Minitab (Minitab, State College, PA).


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 
IVGTT studies.   The HGD dogs were studied at their prevailing fasting hyperglycemia (pre-IVGTT: 18.5 ± 2.4 vs. prealloxan: 5.0 ± 0.1 mM; P < 0.05) and constant basal insulinemia (pre-IVGTT: 9.4 ± 1.2 vs. prealloxan: 9.6 ± 1.5 mU/l; P = NS). Glucose tolerance (not corrected for UrGloss) was severely impaired (0.6 ± 0.1 vs. prealloxan: 4.7 ± 0.4 min–1 x 10–2; P < 0.05), there was no acute endogenous insulin secretion (acute incremental insulin response to the glucose bolus from 0 to 10 min: –0.2 ± 0.2 vs. prealloxan: 37.0 ± 7.0 mU/l; P < 0.05) or delayed endogenous insulin secretion (acute incremental insulin response to the glucose bolus from 0 to 40 min: 0.0 ± 0.2 vs. prealloxan: 17.8 ± 4.3 mU/l; P < 0.05), and there was no change in total body water content (59.0 ± 1.9 vs. prealloxan: 59.3 ± 2.0%; P = NS). Therefore, these alloxan-induced HGD dogs represent a model of chronic (30 ± 2 days) hyperglycemic and hypoinsulinemic diabetes under poor control (low-dose insulin therapy) (3, 5).

Exercise studies: plasma substrate and hormone levels.   As previously reported (5), the HGD dogs exhibited marked fasting (preexercise) hyperglycemia (17.1 ± 1.0 vs. prealloxan: 5.1 ± 0.1 mM; P < 0.05) and modest hypoinsulinemia (6.6 ± 2.1 vs. prealloxan: 9.2 ± 1.4 mU/l; P = NS). However, in the 300-min phlorizin-infused diabetic dogs (NGD), both preexercise plasma glucose (7.7 ± 0.4 mM) and total insulin (3.4 ± 0.8 mU/l) fell significantly (P < 0.05 for both vs. corresponding prealloxan and HGD states). Importantly, this fall in total insulin in the NGD state (~3 mU/l) was similar to that observed when the same HGD dogs were fasted for 5 h (data not shown). During exercise, there was no significant alteration in plasma glucose (prealloxan: 5.0 ± 0.1 vs. HGD: 16.5 ± 1.1 vs. NGD: 8.1 ± 0.3 mM) or total insulin (vs. prealloxan: 10.9 ± 2.6 vs. HGD: 6.9 ± 2.3 vs. NGD: 4.2 ± 0.8 mU/l) in any treatment group. The mean percent coefficients of variation obtained for plasma glucose, total insulin, and [3-3H]glucose-specific activities during the last 30 min of the preexercise period, respectively, in the three treatment groups were prealloxan: 2.8 ± 0.7, 14.7 ± 1.7, and 3.5 ± 0.3%; HGD: 1.8 ± 0.4, 13.5 ± 3.3, and 1.7 ± 0.3%; and NGD: 3.6 ± 1.2, 16.2 ± 2.9, and 4.4 ± 1.2%. In addition, the mean specific activities of plasma [3-3H]glucose obtained during the last 30 min of the preexercise period and at the completion of exercise, respectively, in the three treatment groups were kept constant within each group (prealloxan: 1,469 ± 69 and 1,754 ± 106 dpm/µmol; HGD: 761 ± 82 and 842 ± 91 dpm/µmol; and NGD: 1,184 ± 144 and 1,336 ± 145 dpm/µmol). In 16 of the 18 exercise studies, the plasma [3-3H]glucose-specific activity at the completion of exercise was within the prescribed 30% of the value obtained immediately before exercise (20).

As previously reported (5), preexercise plasma FFA levels were significantly elevated in the NGD state compared with both the prealloxan and HGD states (Fig. 1A). The absolute rise in FFA induced by exercise was similar in the three treatment states, and therefore the FFA levels obtained with exercise in the NGD state were significantly higher than the corresponding prealloxan and HGD values. Preexercise plasma lactate was similar in the three treatment states, but the increment in lactate induced by exercise in the NGD dogs was significantly higher than that observed in the other two states (Fig. 1B). In both the HGD and NGD states, preexercise glucagon levels were significantly elevated compared with the prealloxan state and were ~50% higher (but not significantly) in the NGD state compared with the HGD state (Fig. 1C). With exercise, there was a significant rise in glucagon in both diabetic states but not in the prealloxan state. Similar to that seen with plasma lactate, the increment in glucagon induced by exercise (dglucagon) in the NGD dogs was significantly higher than that observed in the other two states.



View larger version (28K):
[in this window]
[in a new window]
  		Fig. 1. Plasma concentrations of free fatty acids (FFA; A), lactate (B), and glucagon (C) before (open bars) and after 30 min of treadmill exercise at ~65% maximal oxygen consumption (O2 max; hatched bars) in the same 6 dogs as in the prealloxan state and after 4–5 wk of suboptimally controlled diabetes in the absence (hyperglycemia) and presence of acute phlorizin-induced normoglycemia. Values are means ± SE. *Significant difference vs. corresponding preexercise value (P < 0.05). {dagger}Significant difference vs. corresponding prealloxan value (P < 0.05). {ddagger}Significant difference vs. corresponding hyperglycemic diabetic value by Wilcoxon matched-pairs signed-rank test (P < 0.05).

 
In vivo HGP, Rdtissue, GF, UrGloss, and MCRg in response to diabetes, NGD, and exercise.   As shown in Fig. 2A, preexercise HGP was significantly elevated by ~130 and ~160% in the HGD and NGD states, respectively, compared with the corresponding prealloxan state. It is important to note that, in three of the six NGD studies, a variable infusion rate of exogenous 10% glucose prelabeled with 2.0 µCi/g [3-3H]glucose (GINF) was required to maintain NGD (blood glucose ~5.5 mM) during the preexercise period. In addition, a constant infusion rate of GINF, equal to the rate of GINF required to maintain NGD immediately before exercise, was employed during the exercise period. The mean rate of GINF required during the last 30 min of the preexercise period and during the exercise period for the six NGD studies was 7.9 ± 4.6 and 8.4 ± 4.8 µmol·kg–1·min–1, respectively. Therefore, the mean preexercise and exercise-induced values of HGP obtained for the six NGD studies (calculated by subtracting the GINF rate from the corresponding value of Rdtotal) were 30.9 ± 4.8 and 40.3 ± 6.5 µmol·kg–1·min–1, respectively, resulting in an exercise-induced increment in HGP (dHGP) of 9.4 ± 1.7 µmol·kg–1·min–1. This value was significantly lower than that seen in the other two states (prealloxan: 20.4 ± 1.9 and HGD: 14.9 ± 1.8 µmol·kg–1·min–1; P < 0.05 for both) (Fig. 2A). However, in the three NGD dogs that required GINF, the preexercise HGP and dHGP values averaged 22.1 ± 2.6 and 6.6 ± 1.1 µmol·kg–1·min–1, respectively. In contrast, in the three NGD dogs that did not require GINF, the preexercise HGP and dHGP values averaged 39.6 ± 5.7 and 12.2 ± 2.2µmol·kg–1·min–1, respectively. Therefore, the presence of GINF in the NGD dogs resulted in a 45% reduction in preexercise HGP and a 50% reduction in dHGP. However, the GINF infusion did not influence either preexercise or exercise-induced peripheral glucose metabolism (namely Rdtissue, MCRg, and GF) in the NGD state (data not shown). Preexercise HGP was positively correlated with preexercise glucagon levels for the combined groups (r = 0.83, P = 0.001; n = 18 studies) and for the two diabetic states (r = 0.71, P = 0.02; n = 12 studies) but showed no relationship with preexercise FFA.



View larger version (32K):
[in this window]
[in a new window]
  		Fig. 2. Rates of hepatic glucose production (HGP; A), glucose uptake into peripheral tissues (Rdtissue; B), and urinary glucose loss (UrGloss; C) before (open bars) and after 30 min of treadmill exercise at ~65% O2 max (hatched bars) in the same 6 dogs as in the prealloxan state and after 4–5 wk of suboptimally controlled diabetes in the absence (hyperglycemia) and presence of acute phlorizin-induced normoglycemia. Values are means ± SE. *Significant difference vs. corresponding preexercise value (P < 0.05). {dagger}Significant difference vs. corresponding prealloxan value (P < 0.05). {ddagger}Significant difference vs. corresponding hyperglycemic diabetic value by Wilcoxon matched-pairs signed-rank test (P < 0.05).

 
As previously reported (5), preexercise Rdtissue was remarkably similar in the three treatment groups, regardless of the glycemic status and degree of glycosuria observed (Fig. 2, B and C), but both Rdtissue(ex) and dRdtissue were significantly blunted by ~25–30 and ~40–50%, respectively, in the NGD state compared with the prealloxan and HGD states (Fig. 2B). Preexercise UrGloss accounted for 58 and 74% of Rdtotal in the HGD and NGD states, respectively (Fig. 2C). With exercise, the absolute rate of UrGloss did not change in either diabetic state. Preexercise MCRg was markedly reduced by 70% in the HGD state compared with the prealloxan state (Fig. 3). Although phlorizin-induced NGD in the diabetic dogs improved preexercise MCRg, it was still significantly lower (by 37%) compared with the prealloxan state but may reflect the raised plasma glucose level obtained in the NGD state compared with the prealloxan state. However, dMCRg was markedly reduced by ~70–75% in both diabetic states compared with the prealloxan state (Fig. 3, inset). Preexercise GF was similar in all three treatment groups (prealloxan: 9.5 ± 1.1 vs. HGD: 10.3 ± 2.1 vs. NGD: 7.7 ± 2.0 µmol·kg–1·min–1), although it was reduced by ~25% in the NGD group compared with the HGD state (P = non significant). However, dGF was significantly lower by ~30–40% in the NGD state compared with the other treatment group states (Fig. 4).



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 3. Metabolic clearance rate of glucose (MCRg) before (open bars) and after 30 min of treadmill exercise at ~65% O2 max (hatched bars) and the exercise-induced increment (d) in MCRg (dMCRg) (inset) in the same 6 dogs as in the prealloxan state and after 4–5 wk of suboptimally controlled diabetes in the absence (hyperglycemia) and presence of acute phlorizin-induced normoglycemia. Values are means ± SE. *Significant difference vs. corresponding preexercise value (P < 0.05). {dagger}Significant difference vs. corresponding prealloxan value (P < 0.05). {ddagger}Significant difference vs. corresponding hyperglycemic diabetic value by Wilcoxon matched-pairs signed-rank test (P < 0.05).

 


View larger version (10K):
[in this window]
[in a new window]
  		Fig. 4. Exercise-induced increment in the rate of total glycolysis (dGF) induced by 30 min of treadmill exercise at ~65% O2 max in the same 6 dogs as in the prealloxan state and after 4–5 wk of suboptimally controlled diabetes in the absence (hyperglycemia) and presence of acute phlorizin-induced normoglycemia. Values are means ± SE. *Significant difference vs. corresponding prealloxan value (P < 0.05). {dagger}Significant difference vs. corresponding hyperglycemic diabetic value by Wilcoxon matched-pairs signed-rank test (P < 0.05).

 
Ex vivo SkM glucose processing in response to diabetes, NGD, and exercise.   As shown in Table 1, preexercise SkM total glucose and IC glucose concentrations were significantly raised in the HGD state compared with the prealloxan state but were normalized in the NGD state. There was no significant change in either SkM total or IC glucose levels with exercise in any group, but the absolute values obtained during exercise in the HGD and NGD states were no longer significantly different. Preexercise SkM G-6-P concentrations were similar in all groups and rose modestly (but not significantly) with exercise. Preexercise SkM lactate concentrations were similar in all three groups and rose modestly (by ~30–40%) but not significantly with exercise in the prealloxan and NGD groups. However, the ~75% rise in SkM lactate with exercise in the HGD state was significant. Preexercise SkM glycogen content was similar in all groups and did not decrease significantly with the exercise protocol employed.


View this table:
[in this window]
[in a new window]
  		Table 1. Skeletal muscle concentrations of substrates before and after 30 min of treadmill exercise at ~65% O2 max in the same 6 dogs as in the prealloxan state and after 4–5 wk of suboptimally controlled diabetes in the absence and presence of acute phlorizin-induced normoglycemia

 
Determinants of the metabolic responses to exercise and utilization of fuel sources.   First, we performed nonparametric correlations to determine which preexercise metabolic parameters were associated with the altered Rdtissue(ex) and dGF obtained in the HGD and NGD states compared with the prealloxan state and which fuel sources (carbohydrate vs. fat) were utilized during exercise. We found that preexercise plasma glucose showed significant positive correlations with Rdtissue(ex) and dGF for the two diabetic states (n = 12 studies) but not for the combined groups (n = 18 studies) (Table 2). Preexercise plasma total insulin correlated positively with Rdtissue(ex) for both the combined and diabetic states and negatively with preexercise plasma FFA for the combined (r = –0.66, P = 0.003) and diabetic (r = –0.64, P = 0.024) groups. Preexercise plasma FFA showed a negative correlation with Rdtissue(ex) for both the combined and diabetic groups (Table 2). In addition, FFA with exercise negatively correlated with Rdtissue(ex) for the combined (r = –0.67, P = 0.002) and diabetic (r = –0.59, P = 0.045) groups. Preexercise GF showed a positive correlation with Rdtissue(ex) for both the combined and diabetic groups and a negative correlation with both preexercise FFA and FFA with exercise for the combined (r = –0.65, P = 0.004; and r = –0.51, P = 0.030) and diabetic (r = –0.68, P = 0.015; and r = –0.65, P = 0.022) groups.


View this table:
[in this window]
[in a new window]
  		Table 2. Correlation analyses of preexercise parameters vs. altered in vivo metabolic responses to 30 min of treadmill exercise at ~65% O2 max for the combined and diabetic states

 
We also found that both preexercise plasma glucose and SkM IC glucose showed significant positive correlations with dRdtissue for the two diabetic states (r = 0.81, P = 0.001; and r = 0.59, P = 0.045, respectively). Both dRdtissue and dGF negatively correlated with dglucagon (r = –0.61, P = 0.009; and r = –0.63, P = 0.006, respectively) for the combined groups only, and dGF showed a positive correlation with the increment in SkM lactate with exercise (dSkM lactate) for the two diabetic states only (r = 0.60, P = 0.039).

Second, we examined in greater detail the preexercise metabolic parameters that could account for the majority of the abnormal Rdtissue(ex) observed in the diabetic states. We found that Rdtissue(ex) significantly correlated with four preexercise metabolic parameters, namely plasma glucose, total insulin, FFA, and GF, using nonparametric analysis (Table 2). To more closely examine these four basal variables and their relationship with Rdtissue(ex), we confirmed that each variable was normally distributed and that each parameter fit the simple linear regression model with Rdtissue(ex) (19). These four basal variables were then combined against Rdtissue(ex) using multiple linear regression, yielding an r2 (adjusted) value of 86% (r = 0.93). We then performed stepwise (default) regression analysis to select the independent predictors that could account for the majority of Rdtissue(ex) in the two diabetic states and found that only two of the predictors, namely preexercise plasma glucose and GF, combined to fit the model, yielding an r2 (adjusted) value of 88% (r = 0.94). The regression equation is Rdtissue(ex) = 11.2 + 0.506(preexercise plasma glucose) + 0.840(preexercise GF).


   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 
In the present study, we examined the separate impacts of low-dose, alloxan-induced, chronic, poorly controlled diabetes itself and the glycemic state (hyperglycemia vs. phlorizin-induced NGD) on basal and exercise-stimulated, in vivo, whole body glucose turnover and IC glucose metabolism and ex vivo SkM glucose metabolism. Our major findings were that the prevailing hyperglycemia in poorly controlled diabetes was critical in ensuring normal preexercise and exercise-induced Rdtissue and GF and involved a compensatory increase in preexercise HGP and SkM total glucose and IC glucose levels but a decrease in preexercise MCRg and dMCRg. When the hyperglycemia was corrected by acute infusion of the glycosuric agent phlorizin (5), the response of the NGD dog to exercise was compromised and included a marked reduction in dRdtissue and dGF, and a sustained decrease in dMCRg, which was accompanied by excessive dglucagon and absolute FFA levels. In addition, despite the apparent normalization of preexercise SkM IC glucose levels with phlorizin, the mean SkM IC glucose level almost doubled (P = nonsignificant) with exercise in the NGD dogs, and, importantly, this value was similar to the value obtained in the exercising HGD state.

Therefore, the degree of hyperglycemia in the severely insulin-resistant hypoinsulinemic diabetic dogs is precisely balanced by the raised HGP and ongoing UrGloss and plays an important compensatory metabolic role in maintaining normal preexercise and exercise-induced Rdtissue and plasma-derived GF. When the same diabetic dogs underwent acute phlorizin-induced NGD, the importance of the precise balance between preexercise Ratotal, UrGloss, and Rdtissue was reinforced. Under these conditions, both Ratotal and UrGloss increased significantly by a further ~10 µmol·kg–1·min–1 but still resulted in normal preexercise Rdtissue and GF. However, with the extra stress of exercise during phlorizin infusion, both dRdtissue and dGF were severely reduced in the NGD state. Interestingly, we still found a strong positive correlation between preexercise glucagon and HGP levels for both the combined and diabetic states, even when glucose was infused in three of the six NGD studies, which supports the role of glucagon in regulating preexercise plasma glucose and HGP (2).

Importantly, our matched studies in the same dogs revealed that both preexercise MCRg and dMCRg were significantly reduced in both diabetic states compared with the prealloxan state, although an improvement in preexercise MCRg did occur in the NGD state. These data imply that the chronic hyperglycemic and insulin-resistant diabetic state itself may lead to an underlying chronic glucose-toxic state (12, 38, 39). Fisher et al. (14) argued that the impaired MCRg and dMCRg in the hyperglycemic diabetic dogs indicated reduced SkM transmembrane glucose transport and GLUT4 availability. However, when the exercise intensity surpasses ~40% O2 max, IC glucose processing itself becomes rate determining for SkM Rdtissue(ex) (15, 36, 49) and presumably for dMCRg, a fact acknowledged by Zinker et al. (56). Given that the activation of SkM hexokinase II is critical to Rdtissue(ex) (15), the reduced activation of SkM hexokinase II and PDH that occurs in states of chronic diabetes (18, 50) would limit the supply of pyruvate (i.e., reduced GF) (30) and necessitate an increased availability of FFA (17, 53). This situation would lead to a rise in SkM IC glucose, a scenario that was observed in our HGD dogs both before and during moderate exercise (~65% O2 max) and in the same phlorizin-induced NGD dogs during the extra stress of exercise. These observations are also consistent with our failure to observe a significant improvement in dMCRg in the NGD state. Our findings of impaired dMCRg and dRdtissue in our NGD dogs contrast with those of Fisher et al. (14) and may reflect differences in the intensity of exercise (~65 vs. ~45% O2 max), the metabolic status of the animals, and the availability of GLUT4 receptors at the onset of exercise. It also cautions against overemphasizing the change or lack of change in dMCRg as a marker of GLUT4 availability in diabetic states, because the postmembrane defects in SkM glucose metabolism that characterize the diabetic state itself (e.g., impaired activation of hexokinase II and/or PDH) may persist and impact on dMCRg and dRdtissue (15), despite the restoration of near-NGD with phlorizin. Therefore, although lowering plasma glucose by various methods (e.g., by increasing insulin secretion or decreasing glucose production) is important in reducing glucose toxicity, our data indicate that glucose toxicity cannot account for all of the abnormalities of diabetes. However, regardless of the difficulties in interpreting the pathophysiological significance of MCRg and dMCRg, our findings clearly support an important compensatory role for the hyperglycemia per se in poorly controlled diabetes in maintaining a sufficient supply of plasma glucose to the working muscle. This supports the original conclusion of Wahren et al. (51) in hyperglycemic Type 1 diabetic subjects.

The other important aim of the present study was to determine whether the source of energy for the working SkM is altered by phlorizin-induced NGD compared with the HGD state. During moderate exercise (~55–65% O2 max), ATP in working SkM is generated from glucose and fat oxidation, with carbohydrate and fat substrates contributing approximately equally to the total energy expenditure (36, 49). At this exercise intensity, the supply of glucose from SkM glycogen and plasma glucose for oxidation in SkM occurs at a ratio of ~3 to 1 (36, 49). We found that preexercise GF (derived from plasma glucose) was similar in the three treatment groups, but dRdtissue and dGF (a measure of exercise-induced plasma glucose oxidation) (36, 49) were critically dependent on the prevailing glycemia, both being significantly reduced in the NGD state. Moreover, preexercise plasma glucose positively correlated with both dRdtissue and dGF for the diabetic states. In addition, we found a strong positive correlation between preexercise GF and Rdtissue(ex) for the combined (n = 18) and two diabetic states (n = 12), and both parameters were inversely related to preexercise FFA and FFA with exercise levels. We also found that preexercise SkM IC glucose positively correlated with dRdtissue, and dGF showed a positive correlation with dSkM lactate for the diabetic states. Together, these data suggest that the preferred fuel substrate for oxidation (plasma glucose or FFA) in working diabetic SkM is dependent on the prevailing glycemia, availability of FFA, and SkM IC glucose, resulting in alterations in dRdtissue, dGF, and dSkM lactate. That is, on removing the hyperglycemic pump (mass action of glucose) in diabetes, dRdtissue and dGF are impaired due to a diminished fuel supply from plasma glucose and SkM IC glucose, resulting in a shift away from carbohydrate utilization during moderate exercise. This also suggests that the fuel requirements of working diabetic SkM are regulated by IC glucose metabolism rather than by Rdtissue(ex). Whether this can be attributed directly to the increased availability of FFA in the NGD state, as postulated originally by Randle et al. (35), or to the reduced mass-action effect of glucose itself, as suggested by the reverse-Randle hypothesis (42), is uncertain. Additionally, the state of reduced mass action of glucose, impaired GF, and excess FFA availability in contracting SkM, as seen in our exercising NGD dogs, may further limit the activation of PDH, which may contribute to a reduction in glucose oxidation (4, 15, 30).

It is unlikely that the relatively low preexercise total insulin levels observed in the two diabetic states compared with the prealloxan state (~3–6 vs. 9 mU/l) would directly and/or indirectly (via increased lipolysis and raised preexercise FFA levels) influence Rdtissue(ex) and dGF, given the "flatness" of the insulin-dose response curve for Rdtissue and glucose oxidation at low insulin levels (<10 mU/l) (46), the dominance of the exercise vs. insulin-stimulated effects on Rdtissue and glucose oxidation (36, 49), and the severely reduced insulin sensitivity seen in our poorly controlled diabetic dogs. Furthermore, although we found that preexercise total insulin showed a significant positive correlation with Rdtissue(ex) in both the combined and diabetic states, stepwise regression analysis revealed that it was not an independent predictor of Rdtissue(ex) in the diabetic states, confirming earlier reports (23, 34), and did not correlate with dGF.

The measurement of SkM IC glucose concentrations in subjects with differing plasma glucose concentrations may be considered problematic. However, the theoretical difficulties involving the key assumptions relating to the interstitial glucose concentration and the extracellular and IC water contents in the SkM biopsies have been extensively addressed in the past (32, 48). In addition, using these assumptions, we recently calculated SkM IC glucose concentrations in nondiabetic dogs at rest (plasma glucose ~5 mM) and during somatostatin-infused basal insulin replaced hyperglycemic clamps (plasma glucose ~13 mM), and in fasting alloxan-induced hyperglycemic diabetic dogs (plasma glucose ~14 mM), and found SkM IC glucose levels to be 0.7 ± 0.2, 1.0 ± 0.3, and 2.3 ± 0.4 mM IC water, respectively (Christopher MJ, Rantzau C, and Alford F, unpublished observations). These results confirm the findings of Katz et al. (21) and indicate that our measurements of SkM IC glucose concentrations appear to reflect the IC pathophysiological metabolic conditions present in SkM at rest and under conditions of hyperglycemia.

Finally, when the two diabetic states were combined, we found that Rdtissue(ex) significantly correlated with four preexercise metabolic parameters, namely plasma glucose, total insulin, total FFA, and GF, using nonparametric analysis. However, we confirmed that the only two independent predictors, namely preexercise GF and plasma glucose, combined to fit the stepwise regression model, yielding an r2 (adjusted) value of 88.1% (r = 0.94). In an earlier paper (5), our laboratory reported that preexercise SkM AMPK{alpha}1 and AMPK{alpha}2 isoform activities and site-specific phosphorylation of acetyl-CoA carboxylase were markedly increased in both the HGD and NGD states. Whether the elevated levels of these enzymes play a permissive role with the mass action of glucose to normalize the metabolic response to moderate exercise in the HGD state remains speculative (5).

In conclusion, in low-dose, alloxan-induced, 4- to 5-wk poorly controlled DHG dogs, the prevailing hyperglycemia plays a critical metabolic role in ensuring normal preexercise and moderate-exercise-induced rates of Rdtissue and GF, despite a markedly reduced dMCRg. When the mass-action effect of glucose is removed by acute phlorizin-induced (glycosuria mediated) NGD, the metabolic response to exercise is compromised, resulting in marked reductions in dRdtissue and dGF, and a sustained decrease in dMCRg, in the presence of partially normalized SkM IC glucose levels and increased circulating FFA. Although Rdtissue(ex) in the diabetic states significantly correlated with four preexercise metabolic parameters, namely plasma glucose, total insulin, total FFA, and GF, subsequent stepwise regression analysis revealed that only preexercise plasma glucose and GF combined to fit the model, accounting for 88% of Rdtissue(ex). Whether the impaired dGF in the DNG dogs can be directly attributed to the reduced availability of plasma-derived glucose (combined with the sustained reduction of dMCRg) or the increased availability of FFA remains to be determined. Therefore, the prevailing hyperglycemia in poorly controlled diabetes plays a critical compensatory role in maintaining a sufficient fuel supply from plasma glucose to meet the increased energy demands in working SkM.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
ACKNOWLEDGMENTS
 REFERENCES
 
This study was supported by grants from the Juvenile Diabetes Foundation International/Australia, the National Health and Medical Research Council of Australia, and the Diabetes Australia Research Trust.


   ACKNOWLEDGMENTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 ACKNOWLEDGMENTS
REFERENCES
 
The authors are grateful to Sue McKay and Liliana Pepe for performing arteriovenous shunts and venous catheterizations on the dogs and for assistance with muscle biopsies. We thank Selma Gotsbacher and Sally Mifsud and the staff of the Experimental Medical and Surgical Unit for care, health monitoring, and exercising of the dogs, and Jane Ford for secretarial assistance.


   FOOTNOTES
 

Address for reprint requests and other correspondence: M. J. Christopher, Dept. of Endocrinology and Diabetes, St. Vincent's Hospital Melbourne, 35 Victoria Parade, Fitzroy 3065, Victoria, Australia (E-mail: alfordfp{at}svhm.org.au)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 

  1. Bak JF, Jacobsen UK, Jorgensen FS, and Pedersen O. Insulin receptor function and glycogen synthase activity in skeletal muscle biopsies from patients with insulin dependent diabetes mellitus. J Clin Endocrinol Metab 69: 158–164, 1989.[Abstract]
  2. Brand CL, Jorgensen PN, Svendsen I, and Holst JJ. Evidence for a major role for glucagon in regulation of plasma glucose in conscious, non-diabetic and alloxan-induced diabetic rabbits. Diabetes 45: 1076–1083, 1996.[Abstract]
  3. Caruso G, Proietto J, Calenti A, and Alford F. Insulin resistance in alloxan-diabetic dogs: evidence for reversal following insulin therapy. Diabetologia 25: 273–279, 1983.[Medline]
  4. Challis RAJ, Vranic M, and Radda GK. Bioenergetic changes during contraction and recovery in diabetic rat skeletal muscle. Am J Physiol Endocrinol Metab 256: E129–E137, 1989.[Abstract/Free Full Text]
  5. Christopher MJ, Chen ZP, Rantzau C, Kemp BE, and Alford FP. Skeletal muscle basal AMP-activated protein kinase activity is chronically activated in alloxan-diabetic dogs: impact of exercise. J Appl Physiol 95: 1523–1530, 2003.[Abstract/Free Full Text]
  6. Christopher MJ, Rantzau C, Ward GM, and Alford FP. Impact of variable insulinemia and glycemia on in vivo glycolysis and glucose storage in dogs. Am J Physiol Endocrinol Metab 266: E62–E71, 1994.[Abstract/Free Full Text]
  7. Christopher MJ, Rantzau C, Ward GM, and Alford FP. Hypoinsulinemia and hyperglycemia influence the in vivo partitioning of GE and SI. Am J Physiol Endocrinol Metab 268: E410–E421, 1995.[Abstract/Free Full Text]
  8. Christopher MJ, Sleeman MW, Alford FP, and Best JD. Contrasting action of short and long-term adrenaline infusion on dog skeletal muscle glucose metabolism. Diabetologia 35: 399–405, 1992.[CrossRef][ISI][Medline]
  9. Coker RH, Krishna MG, Brooks Lacy D, Allen EJ, and Wasserman DH. Sympathetic drive to liver and nonhepatic splanchnic tissue during heavy exercise. J Appl Physiol 82: 1244–1249, 1997.[Abstract/Free Full Text]
  10. Crowther GJ, Milstein JM, Jubrias SA, Kushmerick MJ, Gronka RK, and Conley KE. Altered energetic properties in skeletal muscle of men with well-controlled insulin-dependent (Type 1) diabetes. Am J Physiol Endocrinol Metab 284: E655–E662, 2003.[Abstract/Free Full Text]
  11. Del Prato S, Bonadonna RC, Bonora E, Gulli G, Solini A, Shank M, and DeFronzo RA. Characterization of cellular defects of insulin action in Type 2 (non-insulin-dependent) diabetes mellitus. J Clin Invest 91: 484–494, 1993.[ISI][Medline]
  12. Dimitrakoudis D, Ramlal T, Rastogi S, Vranic M, and Klip A. Glycemia regulates the glucose transporter number in the plasma membrane of rat skeletal muscle. Biochem J 284: 341–348, 1992.
  13. Douen AG, Ramlal T, Rastogi S, Bilan PJ, Cartee GD, Vranic M, Holloszy JO, and Klip A. Exercise induces recruitment of the "insulin responsive glucose transporter." Evidence for distinct intracellular insulin- and exercise-recruitable transporter pools in skeletal muscle. Biol Chem 265: 13427–13430, 1990.
  14. Fisher SJ, Lekas M, Shi ZQ, Bilinski D, Carvalho G, Giacca A, and Vranic M. Insulin-dependent acute restoration of euglycemia normalises the impaired glucose clearance during exercise in diabetic dogs. Diabetes 46: 1805–1812, 1997.[Abstract]
  15. Fueger PT, Bracy DP, Malabanan CM, Pencek RR, and Wasserman DH. Distributed control of glucose uptake in working muscles of conscious mice: roles of transport and phosphorylation. Am J Physiol Endocrinol Metab 286: E77–E84, 2004.[Abstract/Free Full Text]
  16. Goodyear LJ, Hirshman MF, King PA, Thompson CM, Horton ED, and Horton ES. Skeletal muscle plasma membrane glucose transport and glucose transporters after exercise. J Appl Physiol 68: 193–198, 1990.[Abstract/Free Full Text]
  17. Hagenfeldt L. Metabolism of free fatty acids and ketone bodies during exercise in normal and diabetic man. Diabetes 28: 66–70, 1979.
  18. Hagg SA, Taylor SI, and Ruderman NB. Glucose metabolism in perfused skeletal muscle: pyruvate dehydrogenase in starvation, diabetes and exercise. Biochem J 158: 203–210, 1976.[ISI][Medline]
  19. Hew FL, Koschmann M, Christopher M, Rantzau C, Vaag A, Ward G, Beck-Nielsen H, and Alford F. Insulin resistance in growth hormone-deficient adults: defects in glucose utilization and glycogen synthase activity. J Clin Endocrinol Metab 81: 555–564, 1996.[Abstract]
  20. Hother-Nielsen O, Henriksen JE, Holst JJ, and Beck-Nielsen H. Effects of insulin on glucose turnover rates in vivo: isotope dilution versus constant specific activity technique. Metabolism 45: 82–91, 1996.[CrossRef][ISI][Medline]
  21. Katz A, Raz I, Spenser MK, Rising R, and Mott DM. Hyperglycemia induces glucose accumulation in human skeletal muscle. Am J Physiol Regul Integr Comp Physiol 260: R698–R705, 1991.[Abstract/Free Full Text]
  22. Kennedy JW, Hirshman MF, Gervino EV, Ocel JV, Forse RA, Hoenig SJ, Aronson D, Goodyear LJ, and Horton ES. Acute exercise induces GLUT4 translocation in skeletal muscle of normal human subjects and subjects with Type 2 diabetes. Diabetes 48: 1192–1197, 1999.[Abstract]
  23. Klip A, Marette A, Dimitrakoudis D, Ramlal T, Giacca A, Shi ZQ, and Vranic M. Effect of diabetes on glucoregulation. From glucose transporters to glucose metabolism in vivo. Diabetes Care 15: 1747–1766, 1992.[Abstract]
  24. Martin IK, Weber KM, Boston RC, Alford FP, and Best JD. Effects of epinephrine infusion on determinants of intravenous glucose tolerance in dogs. Am J Physiol Endocrinol Metab 255: E668–E673, 1988.[Abstract/Free Full Text]
  25. McConell GK, Canny BJ, Daddo MC, Nance MJ, and Snow RJ. Effect of carbohydrate ingestion on glucose kinetics and muscle metabolism during intense endurance exercise. J Appl Physiol 89: 1690–1698, 2000.[Abstract/Free Full Text]
  26. Nankervis A, Proietto J, Aitken P, and Alford FP. Impaired insulin activation in newly diagnosed Type 1 (insulin-dependent) diabetes mellitus. Diabetologia 27: 497–503, 1984.[CrossRef][Medline]
  27. Nankervis A, Proietto J, Aitken P, Harewood M, and Alford F. Differential effects of insulin therapy on hepatic and peripheral insulin sensitivity in Type 2 (non-insulin-dependent) diabetes. Diabetologia 23: 320–325, 1982.[ISI][Medline]
  28. Nuutila P, Peltoneimi P, Oikonen V, Larmola K, Kemppainen J, Takala T, Sipila H, Oksanen A, Ruotsalainen U, Bollie GB, and Yki-Jarvinen H. Enhanced stimulation of glucose uptake by insulin increases exercise-stimulated glucose uptake in skeletal muscle in humans. Diabetes 49: 1084–1091, 2000.[Abstract]
  29. Peltoneimi P, Yki-Jarvinen H, Oikonen V, Oksanen A, Takala TO, Ronnemaa T, Erkinjuntti M, Knuuti MJ, and Nuutila P. Resistance to exercise-induced increase in glucose uptake during hyperinsulinemia in insulin resistant skeletal muscle of patients with Type 1 diabetes. Diabetes 50: 1371–1377, 2001.[Abstract/Free Full Text]
  30. Peters SJ, Harris RA, Wu P, Pehleman TL, Heigenhauser GJF, and Spriet LL. Human skeletal muscle PDH kinase activity and isoform expression during 3 days of a high fat low CHO diet. Am J Physiol Endocrinol Metab 281: E1151–E1158, 2001.[Abstract/Free Full Text]
  31. Ploug T, Galbo H, Ohkuwa T, Tranum-Jensen J, and Vinten J. Kinetics of glucose transport in rat skeletal muscle membrane vesicles: effects of insulin and contractions. Am J Physiol Endocrinol Metab 262: E700–E711, 1992.[Abstract/Free Full Text]
  32. Poulin RA, Steil GM, Moore DM, Ader M, and Bergman RN. Dynamics of glucose production and uptake are more closely related to insulin in hindlimb lymph than in thoracic duct lymph. Diabetes 43: 180–190, 1994.[Abstract]
  33. Putman CT, Jones NL, Lands LC, Bragg TM, Hollidge-Horvat MG, and Heigenhauser GJ. Skeletal muscle pyruvate dehydrogenase activity during maximal exercise in humans. Am J Physiol Endocrinol Metab 269: E458–E468, 1995.[Abstract/Free Full Text]
  34. Raguso CA, Coggan AR, Castaldelli A, Sidossis LS, Bastyr EJ, and Wolfe RR. Lipid and carbohydrate metabolism in IDDM during moderate and intense exercise. Diabetes 44: 1066–1074, 1995.[Abstract]
  35. Randle PJ, Garland PB, Hales CN, and Newsholme EA. The glucose-fatty acid cycle: its role in insulin sensitivity and metabolic disturbances of diabetes. Lancet 1: 785–789, 1963.[ISI][Medline]
  36. Romijn JA, Coyle EF, Sidossis LS, Gastaldelli A, Horowitz JF, Endert E, and Wolfe RR. Regulation of endogenous fat and carbohydrate metabolism in relation to exercise intensity and duration. Am J Physiol Endocrinol Metab 265: E380–E391, 1993.[Abstract/Free Full Text]
  37. Rossetti L and Giaccari A. Relative contribution of glycogen synthesis and glycolysis to insulin-mediated glucose uptake. A dose-response euglycemic clamp study in normal and diabetic rats. J Clin Invest 85: 1785–1792, 1990.[ISI][Medline]
  38. Rossetti L, Giaccari A, and De Fronzo RA. Glucose toxicity. Diabetes Care 13: 610–630, 1990.[Abstract]
  39. Rossetti L, Smith D, Shulman GI, Papachristou D, and De Fronzo RA. Correction of hyperglycemia with phlorizin normalises tissue sensitivity to insulin in diabetic rats. J Clin Invest 79: 1510–1515, 1987.[ISI][Medline]
  40. Schalin-Jantti C, Harkonen M, and Groop LC. Impaired activation of glycogen synthase in people at increased risk for developing NIDDM. Diabetes 41: 598–604, 1992.[Abstract]
  41. Shi ZA, Yamatani K, Fisher SJ, Lavina H, Lickley A, and Vranic M. Effects of subbasal insulin infusion on resting and exercise-induced glucose turnover in depancreatectomized dogs. Am J Physiol Endocrinol Metab 264: E334–E341, 1993.[Abstract/Free Full Text]
  42. Sidossis LS and Wolfe RR. Glucose and insulin-induced inhibition of fatty acid oxidation: the glucose-fatty acid cycle reversed. Am J Physiol Endocrinol Metab 270: E733–E738, 1996.[Abstract/Free Full Text]
  43. Sjogaard G, Adams RP, and Saltin B. Water and ion shifts in skeletal muscle of humans with intense dynamic knee extension. Am J Physiol Regul Integr Comp Physiol 248: R190–R1