Hepatic adaptations to iron deficiency and exercise training W. T. WILLIS,





Department of Pediatrics, School of Medicine, University of California, San Francisco 94143; Department of Human Performance, San Jose State University, San Jose, California 95192; and Exercise and Sport Research Institute, Arizona State University, Tempe, Arizona 85287 WILLIS, W.T.,P.H. JONES,R. CHENGSON,AND P.R. DALLMAN. Hepatic adaptations to iron deficiency and exercise training. J. Appl. Physiol. 73(2): 510~515,1992.-Brooks et al. [Am. J. Physiol. 253 (Endocrinol. Metab. 16): E461-E466,1987] demonstrated an elevated gluconeogenicrate in resting iron-deficient rats. Becausephysical exercise also imposesdemand on this hepatic function, we hypothesized that exercise training superimposedon iron deficiency would augment the hepatic capacity for amino acid transamination/deamination and pyruvate carboxylation. Sprague-Dawley rats (n = 32) were obtained at weaning (21 days of age) and randomly assignedto iron-sufficient (dietary iron = 60 mg iron/kg diet) or iron-deficient (3 mg iron/kg) dietary groups. Dietary groupswere subdivided into sedentary and trained subgroups.Treadmill training was4 wk in duration, 6 days/wk, 1 h/day, 0% grade. Treadmill speedwas initially 26.8 m/min and was decreasedto 14.3 m/ min over the 4-wk training period. The mild exercise-training regimen did not affect any measuredvariable in iron-sufficient rats. In contrast, in iron-deficient animals, training increased endurancecapacity threefold and reducedblood lactate and the lactate-to-alanine ratio during submaximal exercise by 34 and 27%,respectively. The mitochondrial oxidative capacity of gastrocnemius muscle was increased 46% by training. However, the oxidative capacity of liver was not affected by either iron deficiency or training. Maximal rates of pyruvate carboxylation and glutamine metabolismby isolated liver mitochondria were also evaluated. Iron deficiency and training interacted to increasepyruvate carboxylation by intact mitochondria. Glutamine metabolismwasincreasedroughly threefold by iron deficiency alone, and training amplified this effect to a ninefold increase over iron-sufficient animals. It is concluded that superimposing the stressof regular exerciseon dietary iron de& ciency stimulates adaptative increasesin liver mitochondrial functions associatedwith pyruvate carboxylation and hepatic nitrogen metabolism. glutaminase; gluconeogenesis;pyruvate carboxylation; liver; hepatic mitochondria

DIETARY IRON DEFICIENCY results in anemia and marked

deficits in the mitochondrial oxidative capacity of skeletal muscle (8). In contrast, many (10,23,26,34), but not all (5,9,20), studies have shown that the mitochondria of liver are largely spared from the effects of dietary iron deficiency. Furthermore, non-iron-containing liver enzymes involved in the processes of transamination and deamination, specifically, alanine and aspartate aminotransferases and glutamate dehydrogenase, are increased in response to severe dietary iron deficiency (34). 510


When iron deficiency is corrected by intraperitoneal administration of iron dextran, the activities of these transdeaminating enzymes decrease toward control levels along the same time course as the increase in muscle oxidases, raising the possibility of a functional interrelationship (2). It has been suggested (6) that in iron deficiency the liver may play an especially important role in supporting the metabolic needs of skeletal muscle, which depends heavily on glycolytic energy production (11,21) due presumably to the mitochondrial impairment in muscle (12, 33). In normal, i.e., iron-sufficient, mammals, the liver plays a central role in maintaining glucose homeostasis during exercise (36). Hepatic gluconeogenesis is a critical source of blood glucose during exercise (17). Moreover the adaptive response to regular exercise (training) may include increases in hepatic gluconeogenic capacity (27, 29) and alanine and aspartate aminotransferase activities (16). It would seem reasonable to hypothesize that regular exercise might constitute an even greater stimulus for adaptation in the liver of the iron-deficient mammal, given the observations that 1) the iron-deficient rodent demonstrates an increased gluconeogenic flux even at rest (6) and 2) aerobic adaptations in skeletal muscle may be restricted by limited iron availability, resulting in an increased reliance on Cori cycle activity (32, 34). The primary sources of gluconeogenic carbon during exercise are lactate, pyruvate, glycerol, and amino acids. Of the amino acid efflux from exercising muscle, alanine and glutamine, represent -60% of the total carbon (1, 24). Liver mitochondria play a central role in hepatic gluconeogenesis by providing both enzymatic catalysis and ATP to the gluconeogenic pathway (4). The gluconeogenic pathways of lactate, pyruvate, and alanine all include the mitochondrial matrix enzyme pyruvate carboxylase. Groen et al. (13) showed that this enzyme step may dominate the control of gluconeogenic flux from these three carbon precursors. Although glutamine is not considered to be a major gluconeogenic precursor, at least under normal circumstances (24), it is thought to be an important carrier of amino acid nitrogen from skeletal muscle to the liver (15). Hepatic glutamine metabolism also involves an enzyme of the mitochondrial matrix, glutaminase. The purpose of the present investigation was to study the effects of exercise training on the liver mitochondria of iron-deficient rats. Specifically, we wished to determine the effects of training on the mito-

$2.00 Copyright 0 1992 the American Physiological


Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} ( on January 15, 2019.





chondrial content and oxidative capacity of liver and on the mitochondrial capacities for pyruvate carboxylation and glutamine metabolism. METHODS

Animal care and training. Thirty-two female SpragueDawley rats were obtained at weaning (21 days of age) and were randomly assigned to one of two dietary groups. Animals received one of two diets (Teklad, Madison, WI), which were assayed for iron content. The iron-sufficient group received 60 mg iron/kg diet, and the iron-deficient rats were fed an identical diet but containing 3 mg iron/kg. The animals were housed two per cage and had access to food and distilled water ad libitum. After 1 wk the dietary groups were subdivided into cage sedentary and exercise-trained subgroups, generating a 2 X 2 diet X exercise experimental design. Exercise training was carried out using a motorized treadmill (Quinton model 42-15). The design of the training regimen was based on what previous findings have shown to be the time course of the development of dietary iron deficiency (35). Weanling rats fed an irondeficient diet exhibit steady declines in blood hemoglobin level and muscle oxidative capacity. As a result, the treadmill running time at a submaximal aerobic exercise intensity * (endurance ) decreases along a somewhat similar time course (35). Accordingly, our training regimen was designed as a negative ramp; the duration and treadmill grade were held constant at 60 min and 0%, respectively, while the belt speed was set at 28.6 m/min for the first 2 wk, 20.1 m/min the 3rd wk, and 14.3 m/min the final week. The animals were exercised 6 days/wk. Our prediction that this downward sloping regimen would be a more effective training stimulus than our previous protocol (34) was confirmed by its effects on the mu scle oxidativ ‘e capacity of exerci sed iron-deficient rats (see below). Untrained animals were acclimatized to exercise by subjecting them to 5 min of treadmill running at 14.3 m/min, 2 days/wk. Exercise tests. Iron-deficient animals were subjected to two types of exercise tests. On the last training day, irondeficient animals, both trained and untrained, were run to exhaustion, and the time to fatigue was recorded. Fatigue was identified according to criteria established by the University of California, Berkeley, Animal Care and Use Committee as the point when the rat landed on the shock grid five times within 2 min or when the animal was unable to return to the treadmill from the shock grid. We previously observed that the fatigue point in iron-deficient rats is almost always preceded by a gait change from trot to gallop and by the appearance of weaving in the treadmill stall. By carefully observing the iron-deficient animals’ locomotory behavior, we were able to anticipate shock grid landings and thus minimize stress to the animals. On the following day all rats were run at 14.3 m/min for 10 min. At the end of the lo-min period, tail vein blood was collected for analysis of blood lactate and alanine. Iron-sufficient rats underwent only the lo- ,min submaximal exercise test. Tissue preparation. Trained animals were maintained on the training regimen until they were killed; this period





never exceeded 3 training days past the exercise tests. Rats were allowed 224 h after the last bout of exercise before they were killed. At approximately 20:00 on the evening before they were killed, food was withdrawn. At O&O0 on the following day, the rats were weighed and a sample of tail vein blood was taken for hemoglobin determination. The animals were then killed. The liver was quickly excised and placed in ice-cold 250 mM sucrose with 10 mM tris(hydroxy)aminomethane (Tris) base (pH 7.4). One gastrocnemius was then removed and placed in an ice-cold mannitol buffer containing 250 mM mannitol, 50 mM Tris, and 1 mM EDTA (pH 7.4). Both tissues were cleaned of visible connective tissue and fat and blotted dry, weighed, and suspended in their respective buffers at a 10% (wt/vol) concentration. Liver tissue was minced with scissors and homogenized with three complete passes of a Teflon pestle in a 50-ml glass tube. A small portion of this homogenate was used for determination of protein concentration and whole tissue succinate oxidase activity, and the majority was used for the isolation of liver mitochondria. Liver mitochondria were isolated according to standard procedures as established by Ji and co-workers (16). Mitochondrial pellets were washed two times and suspended in 250 mM sucrose with 10 mM Tris (pH 7.4) at a concentration of 40-60 mg protein/ml. Gastrocnemius muscle was minced with scissors and homogenized for 15 s at a power setting of 40% in a Tissumizer (Tekmar, Cincinnati, OH). BiochemicaZ assays. Hemoglobin was measured with the cyanomethemoglobin technique (Data Medical, Arlington, TX). The assays of lactate, alanine, glutamate, malate, and citrate were carried out enzymatically with NAD/NADH as electron acceptor/donor, according to standard spectrophotometric techniques (3). Mitochondrial oxidase activity, in either isolated liver mitochondria or whole tissue (liver or muscle) homogenates, was measured at 37OC in a 2.O-ml final volume of respiration medium with a Clark-type oxygen electrode (Rank, Oxford, UK). The respiration medium contained 12 mM mannitol, 45 mM sucrose, 15 mM KCl, 5 mM MgCl,, 7 mM EDTA, 0.2% bovine serum albumin, 25 mM Tris, and 30 mM KPO, (pH 7.4). The substrates used were 10 mM succinate + 5 PM rotenone, 10 mM glutamate + 2.5 mM malate, 10 mM pyruvate + 2.5 mM malate, and 10 mM 2-oxoglutarate. Maximal (state 3) respiration was induced with additions of 1 pmol ADP (0.5 mM final concn). Maximal pyruvate carboxylation by intact mitochondria was estimated according to the method of Walter and Stucki (30). The reaction mixture, 1.9 ml in a scintillation vial prewarmed to 37°C contained 220 mM sucrose, 10 mM MgSO,, 10 mM pyruvate, 6.6 mM triethanolamine, 6.6 mM KPO,, and 10 mM KHCO, (pH 7.4). A timed lo-min incubation was initiated by adding 4-6 mg of liver mitochondrial protein contained in 100 ~1. The reaction was stopped by adding 0.50 ml of 17.5% perchloric acid with rapid mixing. A portion of the acidified sample was decanted into a 1.5-ml tube and centrifuged (Beckman model B micro centrifuge) for 1.0 min. The supernatant was neutralized with 2 N KOH + 0.3 M 3-(N-morpholino)propanesulfonic acid and used to determine malate and citrate accumulation. In normal liver

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} ( on January 15, 2019.






mitochondria, at least, carbon flux through pyruvate carboxylase is largely accounted for as citrate and malate formation (30). The maximal activity of pyruvate carboxylase was assayed spectrophotometrically in disrupted mitochondria at 37OC by the method of Crabtree et al. (7). The reaction mixture contained 100 mM Tris . HCl, 5 mM MgCl,, 2.5 mM ATP, 0.75 mM acetyl-CoA, 0.20 mM 5,5’-dithiobis(2-nitrobenzoic acid), and 10 mM pyruvate (pH 7.4) with 1.0 unit added citrate synthase. Mitochondrial suspension (-40-60 mg/ml) was prediluted 1:lO in a 0.1% Trition X-100 solution with mixing, and 20 ~1 of diluted disrupted mitochondria were then promptly added to the reaction mix. Background activity was followed at 412 nm for 1.0 min, and then specific pyruvate carboxylase activity was initiated with the addition of NaHCO, (50 mM final concn). Activity was determined by following absorbance at 412 nm (A& in a Hitachi spectrophotometer (model 100-60) by use of a millimolar extinction coefficient of 13.6 for the mercaptide ion formed by condensation of DTNB with coenzyme A liberated in the citrate synthase reaction. The conversion of glutamine to glutamate by intact liver mitochondria was measured in scintillation vials at 37OC by the method of Joseph and McGivan (18). Vials contained 1.9 ml of reaction mixture: 120 mM KCl, 20 mM Tris. HCI, 10 mM KPO,, 20 mM KHCO,, 5 mM succinate, 10 pglml rotenone, and 15 mM glutamine (pH 7.3). A timed lo-min incubation was started with the addition of 100 ~1 of mitochondrial suspension (4-6 mg protein). The reaction was stopped with acidification with subsequent neutralization, as described for the pyruvate carboxylation incubation. In this assay, carbon flux through mitochondrial glutaminase accumulates as glutamate because of the presence of rotenone (18). Protein concentrations in tissue homogenates and mitochondrial suspensions were determined in the presence of 1% (wt/vol) deoxycholate by the biuret method. Statistics. The data were analyzed using two-way analysis of variance (ANOVA) with interactions (Crunch Software, San Francisco, CA). The method of least squares was used to adjust for unequal cell sizes. Differences between two mean values were determined with the Newman-Keuls multiple range test. Significant differences were determined at the 0.05 level. All values are reported as means t SE. RESULTS

Although the exercise-training regimen employed in this study was very demanding for the iron-deficient rats, it was very mild by normal standards (33). Consequently, there were no differences between exercised and nonexercised iron-sufficient rats in any of the measured variables. This finding is consistent with previous studies (32, 34).

The blood hemoglobin concentration of iron-deficient rats was 40% of the value in the iron-sufficient animals, and exercise training did not impact on this deficit (Table 1). The body weights were 14% less in untrained irondeficient rats than in iron-sufficient rats, and exercise training amplified this difference to 22%. The absolute




gastrocnemius muscle weights were significantly less in the iron-deficient groups. However, the gastrocnemius weight relative to body weight was maintained in the untrained iron-deficient rats and was significantly increased with training (Table 1). Skeletal muscle oxidase activity. Meticulous care was employed in the present study to prevent exposure of exercising iron-deficient rats to adventitious iron during treadmill training sessions. These procedures were identical to a previous study that did not show training-induced increases in muscle oxidative capacity (34). We believe that the training protocol used in the present study (see METHODS) provided a more powerful training stimulus and is responsible for the marked increases (45%) in muscle oxidative capacity observed in trained iron-deficient rats (Table 1). It should be noted that evidence of training-induced improvements in iron status is localized to skeletal muscle; no effects of training were observed in the hemoglobin levels of iron-deficient rats (Table 1). Metabolic response to submaximal exercise. The untrained iron-deficient rodents were able to walk on the level at 14.3 m/min for only 13.2 t 0.9 min (n = 6). Trained iron-deficient rats walked 45.0 t 5.1 min (n = lo), a greater than threefold increase. Iron-sufficient rats were not tested for endurance in the present study, because previous data indicated endurance times >5 h at the mild exercise intensities used for iron-deficient rats (34). Blood lactate levels during the lo-min walking exercise at 14.3 m/min were markedly increased (Table 2) by dietary iron deficiency, in agreement with previous findings (33). In iron-deficient rats, training decreased the lactate concentrations during exercise, which also corroborates previous results (34). As a result, in iron-deficient rats training attenuated exercise-induced increases in the blood lactate-to-alanine ratio (Table 2). Liver biochemistry. The central concern of this study was the effects of iron deficiency and exercise training on liver mitochondria. Mitochondrial function was evaluated in three ways: oxidative capacity, pyruvate carboxylation, and glutamine metabolism. There were no effects of either iron deficiency or training on any oxidase pathway in liver mitochondria (Table 3). In addition, the mitochondrial content of liver (mg mitochondrial protein/g wet liver) was calculated using succinate + rotenone oxidase activity measured in both whole tissue homogenate and isolated mitochondria. This procedure also revealed no differences in the mitochondrial content of liver tissue (Table 3). Pyruvate carboxylation was estimated in isolated intact mitochondria by measuring the accumulation of malate and citrate after a lo-min incubation of mitochondria in the presence of 10 mM pyruvate and 10 mM bicarbonate. This estimation is based on the findings of Walter and Stucki (30), who showed that, at least in intact liver mitochondria isolated from normal rats, malate and citrate accounted for most of the carbon flux through pyruvate carboxylase. Although dietary iron deficiency did not significantly affect the formation of malate + citrate, two-way ANOVA indicated that iron deficiency and training interacted to significantly increase pyruvate

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} ( on January 15, 2019.









Gastr Oxidase, pm01 g-l. min-’

203.3t2.2 196.2k9.0

125+_3 127&l

174.2+5.4* 154.0&4.4*-t

are means

k SE; n, no. of rats.

2.77kO.14 2.60t0.12

50*3* 54+3*





is expressed


The data in this report, as well as findings from previous studies (2, 34), support the conclusion that iron deficiency is associated with increases in hepatic enzymes involved in amino acid transamination and deamination and gluconeogenesis. Furthermore the effect of dietary iron deficiency is amplified by the superimposition of exercise training. Moreover, dietary iron deficiency resulted in lower body weight in sedentary rats by 14 and by 22% in trained animals. Recently, Tobin and Beard (28) reported similar body weight deficits in irondeficient rats, although food consumption was similar to that of control animals. In addition, they found that the preponderance of the weight deficit was accounted for by lower whole body protein content (28). Lower whole body protein content and increased hepatic capacity for amino acid transamination and deamination in trained iron-deficient rats suggests the possibility of an increased pro-




1.08t0.04 1.07+0.06

5.44t0.07 5.41+0.23







in g/l blood

and gastrocnemius

Lactate, mM

Alanine, mM


7 6

13.Otl.O* 8.6&1.2*-f



2.07kO.20 1.71+0.30


6.87t_O.50* 5.03+0.47*t


Values are means k SE; n, no. of rats. * Iron deficient vs. iron sufficient, P < 0.05; t iron-deficient trained vs. iron-deficient sedentary, P < 0.05.

10 mM

3. Liver mitochondrial content and oxidase maximal activities




trained vs. iron-defi-


1.00+0.06 1.181kO.05




Iron sufficient Sedentary Trained

5 6


Maximal (State 3) Oxidase nmol O2 mg mito protein-l

2. Blood lactate and alanine levels after LO min of running n


teolysis in these animals. It should be noted, however, that the observed relative hypertrophy of the gastrocnemius muscle (Table 1) indicates that iron-deficient skeletal muscles that are involved in exercise training are not diminished in mass through proteolysis. Trained iron-deficient animals demonstrated significantly reduced lactate-to-alanine ratios during submaximal exercise, which may indicate a greater relative deposition of pyruvate into the alanine pool as opposed to the lactate pool. Although the source of the carbon skeleton of alanine is currently controversial, the amino group is generally thought to derive from the transamination of other amino acids (22, 24). Thus a greater tendency to divert pyruvate into the amino acid pool is consistent with a greater overall drive toward the transamination and deamination of amino acids. Because a whole body protein deficit and evidence for an elevated alanine flux during exercise suggest an increased rate of proteolysis, it is interesting that the capacity of liver mitochondria for glutamine metabolism was elevated threefold by iron deficiency and that this effect was amplified to a ninefold increase by the superimposition of training. A consistent observation in our laboratory has been that hepatic mitochondrial glutaminase increases only when iron deficiency is severe enough to result in a body weight deficit (unpublished observations). Increases in glutamine metabolism would provide for an increased nitrogen delivery from muscle to the liver (16). Haussinger (15) proposed that mitochondrial glutaminase acts as an amplification system to enhance the delivery of ammonia to the urea cycle. Given the fact


Sedentary Trained


g wet muscle-‘. * Iron deficient vs. iron sufficient, P < 0.05; t Iron-deficient

carboxylation. The enzymatic basis of this apparent increase, however, is unclear, inasmuch as the maximal activity of pyruvate carboxylase measured in disrupted mitochondria was not significantly altered by the superimposition of training on iron deficiency (Table 4). Glutamine metabolism by intact mitochondria was increased threefold by iron deficiency alone, and the superimposition of training induced a ninefold increase in glutamine utilization in trained iron-deficient rats compared with their iron-sufficient counterparts (Fig. 1). In this assay, glutamine flux through the mitochondrial enzyme glutaminase accumulates as glutamate due to the inclusion of rotenone to block glutamate oxidation. Succinate (10 mM) is included to ensure an intramitochondrial [ATP] that is favorable to glutaminase activity (18).

Iron sufficient Sedentary Trained

Wt, g


%oxoglutarate as substrate as pmol 0, min-’ cient sedentary, P < 0.05.





Iron sufficient Sedentary Trained Iron deficient Sedentary Trained




1. Hemoglobin, muscle oxidase, and body and gastrocnemius muscle weight




Iron deficient Sedentary Trained

Content, wk



Activity, min-’ l






44.Ok2.5 (4) 38.7k3.4 (5)



+ Ma1

78.4k5.0 (5)

12O.Ok7.4 (5)

46.92 1.5 (5)



49.2k2.8 (10)









(9) (9) (9) (9) Values are means t SE of no. of rats in parentheses. Pyr, pyruvate; Mal, malate.

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} ( on January 15, 2019.




that alanine and glutamine can account for -60% of amino acid flux from exercising muscle (1, Z4), an enhanced capacity for glutamine utilization would aptly serve the gluconeogenic and ureogenic needs of the carbohydrate-dependent iron-deficient rodent. Decreased pH inhibits glutaminase in the liver but stimulates it in the kidney (15). An augmented maximal activity of liver glutaminase would be advantageous in the trained iron-deficient mammal, where regular periods of exercise result not only in elevated gluconeog;nic demand (6) but also pronounced elevations in blood lactate (12) (Table 2), suggesting metabolic acidosis. In addition, recently, Welbourne (31) showed that metabolic acidosis (NH&l) results in a fivefold increase in glutamine efflux from perfused hindlimb skeletal muscle. The liver is the primary organ of glucose homeostasis. The observation that liver mitochondria are spared from iron deficiency supports the view that the preservation of adequate liver gluconeogenic function is critically important to support the glycolytic demands of iron-deficient tissues such as skeletal muscle (6). Gluconeogenesis not only requires mitochondrial ATP input, it is also sensitive to the cellular energy state, e.g., ATP/ADP. Groen et al. (13) showed that the mitochondrial matrix enzyme pyruvate carboxylase dominates the control of overall gluconeogenic flux when the carbon source enters the gluconeogenic pathway at this point, e.g., lactate, alanine, and pyruvate. Pyruvate carboxylase not only utilizes ATP in the carboxylation of pyruvate, it is also inhibited by decreased ATP/ADP (30). Moreover, in experiments with isolated liver cells, Berry et al. (4) demonstrated that overall gluconeogenic flux varies as a linear function of mitochondrial energy state. For these reasons, the maximal activities of various oxidase pathways were measured in isolated liver mitochondria; increased maximal respiratory capacity results in a higher extramitochondrial energy state at a given rate of ATP output (19, 33). Curiously, although no changes were found in either liver oxidase activities or maximal pyruvate carboxylase activity in disrupted mitochondria, there was a slight increase in pyruvate carboxylation by 4. Pyruvate carboxylution by intact and disrupted liver mitochondriu TABLE

Intact Group



Iron sufficient Sedentary





Trained Iron deficient Sedentary





Disrupted Ma1 + Cit

Pyr Carboxylase



(5) 33.4t2.7



(6) (6)

14.7kO.9 17.4t1.6 32.1t1.8 171.2t10.5 (9) (9) (9) (10) Trained 19.9t2.9 22.0t2.9 41.9t5.8* 191.4t16.9 (7) (7) (7) (9) Values are means t SE of no. of rats in parentheses, expressed as nmol . mg mitochondrial protein-‘. min? In intact liver mitochondria, pyruvate (Pyr) carboxylation was estimated by measuring accumulation of malate (Mal) and citrate (Cit). Maximal pyruvate carboxylase activity was measured spectrophotometrically in mitochondria disrupted in 0.1% Triton X-100. * Diet X training interaction, P < 0.05.





Sed Tr Sed Tr Iron Sufficient Iron Deficient FIG. 1. Glutamine metabolism by isolated intact liver mitochondria. Liver mitochondria were incubated for 10 min at 37OC in presence of 15 mM glutamine and rotenone (10 pglml) to block glutamate oxidation. Flux through mitochondrial glutaminase was measured by glutamate formation. Succinate (5 mM) was added as a respiratory substrate. *Iron deficient vs. iron sufficient, P < 0.05; **iron-deficient trained (Tr) vs. iron-deficient sedentary (Sed), P < 0.05.

intact mitochondria due to the interaction of iron de& ciency and training. The mechanism of this increase awaits further research. In summary, we found that dietary iron deficiency by itself resulted in a 14% lower body weight and, concomitantly, a threefold rise in liver mitochondrial capacity for glutamine utilization. The superimposition of training was associated with a further 22% body weight deficit, whereas liver glutaminase rose ninefold. In addition, iron deficiency and training interacted to significantly increase the maximal rate of pyruvate carboxylation by intact mitochondria. Given the important role of pyruvate carboxylation in determining the overall rate of hepatic gluconeogenesis (13), this increase suggests an augmented gluconeogenic capacity in trained iron-deficient rats. The authors acknowledge the helpful suggestions of Drs. Steven Gregg, Bradley Zinker, and Karen Klempa Gregg and the continued support and advice of Dr. George Brooks. Technical assistance was provided by Kim Fullerton. This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-13-897. Address for reprint requests: W. T. Willis, Dept. of Exercise Science and Physical Education, Arizona State University, Tempe, AZ 85287. Received 13 November 1990; accepted in final form 21 January 1992. REFERENCES 1. AHLBORG, G., P. FELIG, L. HAGENFELDT, R. HENDLER, AND J. WAHREN. Substrate turnover during prolonged exercise in man: splanchnic and leg metabolism of glucose, free fatty acids, and amino acids. J. Clin. Inuest.-53: 1080-1090, 1974. 2. AZEVEDO, J. L., W. T. WILLIS, L. P. TURCOTTE, A. S. ROVNER, P. R. DALLMAN, AND G. A. BROOKS. Reciprocal changes of muscle oxidases and liver enzymes with recovery from iron deficiency. Am. J. Physiol. 256 (Endocrinol. Metab. 19): E401-E405, 1989. 3. BERGMEYER, H. U. (Editor). Methods of Enzymatic Analysis. New York: Academic, 1985. 4. BERRY, M.N.,R.B. GREGORY, A.R. GRIVELL, D.C. HENLY, J.W. PHILLIPS, P.G. WALLACE,ANDG. R. WELCH. Linear relationships between mitochondrial forces and cytoplasmic flows argue for the organized energy-coupled nature of cellular metabolism. FERS Lett.

224: 201-207,


5. BEUTLER, E., AND R. K. BLAISDELL. Iron enzymes in iron deficiency. III. Catalase in rat red cells and liver with some further observations on cytochrome c. J. Lab. CZin. Med. 52: 694-699,1958. 6. BROOKS,G. A.,S.A. HENDERSON,AND P.R. DALLMAN. Increased glucose dependence in resting iron-deficient rats. Am. J. Physiol. 253 (EndocrinoZ. Metab. 16): E461-E466. 1987.

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} ( on January 15, 2019.





7. CRABTREE, B., S. J. HIGGINS, AND E. A. NEWSHOLME. The activities of pyruvate carboxylase, phosphoenolpyruvate carboxylase, and fructose diphosphatase in muscles from vertebrates and invertebrates. Biochem. J. 130: 391-396, 1972. 8. DALLMAN, P. R. Biochemical basis for manifestations of iron deficiency. Annu. Reu. Nutr. 6: 13-40, 1986. 9. DALLMAN, P. R., AND J. R. GOODMAN. The effects of iron deficiency on the hepatocyte: a biochemical and ultrastructural study. J. Cell BioZ. 48: 79-90, 1971. 10. EVANS, T. C., AND B. MACKLER. Effect of iron deficiency on energy conservation in rat liver and skeletal muscle submitochondrial particles. Biochem. Med. 34: 93-99, 1985. 11. FINCH, C. A., P. D. GOLLNICK, M. P. HLASTALA, L. R. MILLER, E. DILLMAN, AND B. MACKLER. Lactic acidosis as a result of iron deficiency. J. CZin. Invest. 64: 129-137, 1979. 12. FINCH, C. A., L. R. MILLER, A. R. INAMDAR, R. PERSON, K. SEILER, AND B. MACKLER. Iron deficiency in the rat. Physiological and biochemical studies of muscle dysfunction. J. CZin. Inuest. 58: 447-453, 1976. 13. GROEN, A. K., C. W. T. VAN ROERMUND, R. C. VERVOOM, AND J. M. TAGER. Control of gluconeogenesis in rat liver cells. Flux control coefficients of the enzymes in the gluconeogenic pathway in the absence and presence of glucagon. Biochen. J. 237: 379-389,1986. 14. HARLAN, W. R., AND R. S. WILLIAMS. Activity-induced adaptations in skeletal muscles of iron-deficient rabbits. J. Appl. Physiol. 65: 782-787,1988. 15. HAUSSINGER, D. Glutamine metabolism in the liver: overview and current concepts. Metabolism 38: 14-17, 1989. 16. JI, L. L., R. H. MILLER, F. J. NAGLE, H. A. LARDY, AND F. W. STRATMAN. Amino acid metabolism during exercise in trained rats: the potential role of carnitine in the metabolic fate of branchedchain amino acids. Metabolism 36: 748-752, 1987. 17. JOHN-ALDER, H. B., R. M. MCALLISTER, AND R. L. TERJUNG. Reduced running endurance in gluconeogenesis-inhibited rats. Am. J. Physiol. 251 (Regulatory Integrative Comp. Physiol. 20): R137R142,1986. 18. JOSEPH, S. K., AND J. D. MCGIVAN. The effects of ammonium chloride and bicarbonate on the activity of ,glutaminase in isolated liver mitochondria. Biochem. J. 176: 837-844, 1978. 19. KORETSKY, A. P., AND R. S. BALABAN. Changes in pyridine nucleotide levels alter oxygen consumption and extra-mitochondrial phosphates in isolated mitochondria: a 31P-NMR and NAD(P)H fluorescence study. Biochim. Biophys. Acta 893: 398-408, 1987. 20. MCKAY, R. H., D. A. HIGUCHI, W. W. WINDER, R. D. FELL, AND E. B. BROWN. Tissue effects of iron deficiency in the rat. Biochim. Biophys. Acta 757: 352-358,1983. 21. MCLANE, J. A., R. D. FELL, R. H. MCKAY, W. W. WINDER, E. B.

22. 23.








31. 32.









BROWN, AND J. 0. HOLLOSZY. Physiological and biochemical effects of iron deficiency on rat skeletal muscle. Am. J. Physiol. 241 (Cell Physiol. 10): C47-C54, 1981. NEWSHOLME, E. A., AND A. R. LEECH. Biochemistry for the Medical Sciences. New York: Wiley, 1983, p. 419-423. OHIRA, Y., J. HEGENAUER, L. STRAUSE, C.-S. CHEN, P. SALTMAN, AND H. BEINERT. Mitochondrial NADH dehydrogenase in iron-deficient and iron-repleted rat muscle: an EPR and work performance study. Br. J. Huemotol. 52: 623-630, 1982. PALMER, T. N., M. A. CALDECOURT, K. SNELL, AND M. C. SUGDEN. Alanine and inter-organ relationships in branched-chain amino and 2-0~0 acid metabolism. Biosci. Rep. 5: 1015-1033, 1985. PERKKIO, M. V., L. T. JANSSON, S. A. HENDERSON, C. REFINO, G. A. BROOKS, AND P. R. DALLMAN. Work performance in the irondeficient rat: improved endurance with exercise training. Am. J. Physiol. 249 (Endocrinol. Metab. 12): E306-E311, 1985. SALMON, H. A. The cytochrome c content of the heart, kidney, liver, and skeletal muscle of iron-deficient rats. J. Physiol. Land. 164: 17-30,1962. SUMIDA, K. D., J. H. URDIALES, AND C. M. DONOVAN. Enhanced hepatic gluconeogenesis from lactate following endurance training (Abstract). Med. Sci. Sports Exercise 22, Suppl.: S57, 1990. TOBIN, B. W., AND J. L. BEARD. The growth deficit of iron deliciency is exacerbated by intense physical activity (Abstract). FASEB J. 3: A449,1989. TURCOTTE, L. P., AND G. A. BROOKS. Effects of training on glucose metabolism of gluconeogenesis-inhibited short-term-fasted rats. J. AppZ. Physiol. 68: 944-954, 1990. WALTER, P., AND J. W. STUCKI. Regulation of pyruvate carboxylase in rat liver mitochondria by adenine nucleotides and short chain fatty acids. Eur. J. Biochem. 12: 508-519, 1970. WELBOURNE, T. C. Effect of metabolic acidosis on hindquarter glutamine and alanine release. Metabolism 35: 614-618, 1986. WILLIS, W. T., G. A. BROOKS, S. A. HENDERSON, AND P. R. DALLMAN. Effects of iron deficiency and training on mitochondrial enzymes in skeletal muscle. J. AppZ. Physiol. 62: 2442-2446, 1987. WILLIS, W. T., AND P. R. DALLMAN. Impaired control of respiration in iron-deficient muscle mitochondria. Am. J. Physiol. 257 (Cell Physiol. 26): C1080-C1085, 1989. WILLIS, W. T., P. R. DALLMAN, AND G. A. BROOKS. Physiological and biochemical correlates of increased work performance in trained iron-deficient rats. J. AppZ. Physiol. 65: 256-263, 1988. WILLIS, W. T., K. GOHIL, G. A. BROOKS, AND P. R. DALLMAN. Iron deficiency: improved exercise performance within 15 hours of iron treatment in rats. J. Nutr. 120: 909-916, 1990. WINDER, W. W. Control of hepatic glucose production during exercise. Med. Sci. Sports Exercise 17: 2-5, 1985.

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} ( on January 15, 2019.

Hepatic adaptations to iron deficiency and exercise training.

Brooks et al. [Am. J. Physiol. 253 (Endocrinol. Metab. 16): E461-E466, 1987] demonstrated an elevated gluconeogenic rate in resting iron-deficient rat...
2MB Sizes 0 Downloads 0 Views