Journal of Physiology (1995), 489.1, pp. 243-250 Carbohydrate ingestion and glycogen utilization in different muscle fibre types in man Orestis-Konstantinos Tsintzas, Clyde Williams *, Leslie Boobis and Paul Greenhaff Department of Physical Education, Sports Science and Recreation Management, Loughborough University, Loughborough, Leicsestershire LE11 3TU, UK 1. The effect of carbohydrate (CHO) ingestion on muscle glycogen utilization during exercise was examined on seven male subjects completing two 60 min treadmill runs at 70 % maximum oxygen uptake (1702 max), 1 week apart. On each occasion the subjects consumed either water or a 5 5% CHO-electrolyte solution immediately before and during exercise. Muscle samples were obtained from the vastus lateralis by needle biopsy before and immediately after exercise. Venous blood samples were also collected from an ante-cubital vein at rest and at 10, 20, 40 and 60 min into the run. 2. Higher blood glucose concentrations (P < 0-01) were observed throughout the run during the CHO trial compared with the water trial. Serum insulin concentration was only higher after 20 min of exercise (P < 0-01). 3. A 28% reduction in mixed glycogen utilization was observed as a result of CHO ingestion when compared with water ingestion (108-7 + 16-3 vs. 150-9 + 19-9 mmol (kg dry matter)-', respectively; P < 0-01). 4. The ingestion of the CHO solution resulted in sparing of glycogen in type I (slow twitch) fibres only (38 + 7% degradation of glycogen as opposed to 66 + 3% during the water trial; P = 0-01). There is some evidence that carbohydrate (CHO) ingestion during prolonged exercise delays the onset of fatigue by decreasing the rate of muscle glycogen utilization (Bjorkman, Sahlin, Hagenfeldt & Wahren, 1984; Erickson, Schwartzkopf & McKenzie, 1987). Other investigators have shown that carbohydrate ingestion exerts its ergogenic effect by simply complementing the total carbohydrate oxidation rate late in exercise, rather than sparing muscle glycogen (Coyle, Coggan, Hemmert & Ivy, 1986). All these studies have used cycling as the mode of exercise. However, there is a lack of information relating to running exercise. Furthermore, little is known about the effect of carbohydrate ingestion on muscle glycogen utilization in different fibre types. During prolonged running without CHO ingestion blood glucose concentration is maintained at euglycaemic levels (Williams, Nute, Broadbank & Vinall, 1990; Tsintzas, Liu, Williams, Gaitanos & Campbell, 1993). Moreover, the total carbohydrate oxidation rate does not seem to be affected by carbohydrate ingestion (Williams et al. 1990). It was hypothesized that under these conditions sparing of muscle glycogen could be responsible for an improvement in exercise performance as a result of carbohydrate ingestion (Tsintzas et al. 1993). Therefore, the aim of this study was to examine this hypothesis directly using the muscle biopsy technique and also to assess whether or not carbohydrate supplementation during treadmill running has a differential effect on glycogen sparing in type I (slow twitch) and type II (fast twitch) fibres. METHODS Subjects Seven male recreational runners gave their informed consent and volunteered to participate in this study, which was approved by the Loughborough University Ethical committee. The mean (± S.E.M.) age, height, weight, maximum oxygen uptake (V0o,max) and maximum heart rate (HRmax) of the subjects were 29-2 + 2 1 years, 176 3 + 1 8 cm, 75 2 + 4 1 kg, 54 5 + 2 0 ml kg-' min' and 190 0 + 22 beats min-', respectively. Preliminary measurements Following familiarization with treadmill running and experimental procedures, the subjects undertook two preliminary * To whom correspondence should be addressed. 4489 243 ) by guest on July 13, 2011 jp.physoc.org Downloaded from J Physiol (
Transcript
Journal of Physiology (1995), 489.1, pp. 243-250
Carbohydrate ingestion and glycogen utilization in differentmuscle fibre types in man
Orestis-Konstantinos Tsintzas, Clyde Williams *, Leslie Boobisand Paul Greenhaff
Department of Physical Education, Sports Science and Recreation Management,Loughborough University, Loughborough, Leicsestershire LE11 3TU, UK
1. The effect of carbohydrate (CHO) ingestion on muscle glycogen utilization during exercisewas examined on seven male subjects completing two 60 min treadmill runs at 70%maximum oxygen uptake (1702 max), 1 week apart. On each occasion the subjects consumedeither water or a 5 5% CHO-electrolyte solution immediately before and during exercise.Muscle samples were obtained from the vastus lateralis by needle biopsy before andimmediately after exercise. Venous blood samples were also collected from an ante-cubitalvein at rest and at 10, 20, 40 and 60 min into the run.
2. Higher blood glucose concentrations (P < 0-01) were observed throughout the run duringthe CHO trial compared with the water trial. Serum insulin concentration was only higherafter 20 min of exercise (P < 0-01).
3. A 28% reduction in mixed glycogen utilization was observed as a result of CHO ingestionwhen compared with water ingestion (108-7 + 16-3 vs. 150-9 + 19-9 mmol (kg dry matter)-',respectively; P < 0-01).
4. The ingestion of the CHO solution resulted in sparing of glycogen in type I (slow twitch)fibres only (38 + 7% degradation of glycogen as opposed to 66 + 3% during the watertrial; P = 0-01).
There is some evidence that carbohydrate (CHO) ingestionduring prolonged exercise delays the onset of fatigue bydecreasing the rate of muscle glycogen utilization(Bjorkman, Sahlin, Hagenfeldt & Wahren, 1984; Erickson,Schwartzkopf & McKenzie, 1987). Other investigators haveshown that carbohydrate ingestion exerts its ergogeniceffect by simply complementing the total carbohydrateoxidation rate late in exercise, rather than sparing muscleglycogen (Coyle, Coggan, Hemmert & Ivy, 1986). All thesestudies have used cycling as the mode of exercise. However,there is a lack of information relating to running exercise.Furthermore, little is known about the effect ofcarbohydrate ingestion on muscle glycogen utilization indifferent fibre types.
During prolonged running without CHO ingestion bloodglucose concentration is maintained at euglycaemic levels(Williams, Nute, Broadbank & Vinall, 1990; Tsintzas, Liu,Williams, Gaitanos & Campbell, 1993). Moreover, the totalcarbohydrate oxidation rate does not seem to be affected bycarbohydrate ingestion (Williams et al. 1990). It washypothesized that under these conditions sparing of muscleglycogen could be responsible for an improvement in
exercise performance as a result of carbohydrate ingestion(Tsintzas et al. 1993).
Therefore, the aim of this study was to examine thishypothesis directly using the muscle biopsy technique andalso to assess whether or not carbohydrate supplementationduring treadmill running has a differential effect onglycogen sparing in type I (slow twitch) and type II (fasttwitch) fibres.
METHODSSubjectsSeven male recreational runners gave their informed consent andvolunteered to participate in this study, which was approved bythe Loughborough University Ethical committee. The mean
(± S.E.M.) age, height, weight, maximum oxygen uptake (V0o,max)and maximum heart rate (HRmax) of the subjects were29-2 + 2 1 years, 176 3 + 1 8 cm, 75 2 + 4 1 kg, 54 5 +2 0 ml kg-' min' and 190 0 + 2 2 beats min-', respectively.
Preliminary measurements
Following familiarization with treadmill running andexperimental procedures, the subjects undertook two preliminary
* To whom correspondence should be addressed.
4489 243
) by guest on July 13, 2011jp.physoc.orgDownloaded from J Physiol (
O.-K. Tsintzas, C. Williams, L. Boobis and P Greenhaff
tests in order to determine: (i) the relationship between runningspeed and oxygen uptake (Vo2) and (ii) the V02 max using methodsdescribed elsewhere (Williams et al. 1990). One week before thefirst experimental trial, subjects undertook a 60 min treadmill runat 70% V02 max in order to fully familiarize themselves with thedrinking pattern and the measurements used during theexperimental trials.
Experimental designAll subjects completed two 60 min runs at 70% V02 max on amotorized treadmill 7 days apart. On each occasion the subjectsconsumed either an isotonic, 55 % carbohydrate-electrolytesolution (CHO) or water throughout the run. The CHO solutionconsisted of glucose (1-7%), fructose (1-1%), maltose (06%),higher saccharides (1-9%) and electrolytes (sodium, 61 mg(100 ml)-', potassium, 10 mg (100 ml)-'). The order of testing wasrandomly assigned. The subjects were asked to refrain from heavyexercise for 2 days before each trial and record their normal dietfor 3 days preceding the first trial. They were required to replicateexactly the same diet for the same period of time before the secondtrial.
ProtocolOn the day of the experiment each subject arrived in thelaboratory following a 12-14 h overnight fast. The subject wasasked to empty his bladder before nude body weight was obtained.Following this, an indwelling catheter (venflon, 16-18 G,Ohmeda, Hatfield, Herts, UK) was inserted in an ante-cubitalvein while the subject was lying on an examination couch. Thecatheter was kept patent by infusion with sterile heparinizedsaline (10 u ml-'). A resting muscle sample was then obtained fromthe vastus lateralis muscle. The subject was then asked to assumean upright position. Following a 15-20 min period in the standingposition, a 10 ml resting venous blood sample was obtainedimmediately before a warm-up which consisted of a 5 min run at60% VO2, max An expired air sample was collected during the lastminute of the warm-up using the Douglas bag method. Thetreadmill was then stopped for 5 min and the subject consumed8 ml (kg body mass (BM))-' of the assigned fluid while standingon the treadmill. A further 2 ml (kg BM)-' of the assigned fluidwas ingested bv the subject at 20 and 40 min into the run.
Immediately after this 5 min period, the treadmill running speedwas increased to 70% V02, max A 1 min expired gas sample and asubsequent 5 ml venous blood sample were taken every 20 min ofthe run. The Borg scale of perceived exertion (Borg, 1973) and asimple 1-10 scale of perceived thirst were also used at the sameintervals. Following expired air and blood sampling, 2 ml (kgBAI)-' of the assigned fluid was ingested by the subjects. Plasticvolumetric syringes were used by the runners for fluid ingestion toavoid any spillage and ensure that the correct volume wasingested. The last fluid ingestion took place after 40 min ofexercise. Wet sponges were available to the subjects to be used a(dlibitum throughout the runs. Heart rate was recorded continuouslyduring each run by a means of short-range telemetry (PolarElectro sports testers, PE 3000, Polar Electroky, Kempele,Finland). The last expired gas and blood samples were collectedduring the final minute of the run. Immediately after completionof the run a second muscle sample was obtained. The subject's drypost-exercise nude body weight was also obtained.
Dry bulb temperatures wN-ithin the laboratory w!ere 22-9 + 0-4and22-7 + 04 0C during the water and CHO trials, respectively.Relative humidity for the two conditions wvere 50 0 + 1P0 and51-3 + 241%, respectively.
Collection and analysis of expired gas samplesExpired gas samples were collected using a low resistancerespiratory valve which was connected to a 150 1 Douglas bag bylight-weight smooth bore tubing. The percentage of oxygen (02)and carbon dioxide (CO2) in each sample was determined using aparamagnetic oxygen analyser (Sybron-Taylor, model 570A,Servomex Ltd, Crowborough, Sussex, UK) and an infrared carbondioxide analyser (Lira, model 303, Mine Safety Appliance Co. Ltd,Coatbridge, UK), respectively. Expired gas volume andtemperature was determined using a dry gas meter (HarvardApparatus Ltd, Edenbridge, Kent, UK) and a digital thermometer(Edale Instruments Ltd, model C, Toft, Cambridgeshire, UK).Using the temperature of the expired gas, the barometric pressure(GIriffen and George Ltd, Loughborough, UK), and the Haldanetransformation formula (Consolazio, Johnson & Pecora, 1963), allgas volumes were corrected to standard temperature and pressure,dry. This allowed the calculation of VO2 carbon dioxide production
and respiratory exchange ratio (RER).Blood sample collection and analysisEach venous blood sample was collected into a lithium-heparinizedtube except for a 2 ml aliquot which was placed into a non-heparinized tube and left to clot for 1 h. Serum was subsequentlyremoved following centrifugation for 15 min at 6000 r.p.m. andwas stored at -70 0C before being analysed for insulin (Soeldner &Sloane, 1965). Duplicate 20,ul aliquots of venous blood weredeproteinized in 200 #41 of perchloric acid (2 5%) and stored at-20 °C before being analysed for blood lactate concentration usinga modification of the method described by Maughan (1982), and forblood glucose using the glucose oxidase method described byWerner, Rey & Wielinger (1970). Another 40 #s1 (2 x 20 lsl)of venous blood was used for the duplicate determination ofhaemoglobin concentration using a cyanomethaemoglobin method(Boeringher Mannheim, GmbH test combination). Triplicate 20 1ulaliquots of blood were analysed for haeinatocrit values using amicrocentrifuge and haematocrit reader (Hawksley Ltd).Percentage changes in plasma volume were estimated from thehaemoglobin and haematocrit values of blood samples obtained at10 min and during the last min of each run (Dill & Costill, 1974).Plasma, obtained by centrifugation of the remaining blood for15 min at 6000 r.p.m. at temperatures of between 3 and 4 0C, wasstored at -20 0C and was later analysed for free fatty acids (FFA)using a commercially available kit (NEFA-C test, Wtako, Osaka,Japan) and for glycerol (Laurell & Tibbling, 1966).
Muscle sample collection and analysisAll muscle samples were obtained from the vastus lateralis muscleusing the needle biopsy technique (Bergstrom, 1962) with suctionapplied. Each sample was taken through a separate skin incision(3-5 mm long). All incisions were made using a surgical bladeunder local anaesthetic (2-3 ml of 1 % lidocaine (lignocaine)) beforethe start of exercise while the subject was lying on an examinationcouch. The size of the muscle samples ranged between 30 and100 mg wet weight.
After removing the biopsy needle from the leg, all muscle sampleswere quickly removed from the needle and immediately immersedin liquid nitrogen. Samples remained in liquid nitrogen until theywerere freeze dried, after which they were stored at -70 0C. At alater date, one part of the freeze-dried muscle was dissected freeof visible blood and connective tissue, ground up and washed twicewNith petroleum ether to remove fat. Muscle metabolites (glycogen,glucose, glucose 6-phosphate (G-6-P), lactate, ATP and phospho-creatine (PCr)) were extracted and determined enzymaticall) as
J Physiol.489.1244
) by guest on July 13, 2011jp.physoc.orgDownloaded from J Physiol (
Values are means +S.E.M.; n= 7; * P < 0-05 from 10 min; ** P <001 from 10 min.
previously described (Lowry & Passonneau, 1972; Harris,Hultman & Nordesjo, 1974). Glycogen was assayed by hydrolysisin 1 mol F-' hydrochloric acid (HCl) and it was determined both as
acid-soluble and acid-insoluble glycogen (Jansson, 1981). The totalmixed muscle glycogen concentration was calculated by adding theacid-soluble and acid-insoluble glycogen concentrations.
The remaining piece of freeze-dried muscle was also washed twicewith petroleum ether to remove fat, and was used for thedetermination of glycogen in pools of type I and type II musclefibres. Fragments of single muscle fibres were dissected from themuscle piece using low power microscopy (magnification x 10). Theends of the individual fibre fragments were then cut off andstained for myofibrillar ATPase in order to identify type I (slow-twitch) and type II (fast-twitch) fibres using a modification of themethod of Brooke & Kaiser (1970). The remaining part of eachfibre was then weighed on a quartz-fibre fishpole balance (Lowry &Passonneau, 1972) before being stored at -70 'C. At a later date,five to ten fibres of each type were pooled, the total weightranging from 15 to 30,ug. Glycogen was then extracted by adding20 #u1 KOH (1 mM). Following this, samples were agitated on a
vortex mixer and were then warmed to 50 'C for 15 min. Thisresulted in the complete digestion of the fibres. The extract was
then neutralized by adding HCl (0-25 mM) and was analysed forglucose using a modification of the method of Harris et al. (1974).Single fibre analysis was performed on the biopsy samples of sixsubjects.
Statistical analysisTwo-way analysis of variance (ANOVA) for repeated measures on
two factors (experimental treatment and sampling time) was used
to assess overall differences between cardiovascular changes, bloodmetabolites and muscle metabolites for both trials. When a
significant difference was obtained Tukey's post hoc test was usedto locate any differences. Body weight changes, plasma volumechanges and volumes of fluid ingested during the trials were
analysed using Student's t test for paired data. Statisticalsignificance was accepted at a 5% level. Results are presented as
means + S.E.M.
RESULTSPercentage t2, max' RER and CHO oxidation rateThe average percentage V02 max values sustained during thewater and CHO trials were 73-1 + 1.0 and 71P9 + 1P7%)respectively. There were no differences in % 1702 max or RERbetween trials (Table 1). The carbohydrate oxidation rates,as calculated from VO2and RER values, averaged 2-4 + 0 3and 2-5 + 0 3 g min'- during the water and CHO trials,respectively. No difference was found between trials(Table 1).
Fluid ingestion, body weight loss and change in plasmavolumeThe subjects ingested a total of 894 + 50 and 896 + 49 mlof fluid during the water and CHO trials, respectively. Inthe latter trial, this fluid ingestion resulted in theconsumption of 49-3 + 2-7 g of CHO. In both trials,60 min of running induced a body weight loss of
6
Figure 1. Blood glucose concentrations during water andCHO trials
0, CHO; 0, water. Values are means + S.E.M.; n = 7.* P < 0.01 from water.
--5
EE
co004
02
m 300
2 -
As"'.~
40Time (min)
%'2, max
Water
10 min20 min40 min60 min
71P3 + 1P071P7 + 1.073 9 + 1.0*75-4 + 1 0**
CHO
71P4 + 1870 7 + 1P472-1 + 1P773-4 + 2-0
J Physiol. 489.1 245
Rest 20 60
) by guest on July 13, 2011jp.physoc.orgDownloaded from J Physiol (
Values are means + S.E.M. given in mmol glucosyl units (kg DM)-' for glycogen and mmol (kg DM)-' forother metabolites; n = 7. Significantly different from CHO: * P < 0 05; ** P < 0.01.
1-5 + 0-2 kg, a decrease of 2-1 + 0-2%. Body weight losswas corrected for fluid intake. Percentage change of plasmavolume was also similar between the two trials. Reductionsof 7-3 + 1P6 and 6-2 + 1-5% were observed in the waterand CHO trials, respectively.
Heart rate and perceived rate of exertion and thirstCarbohydrate ingestion did not alter the heart rateresponse to exercise. In both trials, however, heart rate washigher at the end of the run when compared with the first10 min (P < 0-01) (Table 1). Perceived rate of exertionranged from 10 to 13 on the Borg scale and there was nodifferences between trials. Similarly, no difference wasfound between the trials in perceived rate of thirst.
Blood metabolitesBlood glucose concentration was higher (P < 0-01)throughout the run during the CHO trial compared withthe water trial (Fig. 1). However, serum insulinconcentration (Fig. 2) was only significantly higher after20 min of exercise (10-5 + 1-6 vs. 5-2 + 0'3 mu F',P < 0-01) which may be attributed to the large variationbetween subjects in the CHO trial. On the other hand, nodifference was observed between trials in blood lactateconcentration (Fig. 3). Carbohydrate ingestion resulted inlower plasma FFA (Fig. 4) and glycerol (Fig. 5)concentrations throughout the run compared with water(P < 001). In contrast, during the water trial there was asteady increase in the concentration of both metabolites.On the other hand, at 20 min of exercise in the CHO trialplasma FFA concentration declined below resting values
(P < 0 05). At the end of the run, however, the plasmaFFA values returned to pre-exercise values.
Mixed muscle metabolitesTable 2 shows that glycogen utilization was 28% lower inthe CHO trial compared with the water trial (108-7 + 16-3vs. 150-9 + 19 9 mmol (kg dry matter (DM))-', respectively;P < 0-01). The average rates of glycogen utilization duringthe CHO and water trials were also different (1f8 ± 0.3 vs.2f5 + 0 3 mmol (kg DM)-' min-', respectively; P < 0f01).Carbohydrate ingestion also resulted in lower muscle G-6-P(1-9 ± 0 3 vs. 2-6 + 0-2 mmol (kg DM)-1) and lactate(9-8 + 0-6 vs. 11 9 + 0-8 mmol (kg DM)-') concentrations atthe end of the run compared with water (Table 2).
Single muscle fibre glycogenMuscle glycogen concentrations in type I and type II fibresare shown in Table 3. Muscle glycogen concentration at restwas higher (P < 0-01) in type II fibres (367-0 + 25-0 in thewater trial and 377-5 + 30-2 mmol (kg DM)-1 in the CHOtrial) when compared with type I fibres (301-9 + 302 inthe water trial and 305 0 + 34 0 mmol (kg DM)-' in theCHO trial). Exercise with water ingestion resulted in agreater (P < 0-01) degradation of muscle glycogen in type Ifibres (66 + 3% of the resting value, P < 0-01) comparedwith type II fibres (20 + 4% of the resting value,P < 0-01). The ingestion of the CHO solution resulted insparing of glycogen in type I fibres (38 + 7% degradationof glycogen as opposed to 66 + 3% during the water trial,P = 0-01). On the other hand, glycogen utilization was18 + 4% in type II fibres and it was not different from the
Table 3. Muscle glycogen in type I and type II fibres at rest and 60 min in water and CHO trials
O.-K. Tsintzas, C. Williams, L. Boobis and P Greenhaff
water trial (20 + 4%). It is worth noting that a goodagreement (on average within 3 %) was observed betweenthe mean glycogen values of type I and type II fibres andthe mixed muscle glycogen values.
Single muscle fibre glycogen utilization rateIn the water trial, the rate of glycogen utilization in type Ifibres averaged 3-3 + 0 3 mmol (kg DM)-' min-' over the60 min run. In type II fibres the corresponding rate wasone-third of what it had been in type I fibres, averaging1 2 + 0 3 mmol (kg DM)-' min-' (P < 0 01). The ingestionof CHO resulted in lower rate of glycogen utilization intype I fibres compared with water (1P9 + 0 4 vs.3-3 + 0 3 mmol (kg DM)-' min-', respectively, P < 0-01).The rates of glycogen utilization in type II fibres were,however, unaffected by CHO ingestion compared with water(1P + 0 2 vs. 1P2 + 0 3 mmol (kg DM)-' min-', respectively).In the CHO trial, the mean rates of glycogen utilization intype I fibres were higher than in type II fibres (I 9 + 0 4 vs.1P1 + 0-2 mmol (kg DM)- min1, respectively), but justfailed to reach statistical significance (P = 0 06).
DISCUSSIONThe main finding of this study was that the ingestion of a5 5% CHO electrolyte solution during a 60 min runresulted in a 28% sparing of glycogen in the vastuslateralis muscle. This reduced muscle glycogen utilizationwas confined mainly to type I fibres where a 42% reductionin glycogen utilization was observed.
The results of the present study are, therefore, inagreement with some previous cycling studies which haveshown that both intravenous infusion of glucose (Bergstrom& Hultman, 1967) and oral ingestion of CHO solutions(Bjorkman et al. 1984; Hargreaves, Costill, Coggan, Fink &Nishibata, 1984; Erickson et al. 1987) decrease the rate ofglycogen utilization during submaximal exercise.
Other cycling studies, however, have failed to demonstratea glycogen sparing effect as a result of CHO ingestion(Coyle et al. 1986; Flynn et al. 1987; Hargreaves & Briggs,1988; Mitchell, Costill, Houmard, Fink, Pascoe & Pearson,1989). In most of these studies, however, the subjects werewell-trained endurance athletes who exhibited pre-exercisemuscle glycogen concentrations that were above normal andmuch higher than those reported for the present study.Endurance training has been shown to decrease bloodglucose oxidation during prolonged exercise (Coggan,Kohrt, Spina, Bier & Holloszy, 1990). Glucose uptakeduring exercise is also inversely related to the prevailingmuscle glycogen concentration (Richter & Galbo, 1986). It ispossible that these factors might explain the absence of theglycogen sparing effect as a result of CHO ingestion.Furthermore, all the above studies used cycling as a modeof exercise. The existing evidence suggests that bloodglucose concentration and the rate of total CHO oxidationgradually decrease during cycling exercise performed
without CHO ingestion, whereas they are maintained athigher levels when CHO is ingested during exercise. Thus,the subjects are able to exercise longer when receiving CHOby simply complementing their total CHO oxidation ratelate in exercise (Coyle et al. 1986). It seems, however, thatduring running without CHO ingestion the blood glucoseconcentration does not decline prior to fatigue to the sameextent that it does during cycling exercise (Williams et al.1990; Tsintzas et al. 1993). Similarly, the rate of CHOoxidation has also been reported not to decrease prior tofatigue (Williams et al. 1990). Moreover, the results fromthe latter study and the present study have shown thatCHO oxidation rates do not seem to be affected by CHOingestion at a rate up to 50 g h-'. It is not surprising,therefore, that CHO ingestion during running would notaffect exercise performance by simply preventing thedecline in blood glucose concentration and the rate of CHOoxidation. Indeed, the results of this study show thatduring running, sparing muscle glycogen could be themechanism by which CHO ingestion exerts its ergogeniceffect.
Assuming that the dry muscle mass of two legs is equal to5-12% of total body weight (Katz, 1986) and all this muscleis involved in contraction during running, it can becalculated that 3-8 kg of dry muscle was engaged inexercise in the present study. Since 42 mmol (kg DM)-' ofmuscle glycogen was spared when CHO was ingestedduring exercise, this means that the total saving ofglycogen was equal to 161 mmol. Given that the 50 g ofCHO ingested provided 280 mmol of CHO, this means thatabout 60% of the CHO ingested was oxidized, therebyconserving muscle glycogen. Presumably the remaining40% was unabsorbed from the gut and/or was distributedto other tissues. Interestingly, almost all the glycogensparing occurred in type I fibres; type II fibres wereunaffected by CHO ingestion.
Since the total CHO oxidation rates were similar with andwithout CHO ingestion, a decreased glycogen utilization intype I fibres would reflect a greater oxidation of bloodglucose by these fibres as a result of CHO ingestion. Thus, itseems that skeletal muscle fibres are capable of increasingblood glucose uptake not only after the initial 2 h ofprolonged exercise (Coyle et al. 1986) or when their muscleglycogen content is low (Gollnick, Pernow, Essen, Jansson& Saltin, 1981), but also within the initial 60 min ofexercise in the presence of high glycogen content. Duringthe early stages of exercise, blood glucose utilization islimited by the inhibition of the hexokinase enzyme byincreased muscle G-6-P concentrations resulting from rapidglycogen breakdown (Katz, Sahlin & Broberg, 1991).Despite this potential inhibitory effect, a recent study ofCHO ingestion during exercise resulted in an increasedglucose uptake after the first 35 min of exercise comparedwith placebo ingestion (McConell, Fabris, Proietto &Hargreaves, 1994), confirming the hypothesis that elevated
J. Physiol.489.1248
) by guest on July 13, 2011jp.physoc.orgDownloaded from J Physiol (
J Physiol.489.1 Carbohydrate ingestion and muscle glycogen 249
blood glucose and plasma insulin concentrations couldsubstantially enhance blood glucose uptake during both theearly and later stages of exercise.
It has been shown in rat skeletal muscle that type I fibrespossess a higher content of insulin-recruitable glucosetransporter protein (GLUT 4) compared with type II fibres(Henrikssen, Bourey, Rodnick, Koranyi, Permutt &Holloszy, 1990). It is not surprising, therefore, that a strongpositive correlation has been observed between bothinsulin-stimulated (Brozinick, Etgen, Yaspelkis & Ivy,1992) and contraction-induced (Henrikssen et al. 1990)glucose uptake and GLUT 4 content in rat muscle. Incontrast, human studies have shown that elevated GLUT 4content is associated with a higher insulin-stimulatedglucose uptake (Dela, Handberg, Mikines, Vinten & Galbo,1993) but not necessarily with increased contraction-induced glucose uptake during exercise (McConell, McCoy,Proietto & Hargreaves, 1994). Although other factors mightalso be involved in the regulation of glucose uptake duringexercise, further research is needed on the relationship ofGLUT 4 content in different fibre types, in particulartype I fibres, of human muscle and glucose uptake duringexercise.
In this study, glycogen utilization was almost three timeshigher in type I than in type II fibres in the water trial. Asimilar finding was reported by Essen (1978) duringcycling exercise of similar duration but lower intensity(50-60% V02 max). In that study, however, the actual rate ofglycogen utilization in type I fibres was higher comparedwith the present study (4-6 vs. 3*3 mmol (kg DM)-' min',respectively). Considering the lower intensity in the studyby Essen (1978), it could be argued that a heavier workloadis probably imposed on the quadriceps muscle duringcycling compared with running exercise. In another study,a much higher rate of glycogen utilization (i.e.8 6 mmol (kg DM)-' min') in type I fibres was found after60 min of cycling exercise at 61 % V02 max (Vollestad, Vaage& Hermansen, 1984). This discrepancy in the resultsbetween the studies by Essen (1978) and Vollestad et al.(1984) could be explained by the differences inmethodology. A direct biochemical analysis on single fibreswas performed by Essen (1978), as in the present study. Onthe other hand, Vollestad et al. (1984) employed ahistochemical method to quantify glycogen content insingle fibres, which is known to be semi-quantitativein nature.
The higher glycogen concentration found at rest in type IIfibres compared with type I fibres is in agreement with aprevious study (Greenhaff, Ren, Soderlund & Hultman,1991). The contribution, however, of type II fibres towardsenergy supply to the working muscles at the exerciseintensity employed in this study (70% V02 max) was small asreflected by the small amount of glyrcogen breakdown inthese fibres. Furthermore, it is possible that thiscontribution was confined within the first 10-15 min of
exercise, a period where glycogen breakdown andconsequently lactate accumulation is very high (Spriet,MacLean, Dyck, Hultman, Cederblad & Graham, 1992).Nevertheless, these results clearly show that one must becareful when drawing conclusions about muscle recruitmentand glycogen utilization from mixed-muscle analysis ofglycogen during studies involving prolonged exercise.
In conclusion, the results of this study show that drinkinga 5-5 % CHO-electrolyte solution before and during a60 min treadmill run at 70% V02 max decreases glycogenutilization compared with water ingestion. Moreover, thissparing of muscle glycogen is restricted to the type I musclefibres. These effects seem to be mediated by the alteredblood glucose and serum insulin concentrations early inexercise.
BERGSTROM, J. (1962). Muscle electrolytes in man. ScandinavianJournal of Clinical and Laboratory Investigation 14, suppl. 68,1-100.
BERGSTROM, J. & HULTMAN, E. (1967). A study of glycogenmetabolism during exercise in man. Scandinavian Journal ofClinical and Laboratory Investigation 19, 218-228.
BJORKMAN, O., SAHLIN, K., HAGENFELDT, L. & WAHREN, J. (1984).Influence of glucose and fructose ingestion on the capacity for long-term exercise in well-trained men. Clinical Physiology 4, 483-494.
BORG, G. A. V. (1973). Perceived exertion: a note on 'history' andmethods. Medicine and Science in Sports and Exercise 5, 90-93.
BROOKE, M. H. & KAISER, K. K. (1970). Muscle fibre types: howmany and what kind? Archives of Neurology 23, 369-379.
BROZINICK, J. T. JR, ETGEN, G. J. JR, YASPELKIS, B. B. III & Ivy,J. L. (1992). Contraction-activated glucose uptake is normal ininsulin-resistant muscle of the obese Zucker rat. Journal of AppliedPhysiology 73, 382-387.
COGGAN, A. R., KOHRT, W. M., SPINA, R. J., BIER, D. M. & HOLLOSZY,J. 0. (1990). Endurance training decreases plasma glucose turnoverand oxidation during moderate intensity exercise in men. Journalof Applied Physiology 68, 990-996.
CONSOLAZIO, C. F., JOHNSON, R. E. & PECORA, L. J. (1963).Physiological Measurements of Metabolic Function in Man, pp. 6-9.McGraw-Hill, New York.
COYLE, E. F., COGGAN, A. R., HEMMERT, M. K. & Ivy, J. L. (1986).Muscle glycogen utilization during prolonged strenuous exercisewhen fed carbohydrate. Journal of Applied Physiology 61,165-172.
DELA, F., HANDBERG, A., MIKINES, K. J., VINTEN, J. & GALBO, H.(1993). GLUT 4 and insulin receptor binding and kinase activity intrained human muscle. Journal of Physiology 469, 615-624.
DILL, D. B. & COSTILL, D. L. (1974). Calculation of percentage changesin volumes of blood, plasma, and red cells in dehydration. Journalof Applied Physiology 37, 247-248.
ERICKSON, M. A., SCHWARTZKOPF, R. J. & MCKENZIE, R. D. (1987).Effects of caffeine, fructose and glucose ingestion on muscleglycogen utilization during exercise. Medicine and Science in Sportsand Exercise 19, 579-583.
ESSEN, B. (1978). Glycogen depletion of different fibre types in humanskeletal muscle during intermittent and continuous exercise. ActaPhysiologica Scandinavica 103, 446-455.
) by guest on July 13, 2011jp.physoc.orgDownloaded from J Physiol (
250 O.-K. Tsintzas, C. Williams, L. Boobis and P Greenhaff J Physiol.489.1
FLYNN, Al. G., COSTILI, D. L., HAWLEY, J. A., FINK, WV. J., NEUFER,P. D., FIELDING, R. A. & SLEEPER, Al. D. (1987). Influence ofselected carbohydrate drinks on cycling performance and glycogenuse. Medicine anid Science itn Sports and Exercise 19, 37-40.
GOLLNICK, P. D., PERNOw, B., ESSEN, B., JANSSON, E. & SALTIN, B.(1981). Availability of glycogen and plasma PFA for substrateutilization in leg mutscle of man during exere ical Physiology1, 27-42.
GREENHAFF, P. L., REN, J. Al., SODERLUND, K. & HULTMAN, E.(1991). Energy metabolism in single human muscle fibres duringcontraction without and with epinephline infusion. A mericanJournial of Physiology 260, E713-718.
HARGREAVES, AM. & BRIGCS, C. A. (1988). Effect of carbohydrateingestion on exercise metabolism. Journal of Applied Physiology65,1533-1555.
HARGCREAVES, AI., COSTILL, D. L., COGGAN, A., FINK, WV. J. &NISHIBATA, I. (1984). Effect of carbohydrate feedings on muscleglycogen utilization and exercise performance. M'Vedicine and Sciencein Sports and Exercise 16, 219-222.
HARRIS, R. C., HULTMAN, E. & NORDESJO, I. 0. (1974). Glycogen,glycogenolytic intermediates and high energy phosphatesdetermined in biopsy samples of musculus quadriceps femoris ofman at rest. Methods and variance of values. Scandinaviani Journialof Clinical and Laboratory Investigation 33, 109-120.
HENRIKSSEN, E. J., BOUREY, R. E., RODNICK, K. J., K,ORANYI, L.,PERMUTT, .i A. & HOLLOSzY, L. 0. (1990). Glucose transportelprotein content and glucose transport capacitv in rat skeletalmuscle. American Journal of Physiology 259, E593-598.
.JANSSON, E. (1981). Acid soluble and insoluble glycogen in lhumanskeletal muscle. Acta Physiologica Scandinaavica 113, 337-340.
KATZ, A. (1986). Regulation of lactic acid production during musclecontraction. PhD Tlhesis, Karolinska Instituite, HudclingeUniversity, Hospital, Sweden.
KATZ, A., SAHLIN, K. & BROBERG, S. (1991). Regulation of glucoseutilization in human skeletal muscle during moderate dy-namicexercise. A merican Journal of Physiology 260, E411-415.
LAURELL, S. & TIBBLING, G. (1966). An enzymatic fluoiometricmicromethod for the determination of glycerol. Clinica ChimiceaActa 13, 312-316.
LOWRY, 0. H. & PASSONNEAU, J. V'. (1972). A F7exible Systemn ofEntzymatic A nialysis. Academic Press, 'New York.
MCCONELL, G., FABRIS, S., PROIETTO, J. & HARGREAVES, Al. (1994).Effect of carbohydrate ingestion on glucose kinetics during exercise.Journial of Applied Physiology 77, 1537-1541.
AICCONELL, G., lICCOY, hi., PROIETTO, J. & HARGREAVES, AI. (1994).Skeletal muscle GIIUT4 and glucose uptake during exercise inhumans. Journal of Applied Physiology 77, 1565-1568.
AIAUG(HAN, R. J. (1982). A simple rapid metlhod for the deteriminationof glucose, lactate, pv-ruv-ate, alanine, 3-hdclroxybutvrate andacetoacetate on a single 20 ,ul blood sample. Clinica Chimica Acta122, 231-240.
MITCHELL, J. B., COSTILL, 1). L., HOUMARI), J. A., FINK, XX. J.,PASCOE, D. D. & PEARS:ON, D. R. (1989). Influence of carbohlydratedosage on exercise performance and glvrcogen metabolism. Journialof Applied Physiology 67, 1843-1849.
RICHTER, E. A. & GALBO, H. (1986). Higlh glycogen levels enhanceglycogen breakdoNwn in isolated contracting skeletal muscle. Joutrnialof Applied Physiology 61, 827-831.
SOELDNER, J. & SLOANE, D. (1965). Critical v-ariables in the radio-immunoassay of seru1m insulin using the double antibody techniqule.Diabetes 14, 771-779.
SPRIET, L. L., MACLEAN, D. A., DYCK, D. J., HULTMAN, E.,CEDERBLAD, G. & GRAHAM, T. E. (1992). Caffeine ingestion andmuscle metabolism during prolonged exercise in humans. AmericanJournal of Physiology 262, E891-898.
TSINTZAS, 0. K., Liu, R., WVILLIAMS, C., GAITANOS, G. C. &CAMPBELL, I. G. (1993). The effect of carbohydrate ingestion duringexercise on a 30-km race. International Journal of Sport Nutrition3, 127-139.
VOLLESTAD, N. K., VAAGE, 0. & HERMANSEN, L. (1984). Muscleglycogen depletion patterns in Type I and subgroups of Type IIfibres during prolonged severe exercise in man. Acta PhysiologicaScandinavica 122, 433-441.
WERNER, WV. H., REY, H. & WIELINGER, H. (1970). Uber dieEigenschaften eines neuen chromogens fur die biutzucker-bestimmung nach GOD/POD methode. Zeitschrift fur AnalyticalC(hemistry 252, 224-228.
WILLIAMS, C., NUTE, A1. G., BROADBANK, L. & VINALL, S. (1990).Influence of fluid intake on endurance running performance.European Journal of Applied Physiology 60, 112-119.
AcknowledgementsThe authors wish to acknowledge the support of this study bySmithKline-Beecham.
Authors' present addressesL. Boobis: Sunderland General Hospital, Sunderland, SR4 7TP,UK.
P. Greenhaff: Department of Physiology and Pharmacology,Nottingham University Medical Scchool, Nottingham NG7 2UH,UK.
Received 7 April 1995; accepted 18 M1ay 1995.
) by guest on July 13, 2011jp.physoc.orgDownloaded from J Physiol (
