v

2

(

t~

SHONA L. HALSON1,2, GRAEME I. LANCASTER1, ASKER E. JEUKENDRUpl, and MICHAEL GLEESON1J Human Peiformance Laboratory, School of Sport and Exercise Sciences, University of Birmingham. Edgbaston, UNITED

KINGDOM,. and 2School of Human Movement Studies, Queensland University of Technology, Kelvin Grove, Queensland,AUSTRAliA

ABSTRACT

HALSON, S. L., G. I. LANCASTER, A. E. JEUKENDRUP, and M. GLEESON. Immunological Responses to Overreaching inCyclists. Med. Sci. Sports Exerc., Vol. 35, No.5, pp. 854-861, 2003. Introduction: Acute bouts of prolonged strenuous exercise areoften associated with immune suppression and an increased risk of infection. However, few studies have examined immunologicalresponses to intensified training that results in overreaching or ovenraining. We investigated the effects of intensified training on plasma

cytokines, glutamine, glutamate, and other related immunological variables in endurance-trained cyclists. Methods: Eight male subjects(age 27.0 :t 3.0 yr, VO2max 58.0 :t 1.7 mL.kg-l.min-I, mass 73.7 :t 2.1 kg) completed 6 wk of training: 2 wk each of normal training(N,7 :t 2 h.wk-I), intensified training (ITP, 14 :!: 5 h.wk-1 and recovery training (R, 3.5 :!: 2.5 h.wk-1. During the study period,

subjects completed six graded cycle ergometer tests to exhaustion (MT), six simulated time trial tests (TT), and eight 2 X IO-minmaximal effon bouts (IT). Subjects also completed questionnaires to assess mood state. Plasma concentrations of tumor necrosisfactor-a (TNF-a) and interleukin-6 (lL-6), salivary IgA, plasma glutamine, glutamate, ammonia, urea, creatine kinase activity, androutine hematological measures were determined once per week. Results: ITP resulted in overreaching in all subjects identified by asignificant decline in performance and disturbances of mood state. Significant increases during the ITP were observed in creatine kinaseactivity and glutamate, whereas the glutamine/glutamate ratio (Gln/Glu ratio), red blood cell numbers (RBC), hemoglobin concentration(Hb), and packed cell volume (PCV) declined after ITP. No significant changes were observed in TNF-a, lL-6, salivary IgA, glutamine,ammonia, urea and various routine hematological measures. Conclusion: Alterations in plasma cytokines do not appear to be relatedto the decline in performance and increased mood state characteristic of overreaching; however, the Gln/Glu ratio may be of use asa marker of overreaching and/or ovenraining. Key Words: CYTOKINES, GLUTAMINE/GLUTAMATE, IMMUNE SYSTEM,

CYCLING, OVERTRAINING

L. O vertraining may .be be~t desc~~ed as an imbalancebetween stress, InvolvIng traInIng as well as non-training sources, and recovery. As many athletes

incorporate high training volumes and limited recovery pe-L nods into their training regimen, they risk the development

of overreaching. Overreaching is defined as an accumula-(l tion of training and/or nontraining stress resulting in a[j short-tenn decrement in perfonnance capacity, in whichrestoration of perfonnance capacity may take from severaln days to several weeks (OR) (9). It is generally believed thatlJ if the imbalance between training and recovery persists thismay result in a long-tenD decrement in perfonnance capac-[ity, in which restoration of perfonnance capacity may takeseveral weeks or months. This condition is tenDed over-

training (OT) (9).

Address for correspondence: Dr. Asker Jeukendrup, School of Sport andExercise Sciences, University of Birmingham, Edgbaston, Bl5 217, Bir-mingham, United Kingdom; E-mail: A.E.Jeukendrup@bham.ac.uk.Submitted for publication July 2002.Accepted

for publication December 2002.

Long-duration, high-intensity exercise has been associ-ated with immunosuppression (18,20), including a reductionin the circulating lymphocyte concentration and suppressionof natural killer cell activity and secretory IgA in mucosalfluid (12,20). Given the high volume of training and limitedrecovery periods often associated with overreaching andovertraining, it has been suggested that immunosuppressionmay occur in overtrained athletes.

Glutamine is a neutral amino acid found in high levels if.a number of human tissues (21) and is the most abundantamino acid in human muscle tissue and plasma (1). Undernormal conditions, glutamine levels are maintained by abalance between the release and utilization of glutamine byvarious organs (21). The brain, lungs, liver, skeletal muscle,and possibly adipose tissue release glutamine, whereas cellsof the immune system, the liver, kidneys, and gastrointes-tinal tract are the primary utilizers (21). The determinationof whether a cell is a net producer or consumer of glutamineis based on the direction of a single reversible reaction(21,28). Glutamine is synthesized from ammonia and glu-tamate by glutamine synthetase. Glutaminase catalyzes thereverse reaction to form ammonia and glutamate from glu-tamine (21).

According to Rowbottom et al. (21), glutamine may bethe most versatile of the amino acids. This is evidenced bythe differing roles that glutamine has in a number of tissues

",

'".1:.,".

854

0195-9131/03/3505-0854MEDICINE & SCIENCE IN SPORTS & EXERCIS~Copyright @ 2003 by the American College of Sports Medicine

DOl: IO.1249/0.1.MSS.OOOOO64964.80040.E9

DId,

FIGURE l-Study design. MT: maximal cycle ergometer test; TT:time trial; IT: intermittent test. Shaded area represents high-intensitytraining sessions. * Days on which resting blood samples were collectedfor all analyses with the exception of salivary IgA; # days on whichsaliva for IgA analysis was collected before and after exercise test

ond, to determine whether the GIn/Glu ratio is altered asresult of intensified training. Finally, measurements of arange of immunological, biochemical, and hematologicalparameters were performed to examine other possible indi-cators of overreaching. It is hypothesized that plasma cyto-kines will remain unchanged after a 2-wk period of inten-sified training and the ratio of glutamine to glutamate isexpected to decline.

METHODS

Approach to the Problem and ExperimentalDesign

To determine whether changes in various immunologicaland hematological indices occur with overreaching, a 6-wktraining period was employed (Fig. 1). During this period,training was manipulated to induce a state of overreachingidentified by a decline in performance and an increase inmood disturbance. Immunological and hematological mea-sures were made before intensified training, after intensifiedtraining, and during a period of recovery to determinewhether changes in these measures occurred alongside thereduction in performance. Resting plasma cytokines wereexamined during normal, intensified and recovery trainingto investigate possible changes in cytokines during over-reaching. Additionally, resting GIn and Glu concentrationswere measured to ascertain their use as a marker of over-reaching. Routine hematological measures were also madeas such measures can be relatively easily obtained and

monitored in athletes.Eight endurance-trained cyclists completed a 6-wk train-

ing protocol described below. Physical characteristics are

855Medicine & Science in Sports & Exercise~

and organs. These include the transfer of nitrogen betweenorgans and detoxification of ammonia, maintenance of acid-base balance during acidosis, as a nitrogen precursor for thesynthesis of nucleotides, a fuel for gut mucosal cells, a fuelfor cells of the immune system, and as a possible directregulator of protein synthesis and degradation (21).

A decline in plasma glutamine concentration as a result ofintensified training suggests that there is either an increaseddemand for glutamine by tissues that require glutamine as afuel and/or a decreased production or altered transport ki-netics of this amino acid (28). A recent hypothesis suggeststhat increased hepatic and gastrointestinal uptake of glu-tamine for gluconeogenesis is occurring at a time whenmuscle release of glutamine remains constant or is decreas-ing (28), thereby explaining the fall in postexercise plasmaglutamine.

As the mechanisms behind the performance decrementsassociated with overtraining are unclear, a combination of anumber of markers is needed for early diagnosis. Changes inthe plasma glutamine/glutamate ratio (Gln/Glu) have re-cently been suggested as a predictor of overreaching orovertraining in athletes (23). Elevated plasma glutamate andhence a reduced GIn/Glu ratio was observed in athletes whowere classified as overtrained (23), and Parry-Billings et al.(19) reported lower glutamine and increased glutamate lev-els in overtrained athletes. However, no studies have inves-tigated changes in glutamine, glutamate, and reported con-current performance measures during a period of intensifiedtraining that has resulted in overreaching.

Currently, there is no unifying hypothesis to adequatelyexplain the mechanisrn/s behind the variety of changes thatare associated with overtraining. Recently in an attempt tointegrate the available information regarding overtraining,Smith (24) proposed the cytokine hypothesis of overtrain-ing. It is suggested that exercise-induced microtrauma to themusculoskeletal system leading to a local inflammatoryresponse is the initiating event in the development of over-training. Inadequate recovery and a continuation of theathletes' training regimen compound this initial local in-flammation leading to chronic inflammation. This results inthe release of inflammatory mediators and the subsequentrelease of pro-inflammatory cytokines from activated mono-cytes, which causes systemic inflammation. This induces"sickness" behavior (fatigue, appetite suppression, depres-sion), activation of the sympathetic nervous system and thehypothalamic-pituitary-adrenal-axis, suppression of the hy-pothalamic-pituitary-gonadal-axis, up-regulation of liverfunction, and possibly immunosuppression (24). Only oneinvestigation has examined interleukin-6 (IL-6) in over-trained athletes (N = 4) and found that levels were withinnormal ranges and unaffected by overtraining (19). How-ever, until now there has been no assessment of the changesin plasma cytokines and performance levels in response tooverreaching, a state considered to be the precursor of theovertraining syndrome.

The aims of the present study were threefold. First, toinvestigate the possibility of altered plasma cytokine con-centrations in response to overreaching and recovery. Sec-

IMMUNOLOGICAL RESPONSES TO OVERREACHING

Subjects completed a 5-min warm-up at 50% W max fol-lowed by two 10-min bouts of maximal exercise. Eachsubject was given 5-min rest between bouts. Subjects wereasked to produce a maximal effort for each of the 10-millbouts, i.e., to produce the maximal amount of work possible.which could be viewed on a computer screen in front of the

subject.

arnrn2-oxcand

PaolIwitharnm

tionsubtlfroniglut~

P1

enz)and

HanalH-2Thi~

neU1p

turn

quoenz:ternana

~

VD2ma.(mL'kg-1,min-1)

TABLE 1. Selected subject characteristics at baseline.-Age Height Body Mass Body Fat

" (yr) (em) (kg) (%):.--

r -27.1 179.7 j 14.6~ 3.0 1.9 1.1

13.72.5

~

58.01.7

MeanSE

outlined in Table 1. The study was approved by the SouthBinningham Local Research Ethics Committee. Before par-ticipation, and after both comprehensive verbal and writtenexplanations of the study, all subjects gave written informedconsent. Each subject completed six incremental maximaloxygen uptake tests to volitional exhaustion (MT), six sim-ulated time trials (TT), and eight intermittent tests (IT).

Questionnaires

Every day for the duration of the study, subjects com-pleted both the Daily Analysis of Life Demands of Athletes(DALDA) (22) and Profile of Mood States Short FormQuestionnaire (POMS-22) (16). The DALDA is divided intcparts A and B, which represent the sources of stress and themanifestation of this stress in the form of symptoms, re-spectively. Subjects were asked to complete these question-naires at the same time of each day before training. Subjectsalso completed the 65-question version of the POMS (16)once a week on the morning of the MT. Global mood statewas determined using the method described by Morgan etal. (17).

Training

Subjects completed 6 wk of training, which consisted oftwo weeks of normal training (N), 2 wk of an intensifiedtraining period (ITP), and 2 wk of recovery (R). Subjectswore a heart rate monitor during all training sessions. Thiswas so the researchers could document training intensityand to ensure that all subjects completed the prescribed.training.

Training during the ITP was based on each indi-vidual's normal training, which was quantified by heart rate'monitoring

(Vantage NY, Polar, Kempele, Finland) duringtraining during N as well as questionnaire assessment ofnormal training volumes. Blood lactate concentrations andheart rate responses from maximal cycle ergometer testswere used to calculate training zones. For additional infor-mation see Halson et al. (5).

anaBri,

pIalwit

SigAft7.2(2~duJize1-2incaswc

peSi1miAJ

Maximal Cycle Ergometer Test (MT)

Subjects attended the laboratory after an overnight fast,and a Teflon catheter (Becton Dickinson, Quickcath) wasinserted into an antecubital vein. After this, the subjectsperfornled an incremental test to exhaustion on an electri-cally braked cycle ergometer (Lode Excalibur Sport, Gro-ningen, The Netherlands) to deternline maximal aerobicpower output (W max)' submaximal and maximal oxygenconsumption, and heart rate. Work rate began at 95 Wandincreased by 35 W every 3 min until volitional exhaustion.Blood samples were collected at the end of each stage andblood lactate concentration was immediately deternlined(YSI 2300 STAT Plus, Yellow Springs, OH).

81Time Trial (TT)

After a 5-min warm-up at 50% W max' subjects performed[a simulated time trial in which they were asked to completea target amount of work as fast as possible. The amount ofwork to be performed was calculated by assuming that[subjects could cycle at 75% of their W max for -60 min at acadence of 80 rpm and thus these time trials lasted approx-imately 60 min for all subjects.

c.

c~

8t

~Intermittent Test (IT)

Unlike the TT, the IT was of a set duration and the changein work produced and mean power output was assessed.

Blood Handling, Storage, and Analysis

Measurements were performed on resting, overnight-fasted samples, which were collected in the morning, onceper week over the 6 wk of the study. Samples were collectedimmediately before the maximal cycle ergometer tests afterinsertion of a Teflon catheter (Becton Dickinson, Quick-cath) into an antecubital vein. Venous blood was collectedinto KJEDTA tubes and centrifuged at 1500 g for 10 min at4°C; plasma was stored at -20°C. All samples were mea-sured in duplicate with the exception of hematological vari-ables. To avoid interassay variation, all samples were ana-lyzed in one batch at the end of the study, with the exceptionof hematological measures, which were performed on theday of collection. The intra-assay coefficient of variation forthe metabolites, creatine kinase, and cytokines measuredwere all less than 5%, with the exception of glutamine andglutamate, which was 7% and salivary IgA which was 10%.

Saliva samples were collected before and after the eighrIT. Subjects were fasted for at least 3 h before testing, andsaliva samples were collected immediately before the fIrstbout and immediately after the second bout. Subjects wereinstructed to swallow, and then unstimulated whole salivawas collected over a 3-rnin period into tubes before exerciseand immediately postexercise. Subjects were instructed toallow saliva to dribble into the collecting tubes unaided byspitting. All saliva collections were made with subjectsseated, leaning forward with their heads down.

Plasma urea. Plasma urea was measured using an en-zymatic colorimetric endpoint method (Kit No. 640-A,

Sigma, Poole, UK).Plasma creatine kinase activity. Plasma creatine ki-

nase (CK) activity was determined at 30°C using an enzy-matic kit (No. 47-10, Sigma, Poole, UK).

Plasma ammonia and glutamine. Plasma glutaminewas analyzed enzymatically by first determining the plasma

ntl

856 I~Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

;~~~~L~'c~:t11

1' 1 "Lto -,:ii)

E~ch ?:~I:;'~were ;]~

.~~~nln ."",,"'""~*"1" ...>Ie, ..",:;/;:

'i;,,f the ';""1

\ n-

L-es

OrIn

C.0le

re-

C~:ate

ammonia concentration based on the reductive amination of

2-oxoglutarate, using glutamate dehydrogenase (EC 1.4.1.3)and reduced nicotinamide adenine dinucleotide (Sigma,Poole, UK). Plasma was then incubated for 60 min at 37°Cwith glutaminase (EC 3.5.1.2), converting free glutamine toammonia and glutamate (10), and the ammonia concentra-tion measured. Plasma glutamine levels were calculated bysubtracting the untreated plasma ammonia concentrationfrom the ammonia concentration in the sample treated with

glutaminase.Plasma glutamate. Plasma glutamate was analyzed

enzymatically using glutamate dehydrogenase (EC 1.4.1.3)and nicotinamide adenine dinucleotide.

Hematology. Venous blood was used for hematologicalanalysis of differential leukocyte counts using a Technicon

H-2 laser system (Bayer Diagnostics, Basingstoke, UK).This included determination of the total leukocyte count andneutrophil, lymphocyte, and monocyte counts.

Plasma cytokines. Plasma concentrations of IL-6 andtumor necrosis factor-a (TNF-a) were determined in ali-quots of plasma with the use of quantitative sandwich-typeenzyme-linked immunosorbent (ELISA) kits (R&D Sys-tems, Abingdon, UK). A high-sensitivity kit was used for

analysis of IL-6.Salivary IgA. After thawing, stored saliva samples were

analyzed for IgA using a sandwich-ELISA method (27).Briefly, flat-bottomed microtitration plates (Linbro EIAplates, Flow Laboratories Inc., McLean, V A) were coatedwith the primary antibody, rabbit antihuman IgA (1-8760,Sigma), at a dilution of 1 in 800 in carbonate buffer, pH 9.6.After washing with phosphate-buffered saline (PBS, pH7.2), the plates were coated with blocking protein solution(2% w/v casein in PBS). Sample analysis was performed induplicate using saliva samples diluted I in 1000 with deion-ized water and a range of standards (Human colostrum IgA,1-2636, Sigma) up to 400 IJ-g.L -I. A reference sample wasincorporated into each microwell plate, and all samples froma single subject were analyzed on a single plate. The plateswere incubated for 60 min at 20°C. After a washing step,peroxidase-conjugated goat antihuman IgA (A-4l65,Sigma) was added and the plate incubated for a further 60min at 20°C. After another washing step, the substrate,ABTS (Boehringer Mannheim, Lewes, UK), was added andafter 30 min the absorbance was measured at 405 nm.~

ce

r:~

Statistical Analysis

Changes in all variables over time was analyzed using arepeated measures analysis of variance, with least signifi-cance difference comparison performed to identify signifi-cant differences between the individual means. The level ofstatistical significance was set at P < 0.05.

training (R, 3.5 ::t 2 howk-I). Perfonnance on MT, TT, andIT all significantly declined after the intensified trainingperiod and subjects demonstrated altered psychologicalstate, with significant changes in global mood state and boththe source and manifestation of stress. Both perfonnanceand mood state returned to baseline or near baseline valuesafter the recovery period. From this infonnation, it wasconcluded that all subjects were overreached upon comple-tion of the intensified training.

At the end of ITP, maximal power output during MTsignificantly declined by 5.4% and maximal work producedduring the IT was also significantly reduced from 181 ::t 10to 166 ::t 12 °kJ. At the same time point, TT time signifi-cantly increased from 59.4 ::t 1.9 min during N to 64.0 ::t2.3 min. For additional infonnation on perfonnance changessee Halson et al. (5).

Questionnaires. All subjects demonstrated alteredmood state with significantly increased scores on thePOMS-65 from 90.4 during N to 116.4 during ITP. Uponcompletion of R, scores returned to 91.5. From this quc'o.tionnaire, the subscales of tension, fatigue and confusionwere also significantly elevated during ITP, whereas vigorsignificantly declined. No changes were evident in the de-pression or anger subscales. Increased total mood distur-bance was also identified by the short version of the POMSquestionnaire, with significantly elevated total scores duringITP. Parts A and B of the DALDA also increased duringITP. The most common changes in sources of stress, asidentified by part A of the DALDA, were related to sporttraining, sleep, and health. The most common alterations inresponses to part B were increased problems associated with

the following areas: need for a rest, recovery, irritability,between session recovery, general weakness, and training

effort.Immunology. Saliva IgA concentration was 121 ::t 14

mg.L -I during N and fell during ITP to 91 ::t 14 mg.L -I,

with some recovery by the end of R (110 ::t 14 mg.L -I)(Table 2); however, these changes were not statisticallysignificant, even when statistical analysis was perfonned onnonnalized data. No significant changes were observed inresting plasma IL-6 or TNF-a concentrations throughout the

duration of the study (Table 2).Biochemistry. Plasma creatine kinase activity was sig-

nificantly elevated during the ITP and returned to baselinelevels during R (Table 2). Plasma urea concentration tendedto be slightly elevated during the ITP (P = 0.057) and also

declined to preintensive training levels after recovery (Table2). However, this increase was not statistically significant.Plasma ammonia also showed a trend for increased levelsduring ITP (P = 0.067) (Table 2).

There were no significant changes in plasma glutamineconcentration over the 6-wk period; however, values de-

clined to 475 ::t 40 .aM after ITP (Fig. 2a). Plasma gluta-mate was significantly elevated during ITP and returned tobaseline levels during R (Fig. 2b). Hence, the GIn/Glu ratiowas significantly lower in the ITP compared with N. Al-though the ratio had not returned to pretraining values after

857Medicine & Science in Sports & Exercise@

RESULTSResponses to training. Subjects completed 2 wk of

normal training (N. 7 ::!: 2 howk-'). 2 wk of intensifiedtraining (ITP, 14 ::!: 5 h-wk-'). and a final 2 wk'ofrecovery

IMMUNOLOGICAL RESPONSES TO OVERREACHING

TABLE 2. Selected immunological and biomechanical variables during normal training (N), intensified training (ITP), and recovery (R).-

N liP R-~_.Week 1 2 4 5 I

Resting salivary IgA (mg.L-1) 121.4:t14.4 112.5:t14.9 109.9:t13.1 113.5:t15.6105.3:t13.2 91.0:t14.1 108.6:t18.2 109.5:t13.9Maximal salivary IgA (mg.L -1) 89.5:t 11.2 100.6:t 9.6 84.5:t 13.5 100:t 16.8 87.2:t 16.2 95.6:t 11.8 111.9:t 11.6 101.2:t 12.7IL-6 (pg'mL -1) 0.5:t 0.2 0.5:t 0.2 0.8:t 0.2 0.7:t 0.2 0.8:t 0.3 0.6:t 0.2TNF-a(pg.mL-1) 7.1:t1.4 8.3:t2.7 8.0:t1.8 7.4:t1.8 6.6:t2.1 6.3:t1.8Glutamine (p.M) 631 :t 21 521 :t 29 555 :t 31 475 :t 40 515 :t 40 555 :t 39Glutamate (p.M) 158:t 18 164 :t 28 200:t 14.' 235:t 18.' 198:t 15' IS7:t 6GIn/Glu ratio 4.38 :t 0.49 3.97:t 0.73 2.86:t 0.24. 2.13 :t 0.26.' 2.76 :t 0.35. 3.61:t 0.37Plasma CK activity (U.L-1) SS.4:t21.458.3:t14.4 80.9:t15.6 92.9:t18.1.' S4.7:t11.S60.6:t6.9Plasma urea (mmol.L -') 2.2 :t 0.2 2.3 :t 0.1 2.6 :t 0.2 2.8 :t 0.2 2.4 :t 0.1 2.4 :t 0.2Plasma ammonia (p.M) 38.6 :t 7.5 6O.2:t 14.4 45.1 :t 15.3 60.8:t 14.4 40.5 :t 14.9 48.1 :t 12.9

-.Significantly different from normal training (N).

.Significantly different from recovery (R).

3 6 p

0.1150.3470.3360.7850.7000.0050.0040.0380.0570.067

r:I:.,

~~

R, there was no statistically significant difference between Rand N (Fig. 2c).

Hematology. During the lTP red blood cell count(RBC), Hb, and packed cell volume (PCV) significantlydeclined and after R had returned to initial levels (Table 3).No changes were observed in mean red blood cell volume(MCV), platelets, white blood cell count, neutrophils,lymphocytes, monocytes, or neutrophiVlymphocyte ratio(Table 3).

700 ,

600

,-.. 500.

I~ 400oS~ 300

l200

100

i 0

~..,

~

300 1

250 i

501

4,6

l3,4 3,4.5

200~,:. 150=G 100

n~

b O+-

~.~ ]3,4,5n

4

4

.l

0 4.5'D=~.2 3.5

~~ r

1..5

sr1.5 ,--

RecoveryNonml Intensified

Training

FIGURE 2-a. Changes in glutamine concentration during normaltraining, intensified training and recovery. b. Changes in glutamat~concentration during normal training, intel.sified training and recov-ery. 1 indicates significantly different from test 1; 2 indicates signifi-cantly different from test 2; 3 indicates significantly different from test3; 4 indicates significantly different from test 4; 5 indicates signifi-cantly different from test 5; 6 indicates significantly different from test6- c. Changes in glutamine/glutamate ratio (GIn/Glu) during normaltraining, intensified training, and recovery. 1 indicates significantlydifferent from test 1; 2 indicates significantly different from test 2; 3indicates significantly different from test 3; 4 indicates significantlydifferent from test 4; 5 indicates significantly different from test 5; 6indicates significantly different from test 6.

~

DISCUSSION

The present investigation does not provide evidence foralterations in cytokines and other immune system parame-ters during overreaching. However, a decline in the Gln/Gluratio was evident, and this ratio may be useful as a diag-nostic tool for overreaching.

Although in the present study there were no statisticallysignificant changes in resting glutamine concentrations, dur-ing the second week of ITP, glutamine concentrations werelower compared with N. This is similar to two previousinvestigations (7,21) that both reported a decline in glu--tamine

concentration in overreached athletes. Glutamineconcentration remained unchanged in swimmers who were

-classified as overreached after 4 wk of intensified training.However, the well-trained athletes had 20% higher concen-r trations of plasma glutamine compared with those who were

l overreached (13). A decline in glutamine may be the resultof greater uptake of this amino acid for gluconeogenesis

r- (28). An increase in gluconeogenesis may be the result of; glycogen depletion due to continued intensified training;

however, alterations in glucose kinetics as a result of over-reaching has not been examined. Although plasma glu-[tamine concentration mayor may not decrease after periodsof intensified training, there is still little evidence to linklow glutamine levels with impaired immune function and[

, increased susceptibility to illness or infection (1). How-ever, the use of glutamine concentrations as a marker toindicate impending or current overtraining warrants fur-[ ther attention.

The mechanismls for the elevation in plasma glutamatewith intensified training are unknown. High plasma glu-tamate concentrations have been reported in catabolic

I conditions such as cancer, human immunodeficiency vi-

858

Official Journal of the American Colleoe of Snnrt" MM;,.,;no

TABLE 3~~~ematological varia~~uring normal training (N), intensified trai~P), and rec~).

R~ N liP2 3 4 5 6 pWeekiI' ..;;;"~

c,'., "."i" 't"'~;'

R8G (x109 L -1) 4.95 ~ 0.06' 4.58 ~ 0.08 4.40 ~ 0.080' 4.27 ~ o.or'Hb (g.dL -1) 14.7 ~ 0.2' 13.7 ~ 0.2 13.1 ~ 0.30' 12.8 ~ 0.20'PGV(L.L-1) 0.44~0.01' 0.41~0.01° 0.39~0.01°' 0.39~0.01°'MGV (fL) 89.7 ~ 0.7 90.1 ~ 0.9 90.0 ~ 0.5 90.8 ~ 0.5Platelets (x109 L -1) 238 ~ 13 232 ~ 10 240 ~ 11 258 ~ 15"W8G(x109L-1) 5.6~0.4 5.1~0.2 5.3~0.4 5.7~0.7Neutrophils (X109 L -1) 2.9 ~ 0.3 2.5 ~ 0.2 2.6 ~ 0.3 3.1 ~ 0.7Lymphocytes (X109 L -1) 2.0 ~ 0.2 2.0 ~ 0.2 1.9 ~ 0.2 1.8 ~ 0.2Monocytes (X109 L -1) 0.4 ~ 0.04 0.4 ~ 0.03 0.5 ~ 0.02 0.5 ~ 0.06Neut:lymph 2.0~0.2 1.3~0.2 1.4:t0.1 2.0:t0.7

" Significantly different from normal training (N).

.Significantly different from recovery (R).

4.57:!: 0.06'13.6 :!: 0.2'0.41 :!: 0.01"91.1 :!: 0.4275:!: 16'5.2:!: 0.42.8:!: 0.31.7:!: 0.20.5:!: 0.041.8:!: 0.3

4.55 :!: 0.08*13.5 :!: 0.3*0.41:!:0.01*89.6 :!: 0.7238 :!: 215.2:!: 0.52.6:!: 0.32.0:!: 0.20.5 :!: 0.031.2:!: 0.1

<0.001<0.001<0.001

0.1200.0050.8130.7900.8070.0540.192

Our findings of unchanged resting plasma IL-6 andTNF-a concentrations during a 2-wk period of intensifiedtraining despite changes in performance. fatigue. and moodstate do not support the role of cytokines in overreaching.According to the cytokine hypothesis of overtraining. theproinflammatory cytokine IL-l.a. in addition to plasma IL-6or TNF-a. is central to the development of overtraining. Inthe present study, we did not measure IL-l.a. and the pos-sibility cannot be excluded that elevated systemic levels ofIL-l.a may have accounted for some of the changes weobserved in our group of overreached subjects. However. asIL-6 and TNF-a were unchanged. it is doubtful that IL-l.awould have demonstrated considerable changes.

The cytokine hypothesis of overtraining proposes thattrauma to the musculoskeletal system leading to a localinflammatory response is the initiating event in the devel-opment of overtraining. Mechanical injury to the musculo-skeletal system due to high impact forces. and in particularinjury to those muscles that contract eccentrically to absorbfoot strike impact forces (2). are the likely source of micro-trauma and local inflammation during running. However.during cycling. muscle contraction is almost entirely con-centric, and mechanical trauma to the musculoskeletal sys-tem from ground impact forces does not occur. Althoughsimilar CK levels have been observed during concentric,eccentric. and isometric arm flexion exercise (3). it is rec-ognized that eccentric muscular contractions result ingreater muscle fiber injury than concentric contractions. It istherefore proposed that ischemia/reperfusion injury may bea possible source of the microtrauma leading to local in-flammation during activities such as cycling which are pre-dominately composed of concentric muscle contractions.

Exercise-induced muscle ischemia/reperfusion injurymay occur as a result of fibers within the contracting muscleexperiencing hypoxia followed by reoxygenation upon thecessation of exercise and the subsequent generation of re-active oxygen species (29). However. the available scien-tific evidence does not support such a view (6.15). and it hasbeen suggested that during prolonged exercise. athletes arehighly unlikely to experience ischemic muscle injury (2).Without an initial source of microtrauma the developmentof local and chronic inflammation will not occur. Further-more. circulating monocytes will not become activated. and

R

rus infection, and sepsis (4). Elevated plasma glutamateconcentration in cancer patients was reported to indicatedecreased uptake of glutamate into the peripheral muscletissue, possibly as a consequence of reduced transportactivity (4). Kinscherf et al. (8) suggested that highplasma glutamate in combination with insufficient base-line glutamine levels may result in catabolism or a loss ofbody cell mass (cachexia) in healthy subjects after veryhigh intensity exercise. These authors suggested that glu-tamate transport activity may be inhibited when there isa high rate of glycolytic activity in skeletal muscle (8).The significance of the elevated glutamate levels is un-known; however, glutamate has been shown to have noeffect on the rate of T -lymphocyte proliferation in vitro(19). Elevated plasma glutamate concentrations may beassociated with overreaching and overtraining; however,the role of glutamate in the mechanisms of overreachingand overtraining is questionable.

Smith and Norris (23) reported unchanged resting plasmaglutamine concentrations in athletes who were classified asovertrained, yet plasma glutamate concentration was signif-icantly elevated in this group. Thus, our observation of anelevated Gln!Glu ratio after intensified training was alsoshown in this previous investigation. The glutamine andglutamate values reported by Smith and Norris (23) aresimilar to those of the present investigation. Our data sup-ports their values suggested to indicate overreaching, i.e., aratio < 3.58. As performance in the present study returnedto baseline after 2 wk of recovery, it was concluded that thesubjects were overreached as opposed to overtrained. Afterrecovery the Gln/Glu ratio returned to above 3.58 and there-fore supports the classification of overreaching based on theGIn/Glu ratio suggested by Smith and Norris (23).

The decrease in RBC, Hb, and PCV during intensifiedtraining most likely reflects plasma volume expansion. Theunchanged resting blood leukocyte counts are consistentwith previous research in this area (14). However, decliningleukocyte counts after 4 wk of training have been foundalongside lower lymphocyte counts in overreached athleteswhen compared with well-trained athletes after 2 wk ofintensified training (14). This may suggest that leukocytecounts reflect the athlete's training status and/or may beassociated with longer term intensified training, i.e., greaterthan 2 wk.

859Medicine & Science in Sports & Exercise@

-

IMMUNOLOGICAL RESPONSES TO OVERREACHING

sible for the fatigue and decreased performance associatedwith overreaching. It was possible to induce a state ofoverreaching, evident by underperformance and changes inmood state, yet resting plasma cytokine concentrations re-mained unchanged. However, it cannot be stated thatchanges in cytokines will not occur in overtrained athletes.Although it is generally assumed that continued trainingwhile in a state of overreaching will lead to overtraining, itcannot be said that the symptoms and characteristics of bothstates are identical. If the athlete continues to train whileoverreached and does not incorporate adequate recoverybetween exercise sessions, this acute inflammatory responsemay develop into a chronic response, ultimately resulting inthe activation of circulating monocytes. Pro-inflammator;cytokines released by these activated monocytes may resultin systemic inflammation, perhaps accounting for some ofthe multitude of symptoms observed in overtrained athletes.Furthermore, it is possible that during running, where thestimulus for the initial microtrauma to the musculoskeletalsystem is greater, a local acute inflammatory response mayoccur. Finally, subjects in the present study were trainedendurance athletes with a moderately high fitness level andit is not known if the results are applicable to athletes ofdiffering fitness levels.

Taken together, the current information regarding theimmune system and overreaching seems only to confirm therole of intensified training in immune depression. Many cellnumbers do not appear to change during overreaching andthose cells that do alter appear to simply reflect the nature ofthe training performed. Thus, immune parameters maychange in response to intensified training independent ofwhether the training results in overreaching. Hence, the rolcof changes in the immune system in the etiology of over-reaching is in doubt. This study supports the classification ofoverreaching based on changes in the GIn/Glu ratio. Alowering of the GIn/Glu ratio in conjunction with a declinein performance and altered mood state may be a useful toolfor the diagnosis of overreaching.

therefore elevated resting systemic pro-inflammatory cyto-kine concentrations would not be expected.

A tendency toward elevated resting plasma CK was ob-served after the first week of intensified training, and CKwas significantly elevated at the end of the second intensi-fied training week. The current study employed cyclingexercise, which excludes eccentric muscle contractions;therefore, the rise in resting plasma CK we observed isunlikely to have been the result of muscle trauma. Onepossibility might be that the cumulative effect of repeatedbouts of prolonged exercise may induce sufficient oxidativestress to impair the body's antioxidant defense systems andperhaps induce membrane peroxidation resulting in theleakage of CK from the muscle into the circulation (26).Results of a previous study (25) do not support this notion;however, the study employed only three consecutive days ofprolonged cycling exercise compared with 14 d in thepresent study. Although statistically significant, the rise inresting plasma CK was quantitatively small and the abilityof this marker to discriminate between normal, intensifiedtraining and intensified training that results in overreachingis doubtful.

Mucosal IgA is an important factor in host defense andhas been examined in relation to increased upper respiratorytract infection incidence and immune depression in endur-ance-trained athletes (II). To date, there is limited data onchanges in mucosal IgA as a result of overreaching withonly Mackinnon et al. (12) reporting 18-32% lower salivaryIgA concentrations in athletes showing symptoms of over-reaching compared with those who were well trained. Wefound lower IgA concentrations during ITP compared withN; however, this was not statistically significant.

The results of the present investigation do not provideclear evidence to either definitively confirm or refute therecently proposed cytokine hypothesis of overtraining. Al-though the underlying causative mechanismls of overreach-ing and overtraining still remain unclear, it does not appearthat elevations in circulating cytokines are primarily respon-

I

REFEREN CES

I. CASTELL, L. M., and E. A. NEWSHOLME. The relation between

glutamine and the immunodepression observed in exercise. AminoAcids 20:49-61, 2001.

2. CHnD, R. Running and tissue damage. In: Improving Spons Perfor-mance in Middle- alw Long-Distance Running: A Scientific Approachto Race Preparation, J. L. Fallowfield and D. M. Wilkinson (Eds.).Chichester, UK: John Wiley & Sons, 1999, pp. 122-139.

3. CLARKSON, P. M., W. C. ByR'-'ES, K. M. MCCORMICK, L. P. TuR-COTI'E, and J. S. WHI1'E. Muscle soreness and serum creatine kinase

activity following isometric, eccentric, and concentric exercise.Int. J. Sports Med. 7:152-155, 1986.

4. HACK, V., O. STtJ1Z, R. KlNSCHERF, et al. Elevated venous glutamatelevels in (pre)catabolic conditions result at least partly from a de-creased glutamate transport activity. J. Mol. Med. 74:337-343,1996.

5. HAI.SON, S. L., M. W. BRIDGE, R. MEEusEN, et al. Time course of

performance changes and fatigue markers during intensified train-ing in trained cyclists. J. Appl. Physiol. 93:947-956, 2002.

6. JACKSON, M. J., A. MCARDLE, and S. O'FARREL. Free Radicals,muscle fatigue and damage. In: Immunophannacology of FreeRadical Species, D. Blake and P. G. Wingard (Eds.). London:Academic Press, 1995, pp. 175-182.

,2:;. :~"

"1""f~:f*~

7. KEAST, D., D. ARSTEIN, W. HARPER, R. W. FRY, and A. R. MORTO,.:

Depression of plasma glutamine concentration after exercise stres"and its possible influence on the immune system. Med. J. Aust.

162:15-18, 1995.8. KINSCHERF, R., V. HACK, T. FISCHBACH, et al. Low plasma glu-

tamine in combination with high glutamate levels indicate risk forloss of body cell mass in healthy individuals: the effect of N-

acetyl-cysteine. J. Mol. Med. 74:393-400, 1996.9. KREIDER, R., A. C. FRY, and M. O'TOOLE. Overtraining in sport:

terms, definitions, and prevalence. In: Overtraining in Sport, R.Kreider, A. C. Fry, and M. O'Toole (Eds.). Champaign, IL:Human Kinetics, 1998, pp. vii-ix.

10. LUND, P. Glutamine: UV-method with glutaminase and glutamate

dehydrogenase. In: Methods of Enzymatic Analysis. Vol. VIII.Metabolites 3: Lipids, Amino Acids and Related Compounds. B.Hu (Ed.). Weinheim: VCF, 1985, pp. 357-363.

11. MACKINNON. L. T. Chronic exercise training effects on immunefunction. Med. Sci. Sports Exerc. 32:5369-5376, 2000.

12. MACKINNON, L. T., and 5. HOOPER. Mucosal (secretory) immune

system responses to exercise of varying intensity and duringovertraining. Int. J. Sports Med. 15:5179-5183, 1994.

860 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

~13.

MACKINNON, L. T., and S. L. HOOPER. Plasma glutamine and upperrespiratory tract infection during intensified training in swimmers.Med. Sci. Sports Exerc. 28:285-290, 1996.

14. MACKINNON, L. T., S. L. HOOPER, S. JONES, R. D. GORDON, andA. W. BACHMANN. Hormonal, immunological, and hematologicalresponses to intensified training in elite swimmers. Med. Sci.Sports Exerc. 29:1637-1645, 1997.

15. MAIR, P., J. MAIR, J. BLEIER, F. WALDENBERGER, H. ANTRE1TER, D.BALOGH, and B. PUSCHENDORF. Reperfusion after cardioplegic car-diac arrest-effects on intracoronary leucocyte elastase release andoxygen free radical mediated lipid peroxidation. Acta Anaesthe-sial. Scand. 39:960-964, 1995.

16. MCNAIR, D. M., M. LORR, and L. F. DROPPLEMAN. Profile of MoodStates Manual, 2nd Ed. San Diego: Educational and IndustrialTesting Services, 1992, p. 40.

17. MORGAN, W. P., D. L. COSffiL, M. G. FLYNN, J. S. RAGLIN, and P. J.O'CONNOR. Mood disturbance following increased training inswimmers. Med. Sci. Sports Exerc. 20:408-414, 1988.

18. NIEMAN, D. C. Is infection risk linked to exercise workload? Med.Sci. Sports Exerc. 32:S406-S41I, 2000.

19. PARRY-BILLINGS, M., R. BUDGm, Y. KOUTEDAKIS, et aI. Plasmaamino acid concentrations in the overtraining syndrome: possibleeffects on the immune system. Med. Sci. Sports Exerc. 24:1353-

1358,1992.20. PEDERSEN, B. K., T. ROHDE, and M. ZACHO. Immunity in athletes.

J. Sports Med. Phys. Fitness 36:236-245, 1996.21. RowBonoM, D. G., D. KEAST, and A. R. MORTON. The emerging

role of glutamine as an indicator of exercise stress and overtrain-ing. Sports Med. 21:80-97, 1996.

22. RUSHALL, B. S. A tool for measuring stress tolerance in eliteathletes. J. Appl. Sport Psychol. 2:51-66, 1990.

23. SMITH, D. J., and S. R. NORRIS. Changes in glutamine and gluta-mate concentrations for tracking training tolerance. Med. Sci.Sports Exerc. 32:684-689, 2000.

24. SMITH, L. L. Cytokine hypothesis of overtraining: a physiologicaladaptation to excessive stress? Med. Sci. Sports Exerc. 32:317-

331,2000.25. VIGUIE, C. A., B. PREI, M. K. SHIGENAGA, B. N. AMES, L. PACKER,

and G. A. BROOKS. Antioxidant status and indexes of oxidativestress during consecutive days of exercise. J. Appl. Physiol. 75:566-572, 1993.

26. VOLEK, J. S., W. J. KRAEMER, M. R. RUBIN, et aI. L-tartrate supple-mentation favorably affects markers of recovery from exercise stress.Am. J. Physiol. Endocrinol. Metab. 282:E474-E482, 2002.

27. WALSH,N. P., A. K. BLANNIN, A. M. CLARK, L. COOK, P. J. ROBSON,and M. GLEESON. The effects of high-intensity intermittent exerciseon saliva IgA, total protein and a-amylase. J. Sports Sci. 17:129-

134,1999.28. WALSH, N. P., A. K. BLANNIN, P. J. ROBSON, and M. GLEESON.

Glutamine, exercise and immune function: links and possiblemechanisms. Sports Med. 26:177-191, 1998.

29. Wm, E. H., A. Z. REZNICK, C. A. VIGUIE, P. STARKE-REED, and L.PACKER. Exercise, oxidative damage and effects of antioxidantmanipulation. J. Nutr. 122:766-:773, 1992.