Nitric Oxide 23 (2010) 128–135

Contents lists available at ScienceDirect

Nitric Oxide

journal homepage: www.elsevier .com/locate /yniox

Positive correlation between plasma nitrite and performance duringhigh-intensive exercise but not oxidative stress in healthy men

Ulrike Dreißigacker a, Marcel Wendt a, Torge Wittke a, Dimitrios Tsikas b,*, Norbert Maassen a

a Institute of Sport Medicine, Hannover Medical School, Carl-Neuberg-Str. 1, 30625 Hannover, Germanyb Institute of Clinical Pharmacology, Hannover Medical School, Carl-Neuberg-Str. 1, 30625 Hannover, Germany

a r t i c l e i n f o

Article history:Received 16 March 2010Revised 20 April 2010Available online 6 May 2010

Keywords:BiomarkerEndurance exerciseNONitriteNitrateOxidative stress

1089-8603/$ - see front matter � 2010 Elsevier Inc. Adoi:10.1016/j.niox.2010.05.003

* Corresponding author. Fax: +49 511 532 2750.E-mail address: Tsikas.dimitros@mh-hannover.de

a b s t r a c t

Several studies suggest that exercise is associated with elevated oxidative stress which diminishes NObioavailability. The aim of the present study was to investigate a potential link between NO synthesisand bioavailability and oxidative stress in the circulation of subjects performing high-intensive endur-ance exercise. Twenty-two male healthy subjects cycled at 80% of their maximal workload. Cubitalvenous blood was taken before, during and after exercise, and heparinized plasma was generated. Plasmaconcentrations of nitrite and nitrate were quantified by GC–MS and of the oxidative stress biomarker15(S)-8-iso-PGF2a by GC–MS/MS. pH and pCO2 fell and HbO2 increased upon exercise. The duration ofthe 80% phase (d80) was 740 ± 210 s. Subjects cycled at 89.2 ± 3.3% of their peak oxygen uptake. Plasmaconcentration of nitrite (P < 0.01) and 15(S)-8-iso-PGF2a (P < 0.05) decreased significantly during exercise.At the end of exercise, plasma nitrite concentration correlated positively with d80 and performed work(w80) (each P < 0.05). Changes in nitrate concentration also correlated positively with d80 (P < 0.05) andw80/kg (P < 0.01). These findings provide evidence of a favorable effect of nitrite on high-intensive endur-ance exercise. The lack of association between 15(S)-8-iso-PGF2a and NO bioavailability (nitrite concen-tration) and NO biosynthesis (nitrate concentration) suggest that oxidative stress, notably lipidperoxidation, is not linked to the L-arginine/NO pathway in healthy male subjects being on enduranceexercise.

� 2010 Elsevier Inc. All rights reserved.

Introduction

During strenuous exercise there may be a 10- to 20-fold in-crease in molecular oxygen (O2) uptake. It is therefore reasonableto assume that exercise may increase the production of reactiveoxygen species (ROS) which may damage various cell constituents.ROS are produced in intermediate metabolism [1]. Most of O2 isutilized to produce ATP in the mitochondria. During oxidativephosphorylation incomplete reduction of O2 and leak out of theelectron transfer chain in the mitochondria produce differentROS including superoxide (O��2 ), hydrogen peroxide (H2O2) andhighly toxic hydroxyl radicals (�OH) [2]. Irreversible oxidative dam-age of certain vulnerable molecules by ROS is thought to contributeto the degenerative process associated with cell breakdown andaging [3]. Prevention of potential harmful effects of ROS in thehealthy human body is ensured by various antioxidative defensemechanisms including superoxide dismutase, catalase, glutathioneperoxidase and many non-enzymatic antioxidants, notably re-duced glutathione [3].

ll rights reserved.

(D. Tsikas).

Several studies have reported that intense physical activity mayshift the balance between ROS production and ROS inactivation infavor of oxidative stress [3]. ROS can alter cell functions by multi-ple ways, for instance by causing lipid peroxidation of polyunsatu-rated fatty acids in cell membranes [4]. The F2-isorostane 8-iso-prostaglandin F2a (8-iso-PGF2a) belongs to the most frequentlyand best studied biomarkers of oxidative stress [5]. 8-iso-PGF2aand other isoprostanes are generally thought to be specific end-products of non-enzymatic free radical-catalyzed oxidation of ara-chidonic acid esterified to lipids [6]. Many studies, e.g. [1,7], havereported increased levels of lipid peroxidation products in tissuesand plasma during exercise and that exercise volume and intensitymay influence the extent of lipid peroxidation. However, most ofthese studies did not investigate the extent of lipid peroxidationduring exercise but only before the start and after completion ofthe trial. Furthermore, most of the performed exercises were pro-longed but of a low intensity [4,8–10].

ROS produced with elevated formation rates during physicalexercise may decrease the bioavailability of nitric oxide (�NO)[11], one of the most potent endogenous vasodilatators. NO isenzymatically synthesized from L-arginine by different isoformsof NO synthases. In some circumstances, nitrite, the autoxidationproduct of NO, can be reduced back to NO in the circulation [12].

Table 1Demographic and physical characteristics of the study subjects.

Number 22Age (years) 27.8 ± 3.3Height (cm) 185 ± 9.3Weight (kg) 82.9 ± 12.4VO2peak (mL/kg min) 56.8 ± 7.6Mean maximal power (Wlmax/kg) 4.3 ± 0.8Trained group maximal power (Wlmax/kg) 4.9 ± 0.5Untrained group maximal power (Wlmax/kg) 3.7 ± 0.4Exercise duration (s) 740 ± 210Trained group triglycerides (mM) 0.84 ± 0.29Untrained group triglycerides (mM) 1.01 ± 0.48Trained group total glycerol (mM) 0.90 ± 0.30Untrained group total glycerol (mM) 1.13 ± 0.50

Data are shown as mean ± SD.

U. Dreißigacker et al. / Nitric Oxide 23 (2010) 128–135 129

Hence, circulating nitrite could be regarded as a NO reservoir.Within erythrocytes nitrite and NO are oxidized to nitrate [13].NO has been suggested to play a role in adaptation to physicalexercise by modulating blood flow, muscular contraction, glucoseuptake, glycolysis, and cellular respiration. Interestingly, it hasbeen reported that more efficient energy production during exer-cise can be induced by a dietary nitrate/nitrite supplementation[14,15]. It has been assumed that NO is involved in endothelium-mediated vasodilatation and that this is one of the regulatorymechanisms by which substrate supply to working muscles is in-creased, thus allowing prolonged exercise [16]. Because NO ishighly reactive towards oxidants such as superoxide, an associa-tion between NO bioavailability and oxidative stress is reasonableand has been proposed. High-intensive exercise has been reportedto be associated both with elevated oxygen uptake and ROS pro-duction [17,18]. NO reacts very rapidly by superoxide to form per-oxynitrite (ONOO�) [11]. As peroxynitrite most likely does notrelease NO any more but degrades to nitrate and nitrite, the reac-tion of superoxide with NO would diminish favorable effects of NOsuch as increase in muscle perfusion and reduction of oxygen costduring exercise [19].

It is currently discussed that both NO and oxidative stress playmajor roles in adaptation to physical exercise. The aim of the pres-ent study was to examine a potential link between NO synthesis/bioavailability and oxidative stress in healthy young male subjectsduring and after high-intensive exercise on a cycle ergometer. Oneaim of the present study was to monitor potential changes in thestatus of lipid peroxidation. For this, we measured the total plasmaconcentration of the F2-isoprostane 15(S)-8-iso-PGF2a. The secondaim of the study was to examine potential effects of intense phys-ical exercise not only on NO bioavailability but also on NO biosyn-thesis. For this, we measured plasma concentrations of nitrite andnitrate, respectively. We have assumed that measurement of thesebiochemical parameters by thoroughly validated and frequentlyproven GC–MS and GC–MS/MS methods would allow identify apossible link between oxidative stress and NO during a high-inten-sive endurance exercise.

TEST - PROTOCOLPower

0 W10 W

30 %

80 %

Rest 5 7 13 18 23 E1 3 5 10

post-exercise

Fig. 1. Schematic of the endurance exercise test applied in the present study.

Materials and methods

Subjects exercise performance

The study was performed with the local Ethics Committee ap-proval and in accordance with the guidelines of the Helsinki Decla-ration for the Ethical Treatment of Human Subjects.

SubjectsTwenty-two healthy male subjects volunteered for the study

being informed of all risks and gave written informed consent. Sub-jects were trained in different kinds of sports, i.e., handball, longdistance running, triathlon, football, basketball, cycling, swimming,and hockey. Subjects refrained from vigorous activity 2 days beforethe exercise tests and avoided ingesting nicotine. The physicalcharacteristics of the volunteers are summarized in Table 1.

Exercise testingThe subjects were instructed to have a carbohydrate-rich diet the

2 days before both exercise tests. As high dietary fat intake mayincrease oxidative stress [20], we measured plasma triglyceridesand total glycerol before exercise on a HITACHI 150-20 spectropho-tometer. All exercise trials were completed on an electronicallybraked cycle ergometer (LODE Excalibur Sport, Groningen, TheNetherlands). Subjects first performed an incremental exercise testto determine their peak oxygen consumption (VO2peak) and theirmaximal power (Wlmax). Four days later, the subjects completed

an endurance exercise at 80% of their individual Wlmax. In both trials,subjects had to maintain a pedal cadence of 80–90 rpm until subjec-tive exhaustion, defined as an inability to maintain pedal cadenceabove 60 rpm. Subjects were seated at a comfortable seat height thatwas kept constant for both trials. In both tests, oxygen uptake wasrecorded during all test phases (Metalyzer 3b; Cortex, Leipzig,Germany). Analysis was carried out breath-by-breath and themeasured values were averaged over intervals of 30 s for furtherinterpretation. The highest VO2 over a 30-s period during the incre-mental exercise was defined as VO2peak. Heart rate was measuredduring the whole test using a pulse-meter (POLAR Electro Vantage,Kempele, Finland).

Earlobe hyperaemized blood samples (20 lL) for blood lactateconcentration measurement, and blood samples {10 mL, contain-ing 20 lL sodium heparinate (Ratiopharm, Ulm, Germany) as anti-coagulant} from the antecubital vein for plasma parametersanalysis were drawn pre-exercise, after the warm-up period, every5 min during the 80% phase and at the end of exercise (EE), as wellas 1, 3, 5, and 10 min post-exercise in an ease sedentary position onthe ergometer (Fig. 1).

Quantitative analysis of biochemical parameters

Whole blood lactate concentration of capillary blood sampleswas determined using an automated analyzer (BIOSEN S-line,EKF-Diagnostik, Barleben/Magdeburg, Germany). Whole bloodhemoglobin concentration was determined spectrophotometri-cally using an OSM3 (Radiometer Kopenhagen, Denmark). PlasmapH, pO2 and pCO2 were determined using an ABL 505 (RadiometerKopenhagen, Denmark).

0 10 20 30 40 50 600

400

800

1200

1600

2000

Tota

l 15(

S)-

8-is

o-P

GF 2α

in p

lasm

a (p

g/m

L)

Quality control number

0 3 6 9 12 15 18 210

1000

2000

3000

4000

5000

Nitr

ite in

pla

sma

(nM

)

Quality control number

QC3

QC2

QC1

80

100

120

QC2

QC3

QC1sma

(µM

)

B

C

A

130 U. Dreißigacker et al. / Nitric Oxide 23 (2010) 128–135

In each test, venous blood samples withdrawn at different timepoints were gently shaken, put immediately in an ice bath andstored until the exercise was finished. Under these conditions,we did not observe any loss of nitrite in human blood (data notshown). Subsequently, blood samples were centrifuged collectively(10 min, 800g, 4 �C), using a 3–18 k centrifuge (Sigma, Osterode,Germany), plasma was separated from blood cells and stored fro-zen at �80 �C until analysis. Total, i.e., free + esterified 15(S)-8-iso-PGF2a was measured in 1-mL plasma samples by GC–MS/MSafter saponification and immunoaffinity chromatography columnextraction as described elsewhere [21]. Nitrite and nitrate weremeasured simultaneously in 100-lL plasma aliquots by GC–MSas their pentafluorobenzyl derivatives as described by us previ-ously [22]. Study samples were analyzed within several runsalongside quality control (QC) samples as reported elsewhere[21,22]. For nitrite and nitrate each run accompanied with six QCsamples (QC1, QC2, QC3) that were analyzed in duplicate. QC1samples were analyzed without external addition of nitrite and ni-trate. QC2 samples were spiked with 2 lM nitrite and 20 lM ni-trate, whereas QC3 samples were spiked with 4 lM nitrite and40 lM nitrate.

All blood analyzes were corrected for changes in plasma volumeduring exercise by measuring hemoglobin concentrations andhematocrit and applying the method described by Dill and Costill[23].

Statistical analysis

All data are presented as arithmetic mean ± SD. Data on plasmaconcentration changes were analyzed by one-way and two-wayANOVA with repeated measurements. Correlation between param-eters was assessed using linear regression. Differences betweentrained and untrained groups were verified using Student’s t-test.All statistical analyses were performed using SigmaPlot (Version11.0, Systat Software Inc., Richmond, CA). A probability valueP < 0.05 was considered the minimal level of statisticalsignificance.

0 3 6 9 12 15 18 210

20

40

60

Nitr

ate

in p

la

Quality control number

Fig. 2. Quality control profile for (A) total 15(S)-8-iso-PGF2a, (B) nitrite and (C)nitrate in human plasma samples over the whole study period of about two years.Solid and dotted lines indicate the mean values and the ±2 � SD-range. For moredetails see Materials and methods section. Fig. (A) was reconstructed from the dataof Fig. 1 of Ref. [24].

Table 2Plasma total 15(S)-8-iso-PGF2a, nitrite and nitrate concentration at various stages ofthe physical exercise.

15(S)-8-iso-PGF2a Nitrite Nitrate(pg/mL) (nM) (lM)

Rest 87.3 ± 53.3 767 ± 234 32.8 ± 10EE 65.6 ± 19.8 674 ± 204 33.3 ± 10.510 min post-exercise 66.8 ± 21.4 698 ± 262 32.4 ± 11.5DWarm-up versus EE (%) +3.4 ± 32.4 �16.6 ± 16.8* +1.6 ± 8.1P ANOVA P = 0.038 P = 0.005 P = 0.218

Data are shown as mean ± SD.EE, end of exercise.P ANOVA main effect during the whole trial.* P = 0.002.

Results

Quality control

The results of the QC for all analytes are shown in Fig. 2. In theQC plasma samples used for 15(S)-8-iso-PGF2a (n = 57), the concen-tration of total 15(S)-8-iso-PGF2a was determined to be1134 ± 178 pg/mL, i.e., with a precision (relative standard devia-tion, RSD) of 15.7%. Of the 57 QC samples 54 analyses (i.e., 95%)were within the ±2 � SD-range. In the QC plasma samples usedfor the simultaneous analysis of nitrite and nitrate, nitrite wasdetermined with a precision of 22.4% in QC1, 8.7% in QC2, and4.6% in QC3; nitrate was determined with a precision of 5.8% inQC1, 4.9% in QC2, and 4.0% in QC3. The accuracy values for nitriteand nitrate were 85.8% and 94.8% in QC2, and 85.8% and 93.6% inQC3, respectively. These QC data indicate that the nitrite, nitrateand 15(S)-8-iso-PGF2a concentrations measured in the study sam-ples were determined with acceptable precision and accuracy.

Physical characteristics of the subjects exercise testing

Anthropometrical data of the study subjects are summarized inTable 1. Based on the median of the whole group for Wlmax of4.34 W/kg, the subjects were divided into a trained group (T) andan untrained group (UT). The difference in the respective Wlmax

(see Table 2) corresponds to a difference in VOpeak of about 33%. Nosignificant differences were found between trained and untrained

R 6' 5' EE 1' 3' 5' 10'

pH

7.10

7.15

7.20

7.25

7.30

7.35

7.40

30% 80% post-test

R 6' 5' EE 1' 3' 5' 10'

pCO

2 (m

mH

g)

30

35

40

45

50

55

60

65

70

30% 80% post-test

HbO

2 (%

)

30

40

50

60

70

80

90

100

A

B

C

U. Dreißigacker et al. / Nitric Oxide 23 (2010) 128–135 131

subjects with regard to plasma triglycerides (P = 0.38) and plasmatotal glycerol (P = 0.17). Also, there was no correlation betweentraining status and triglycerides (r2 = 0.14) or total glycerol(r2 = 0.15).

Mean relative power during the 80% period of the endurancetest was 3.95 ± 0.42 W/kg (n = 11) for trained and 2.98 ± 0.29 W/kg (n = 11) for untrained subjects. In the 5th minute of the 80%phase, both trained and untrained subjects cycled on average with89% of their VO2peak. The d80 values were not significantly differentin T and UT: d80(T) = 754 ± 173 s versus d80(UT) = 726 ± 250 s.Endurance time did neither correlate to VO2peak nor to the relativepower.

Biochemical parameters

The blood lactate concentration increased significantly from1.0 ± 0.3 mM at rest to 13.7 ± 2.6 mM at EE (P < 0.001; Fig. 3). Fivesubjects were training for and competing in triathlon, cycling orlong distance running. These athletes had almost identical restingplasma 15(S)-8-iso-PGF2a values with those of the non-endur-ance-trained subjects, i.e., 88.3 pg/mL versus 86.8 pg/mL. Duringexercise 15(S)-8-iso-PGF2a plasma concentration decreased signif-icantly (ANOVA main effect P < 0.05; Fig. 3), but there was no sig-nificant change comparing neighbor time points. The time courseof the changes of pH, pCO2 and HbO2 values in venous blood seenduring the exercise are shown in Fig. 4. pH and pCO2 values fellconsiderably, whereas HbO2 increased to about 80%. There wasno significant correlation between 15(S)-8-iso-PGF2a and d80 orw80, accumulated VO2 during the 80% phase, blood lactate concen-tration, and blood pH and pCO2 (data not shown). Moreover, therewas no significant correlation between 15(S)-8-iso-PGF2a plasmaconcentrations and training status at rest as well as at EE.

Nitrite plasma concentration decreased significantly during thetrial (ANOVA main effect P < 0.01) (Table 2). Comparing nitrite con-centrations between neighbor time points, the decrease in nitriteconcentration from the end of the warm-up to the EE phase washighly significant (P < 0.005). Nitrite concentration in plasma atEE positively correlated with d80 (Fig. 5A) and with w80 (eachP < 0.05).

Regarding nitrate plasma concentration, there was no statisti-cally significant change as a result of the high-intensive enduranceexercise during or after the trial compared with resting values intrained and untrained subjects (Table 2). However, the percentagealteration of plasma nitrate from resting values to EE values signif-icantly correlated with d80 (P < 0.05) and equally to w80/kg(P < 0.01) (Fig. 5B).

R 6' 5' EE 1' 3' 5' 10'

Lact

ate

(mM

)

0

2

4

6

8

10

12

14

16

30% 80% post-test

8-is

o-PG

F2α

(pg/

mL)

20

40

60

80

100

lactate8-iso-PGF2α

Fig. 3. Time course of blood lactate and plasma total 15(S)-8-iso-PGF2a concentra-tions during the trial. From rest (R) to exercise end (EE), blood lactate increased(P < 0.001) and 15(S)-8-iso-PGF2a decreased (ANOVA main effect, P = 0.038).

R 6' 5' EE 1' 3' 5' 10' 20

30% 80% post-test

Fig. 4. Time course of venous blood pH (A), pCO2 (B) and HbO2 during the trial.

The resting concentrations of nitrite (r2 = 0.002) and nitrate(r2 = 0.03) did not correlate with Wlmax. Differences in nitrite andnitrate plasma concentrations between trained and untrained sub-jects did not occur during exercise. Moreover, nitrite and nitrateplasma concentrations and their changes were not associated withthe training status (Wlmax) of the subjects. There was no correla-tion between the plasma concentration of nitrite or nitrate andoxygen uptake during the 80% phase, or with blood lactate concen-tration. Also, there was no correlation between plasma nitrite ornitrate and blood pH, pO2 or pCO2 (data not shown).

Our study of exercised healthy young volunteers reveals no cor-relation between NO bioavailability (i.e., plasma nitrite concentra-tion) or NO biosynthesis, i.e., plasma nitrate concentration, and

Plasma nitrite (nM) at EE200 400 600 800 1000 1200

Exer

cise

dur

atio

n (s

)

200

400

600

800

1000

1200

y = 0.52x + 397.56r = 0.49; P <0.05

Plasma nitrate increase (%)-30 -20 -10 0 10 20 30 40

Perf

orm

ed w

ork

(J/k

g)

0

1000

2000

3000

4000

5000

y = 44.54x + 2590.2r = 0.55; P <0.01

A

B

Fig. 5. Relationship (A) between plasma concentrations of nitrite at exercise end(EE) and d80 and (B) between the percentage alteration of plasma nitrate (restingversus EE) and w80/kg (B).

132 U. Dreißigacker et al. / Nitric Oxide 23 (2010) 128–135

oxidative stress, i.e., plasma total 15(S)-8-iso-PGF2a concentration,in the circulation. It is noteworthy that such a correlation was alsomissing in a recent study from our group in smoking and non-smoking young healthy subjects [25].

Discussion

The main aim of the present study was to examine potentialgradual changes in plasma concentrations of 15(S)-8-iso-PGF2a, ni-trite and nitrate during high-intensity exercise. Being establishedmeasures of biomarkers of lipid peroxidation, NO bioavailabilityand NO biosynthesis, respectively, we have assumed that the plas-ma concentration of these parameters would allow to examine apotential association between NO and oxidative stress. Analysesin blood samples were corrected for changes in plasma volumethat occurred during exercise.

15(S)-8-iso-PGF2a and exercise

Especially during exercise the mitochondria of active muscula-ture were identified as the major source of free radicals [26]. Rajg-uru et al. [27] showed enhanced formation of ROS in response to anincreased oxygen uptake. Also, a decrease in pO2 has been pro-posed as a ROS-promoting factor [28]. During exercise, as per-formed in the present study, Medved et al. [18] showed anelongated time to fatigue at 92% VO2peak after infusion of N-acetyl-cysteine which is considered an antioxidant. The concentration of

15(S)-8-iso-PGF2a in the interstitial fluid has been reported to in-crease 3-fold during medium-intensive exercise within 30 min[17]. Appearance of free radicals in the vascular bed was shownby Bailey et al. [7] by electron paramagnetic resonance spectros-copy. In this study free radicals and byproducts of lipid peroxida-tion had been detected early in the exercise, e.g., after 4 min, atexercise intensities comparable to those applied in our study.

The first main finding of the present study is that plasma con-centrations of total 15(S)-8-iso-PGF2a did not only increase duringa high-intensive endurance cycling exercise in trained and un-trained healthy men, but they did rather decrease. In a shortexhaustive treadmill-exercise with intensity and duration compa-rable to those in the present study, Watson et al. [28] have re-ported similar results with regard to 15(S)-8-iso-PGF2a duringand after exercise. These findings could be due to the exercise-in-duced elevation of blood flow which may result in part in a greatermetabolism and excretion of circulating 15(S)-8-iso-PGF2a. It isworth mentioning that Karamouzis et al. [29] found a concentra-tion gradient of 15(S)-8-iso-PGF2a between plasma and muscleinterstitial fluid. Mastaloudis et al. [10] found elevated 15(S)-8-iso-PGF2a levels during and after a 50-km ultra marathon, whereasMargonis et al. [30] showed that the decline of maximal strengthcorrelated with elevated 15(S)-8-iso-PGF2a exclusively after over-training. Possible explanation for differences between studies fromours and other groups could be that changes in plasma concentra-tions of 15(S)-8-iso-PGF2a may primarily occur during longer-last-ing exercises with a large amount of mechanical stress or due tothe use of different analytical methods, i.e. GC–MS/MS in the pres-ent study and commercially available immunoassays by otherswhich are prone to inaccuracy.

Since lactic acidosis has been demonstrated in vitro to be a po-tent pro-oxidant factor, some authors consider proton accumula-tion due to lactic acidosis as an important factor contributing toexercise-induced oxidative stress [31]. Indeed, we measured signif-icantly increased blood lactate levels during and after the exercise,but without any accompanying increase in 15(S)-8-iso-PGF2a plas-ma concentrations. In contrast, we noted even a decrease in 15(S)-8-iso-PGF2a plasma concentrations (Fig. 3). Equally, plasma pH val-ues did not correlate with 15(S)-8-iso-PGF2a plasma concentra-tions. Hence, we can not confirm a correlation between lacticacidosis (in the blood) and oxidative stress in terms of total15(S)-8-iso-PGF2a plasma concentrations. But, we can also not ex-clude that lipid peroxidation measured as plasma 15(S)-8-iso-PGF2a may not be a sensitive biomarker of acute changes in exer-cise-induced oxidative stress.

It should also be mentioned that differences in blood samplingtimes may be associated with diverging results among studies.Thus, Marzatico et al. [3] found increased lipid peroxidationbyproducts after a half-marathon. In that study blood sampleswere taken from 30 min to 48 h after the race, but not during orimmediately after the race. Hence, lipid peroxidation rate mayhave remained constant during exercise [3]. In our study we haveno indication for changes in lipid peroxidation until 10 min post-exercise, but we can not exclude that it could have occurred later.

NO, nitrite, nitrate and exercise

Regarding NO, the main finding of the present study is that ni-trite plasma concentration significantly decreased during high-intensive endurance cycling exercise in trained and untrainedmen. Furthermore, plasma nitrite concentration at EE correlatedpositively with d80 and w80. Circulating nitrite in blood has beensuggested as an ‘‘endocrine” form of NO, as nitrite can be reducedto bioactive NO by enzymatic and non-enzymatic reactions [32].The decrease in the nitrite plasma concentration seen in our studycould be, at least in part, due to exercise-induced enhancement of

U. Dreißigacker et al. / Nitric Oxide 23 (2010) 128–135 133

NO formation from nitrite. Gladwin et al. have postulated that ni-trite is reduced to NO during the rapid hemoglobin deoxygenationfrom artery to vein [32]. These authors observed a positive correla-tion between amounts of NO released from artery to vein and theconcentration of deoxygenated hemoglobin. Hence, nitrite hasbeen suggested as a hypoxic vasodilatator [33]. It is worth men-tioning that nitrite infusion has been reported to cause vasodilata-tion which supports the idea that nitrite may act as NO donorunder certain conditions [12]. Therefore, nitrite-induced vasodila-tation and blood flow increase may occur especially during exer-cise. Such a phenomenon would be concordant with the positivecorrelation we found between plasma nitrite concentration andd80 or w80. In theory, the decrease in plasma nitrite concentrationobserved in our study could also be explained by a distribution ofnitrite between plasma and red blood cells (RBC) in favor of the lat-ter. Vitturi et al. have recently reported that the nitrite transportrate into the RBC increases with the degree of hemoglobin deoxy-genation [34]. This seems to be closely related to deoxyhemoglo-bin-catalyzed reduction of nitrite to NO which is ultimatelydelivered to surrounding smooth muscle cells. Hence, the resultingvasodilatation may increase blood flow according to the local oxy-gen demand.

In contrast to our study, Rassaf et al. found that exercise in-creased plasma nitrite concentration by about 25% in healthyfasting subjects [35]. This remarkable difference is likely to bedue to differences in the study protocols used by us in the pres-ent study and by Rassaf and colleagues in the above mentionedstudy. In particular, it is worth mentioning that our study sub-jects were all male athletes, and their maximal power and exer-cise duration were considerably higher as compared to those inthe study of Rassaf et al. [35]. A possible explanation for this dis-crepancy may be that nitrite consumption is higher in high-intensive physical exercise (e.g., 357 W and 740 s in the presentstudy) than in rather normal-intensive physical exercise (e.g.,181 W and 576 s in [35]).

Many previous studies attempted to understand the formationand the functional importance of NO and its metabolites in hu-mans, especially during exercise. Isolated skeletal muscle cellshave been shown to produce NO at low rates under resting condi-tions and at higher rates during repetitive contraction [36]. NO ap-pears to mediate cell–cell interactions in the muscle, includingvasodilatation and inhibition of leukocyte adhesion [11]. Duringexercise, blood flow in the working muscles increases, accompa-nied by an increase in oxygen consumption. Vasodilatation is oneof the regulatory mechanisms that increase substrate supply tomuscles and allow prolonged exercise [16]. Several factors suchas low oxygen tension, decrease in pH, and increase in pCO2, potas-sium, lactate and osmolality have been shown to contribute tothese mechanisms [37,38]. NO has been suggested to increaseblood flow in limb muscles by causing vasodilatation; enhancedNO release is supposed to be induced by increased shear stresscaused by increased blood flow at the endothelial surface. Thisindicates the existence of a flow-dependent vasodilatation [39]. In-deed, this theory implies that NO production is in part the resultbut not the cause of vasodilatation in exercise.

Besides acting as a vasodilatator, NO is supposed to exert man-ifold further effects such as modulation of muscular contractionand glucose uptake [40], facilitation of contraction-induced trans-location of GLUT4 [41], and activation of glycolysis [42]. Therefore,the lactate increase during the exercise may correlate with the de-crease of plasma nitrite concentration. However, this was not thecase in the present study.

The possibility of in vivo conversion of nitrite and even of nitrateto bioactive NO and the potential favorable effect on exercise hasinitiated considerable work. A further muscular effect was shownby Larsen et al. [19]. This group reported that a reduced oxygen

cost during exercise resulted in a significant increase in muscleefficiency after dietary inorganic nitrate supplementation in hu-mans. This effect occurred without any accompanying increase inlactate concentration. Thus, the authors assumed an increased effi-ciency of energy production which may lead to an improved phys-ical performance [19]. This might be caused by the higher relaxingvelocity of muscle fibers demonstrated by Marechal and Gailly [43]in in situ experiments during administration of a NO donor. Recentstudies by Bailey et al. [14] and Larsen et al. [15] demonstrated thatdietary nitrate supplementation reduces oxygen consumption inexercise. We cannot confirm the relation between increasing ni-trite/nitrate concentration from endogenously produced NO andreduced oxygen consumption in our study under normal dietaryconditions. It is worth mentioning, that, based on a daily formationrate of about 10 lmol NO/kg/day in healthy humans [44], dietarynitrate supplementation in reported studies (e.g., [14,15]) wouldcorrespond to an about 10-fold higher daily NO production rate.

In our study, at the end of exercise, plasma nitrite concentrationcorrelated positively with d80 and w80. The mean nitrate concen-tration remained unchanged. However, in some athletes nitrateconcentration decreased and in others nitrate concentration in-creased. Nevertheless, the percentage change of plasma nitratefrom rest to EE correlated positively with d80, w80, and w80/kg,respectively. Changes in plasma nitrate concentrations may reflectchanges in systemic NO production rates [13]. In some of our ath-letes the decrease in plasma nitrate concentration could be due to asmaller NO production rate which in turn may lead to a smallervasodilatation and thus to a shorter endurance time which is themain factor determining w80.

It is currently discussed that NO may play an important regula-tory role in adaptation to chronic physical training. Wang et al. [45]showed that endothelium-mediated vasodilatation was enhancedin endurance-trained dogs. However, our results are not supportiveof a dependency of elevated systemic NO formation upon the train-ing status, measured by Wlmax.

Oxidative stress and NO in physical exercise

Whether and by which mechanisms physical exercise and oxi-dative stress may influence bioavailability, biosynthesis andmetabolism of NO are poorly understood. It is generally assumedthat oxidative stress decreases NO bioavailability. Gomes et al.[46] suggested that a long-term decrease in oxidative stress maybe one of the mechanisms that are involved in training-inducedelevation of NO biosynthesis. On the other hand and in diametricalcontradiction to this opinion, it is also generally believed that acuteexercise causes oxidative stress which should however decreaseNO bioavailability. Hence, the decrease in NO bioavailabilityshould be smaller in trained compared to untrained subjects. Infact, our study suggests that the biosynthesis of NO, estimated asplasma nitrate concentration as well as the NO bioavailability, esti-mated as nitrite concentration, is closely comparable in trainedand untrained healthy men and does not change significantly uponhigh-intensive exercise.

Should NO be indeed involved in exercise-induced vasodilata-tion we have to assume that NO is additionally formed, most likelyfrom reduction of nitrite. In this process, erythrocytes may play amajor role and involve various proteins and enzymes including dif-ferent hemoglobin forms and carbonic anhydrase. Such a mecha-nism would be in line with an almost constant total plasmaconcentration of nitrite and nitrate but would require an exercise-induced decrease in plasma nitrite concentration. In considerationof the high vasodilatory potency of NO, reduction of only a minor,hardly detectable fraction of circulating nitrite may be required forproduction of additional NO that ensures the increasingly requiredblood flow under high-intensive exercise (Fig. 6). It is worth men-

O=N-O¯

Oxyhemoglobin

NO3¯Urine

NO

O=N-OH

+ H+ −H+Carbonic

Anhydrase

?

Reductases

BiologicalActivity

vasodilationmuscular contraction

glucose uptakeGLUT 4 translocation

glycolysis

Autoxidation

Kidney

L-ArginineNOS

Nutrition

Bacteria

O2•¯

O=NOO¯

INCREASE OF

PHYSICAL EXERCISE

Fig. 6. Proposed mechanism for nitrite-dependent NO formation during high-intensive exercise in healthy young men. Nitrite from various sources, including L-arginine andnutrition, is converted to NO both non-enzymatically and by the catalytical action of enzymes and proteins including carbonic anhydrase and hemoglobin forms. Falls in pHand CO2 concentration in the blood increase the formation rate of nitrite-derived NO. Even small increases in bioactive NO amounts ameliorate NO-dependent biologicalactions which synergistically increase physical performance.

134 U. Dreißigacker et al. / Nitric Oxide 23 (2010) 128–135

tioning that carbonic anhydrases, especially the isoforms presentin erythrocytes, facilitate the conversion of nitrite to NO [47]. Itis interesting that the formation rate of nitrite-derived NO isfavored by low pH values which also occurred in the blood of ourvolunteers.

Conclusion

Our study shows that a high-intensive endurance exercise on acycle ergometer at 89% of the VO2peak does not induce increases inplasma concentrations of 15(S)-8-iso-PGF2a and nitrate, suggestingthat oxidative stress and NO biosynthesis is not elevated duringhigh-intensive exercise in healthy young men. The small but signif-icant decrease in plasma nitrite concentration observed suggeststhat under exercise nitrite is increasingly reduced to NO, mostlikely in erythrocytes. In healthy men, NO additionally producedfrom plasma nitrite enhances relaxation of smooth muscle cells,elevates blood flow, thus taking care for sufficient supply of themuscles with oxygen and substrates.

Acknowledgments

The authors thank H. Konrad and V. Chouchakov for their assis-tance during the exercises. The laboratory assistance by M.-T. Su-chy is gratefully acknowledged. F.M. Gutzki is thanked forperforming GC–MS and GC–MS/MS analyses. Support was givenby the German Federal Institute of Sports Science, Germany(IIA1-070114/0).

References

[1] R. Lovlin, W. Cottle, I. Pyke, M. Kavanagh, A.N. Belcastro, Are indices of freeradical damage related to exercise intensity, Eur. J. Appl. Physiol. Occup.Physiol. 56 (1987) 313–316.

[2] B. Chance, H. Sies, A. Boveris, Hydroperoxide metabolism in mammalianorgans, Physiol. Rev. 59 (1979) 527–605.

[3] F. Marzatico, O. Pansarasa, L. Bertorelli, L. Somenzini, V.G. Della, Blood freeradical antioxidant enzymes and lipid peroxides following long-distance andlactacidemic performances in highly trained aerobic and sprint athletes, J.Sports Med. Phys. Fitness 37 (1997) 235–239.

[4] G.G. Duthie, J.D. Robertson, R.J. Maughan, P.C. Morrice, Blood antioxidantstatus and erythrocyte lipid-peroxidation following distance running, Arch.Biochem. Biophys. 282 (1990) 78–83.

[5] I. Dalle-Donne, R. Rossi, R. Colombo, D. Giustarini, A. Milzani, Biomarkers ofoxidative damage in human disease, Clin. Chem. 52 (2006) 601–623.

[6] J.D. Morrow, K.E. Hill, R.F. Burk, T.M. Nammour, K.F. Badr, L.J. Roberts, A seriesof prostaglandin F2-like compounds are produced in vivo in humans by a non-cyclooxygenase, free radical-catalyzed mechanism, Proc. Natl. Acad. Sci. USA87 (1990) 9383–9387.

[7] D.M. Bailey, I.S. Young, J. McEneny, L. Lawrenson, J. Kim, J. Barden, Regulationof free radical outflow from an isolated muscle bed in exercising humans, Am.J. Physiol. Heart Circ. Physiol. 287 (2004) H1689–H1699.

[8] R.B. Child, D.M. Wilkinson, J.L. Fallowfield, A.E. Donnelly, Elevated serumantioxidant capacity and plasma malondialdehyde concentration in responseto a simulated half-marathon run, Med. Sci. Sports Exer. 30 (1998) 1603–1607.

[9] W.L. Knez, D.G. Jenkins, J.S. Coombes, Oxidative stress in half and full Ironmantriathletes, Med. Sci. Sports Exer. 39 (2007) 283–288.

[10] A. Mastaloudis, S.W. Leonard, M.G. Traber, Oxidative stress in athletes duringextreme endurance exercise, Free Radical Biol. Med. 31 (2001) 911–922.

[11] M.B. Reid, Role of nitric oxide in skeletal muscle: synthesis, distribution andfunctional importance, Acta Physiol. Scand. 162 (1998) 401–409.

[12] K. Cosby, K.S. Partovi, J.H. Crawford, R.P. Patel, C.D. Reiter, S. Martyr, Nitritereduction to nitric oxide by deoxyhemoglobin vasodilates the humancirculation, Nat. Med. 9 (2003) 1498–1505.

[13] K. Chen, R.N. Pittman, A.S. Popel, Nitric oxide in the vasculature: where does itcome from and where does it go? A quantitative perspective, Antioxid. RedoxSignaling 10 (2008) 1185–1198.

[14] S.J. Bailey, P. Winyard, A. Vanhatalo, J.R. Blackwell, F.J. DiMenna, D.P.Wilkerson, J. Tarr, N. Benjamin, A.M. Jones, Dietary nitrate supplementationreduces the O2 cost of low-intensity exercise and enhances tolerance to high-intensity exercise in humans, J. Appl. Physiol. 107 (2009) 1144–1155.

[15] F.J. Larsen, E. Weitzberg, J.O. Lundberg, B. Ekblom, Dietary nitrate reducesmaximal oxygen consumption while maintaining work performance inmaximal exercise, Free Radical Biol. Med. 48 (2010) 342–347.

[16] O. Hudlicka, Regulation of muscle blood flow, Clin. Physiol. 5 (1985) 201–229.[17] I. Karamouzis, K. Christoulas, D. Grekas, K. Giannoulis, E. Vamvakoudis, K.

Mandroukas, The response of muscle interstitial F2-isoprostane (8-ISO-PGF2alpha) during dynamic muscle contractions in humans, ProstaglandinsLeukot. Essent. Fatty Acids 71 (2004) 87–90.

[18] I. Medved, M.J. Brown, A.R. Bjorksten, K.T. Murphy, A.C. Petersen, S. Sostaric, N-acetylcysteine enhances muscle cysteine and glutathione availability andattenuates fatigue during prolonged exercise in endurance-trained individuals,J. Appl. Physiol. 97 (2004) 1477–1485.

[19] F.J. Larsen, E. Weitzberg, J.O. Lundberg, B. Ekblom, Effects of dietary nitrate onoxygen cost during exercise, Acta Physiol. (Oxf.) 191 (2007) 59–66.

[20] E.J. Anderson, M.E. Lustig, K.E. Boyle, T.L. Woodlief, D.A. Kane, C.T. Lin, J.W.Price III, L. Kang, P.S. Rabinovitch, H.H. Szeto, J.A. Houmard, R.N. Cortright, D.H.Wassermann, P.D. Neuffer, Mitochondrial H2O2 emission and cellular redoxstate link excess fat intake to insulin resistance in both rodents and humans, J.Clin. Invest. 119 (2009) 573–581.

[21] D. Tsikas, E. Schwedhelm, M.T. Suchy, J. Niemann, F.M. Gutzki, V.J.Erpenbeck, Divergence in urinary 8-iso-PGF(2alpha) (iPF(2alpha)-III, 15-F(2t)-IsoP) levels from gas chromatography-tandem mass spectrometryquantification after thin-layer chromatography and immunoaffinity columnchromatography reveals heterogeneity of 8-iso-PGF(2alpha). Possible

U. Dreißigacker et al. / Nitric Oxide 23 (2010) 128–135 135

methodological, mechanistic and clinical implications, J. Chromatogr. B 794(2003) 237–255.

[22] D. Tsikas, Simultaneous derivatization and quantification of the nitric oxidemetabolites nitrite and nitrate in biological fluids by gas chromatography/mass spectrometry, Anal. Chem. 72 (2000) 4064–4072.

[23] D.B. Dill, D.L. Costill, Calculation of percentage changes in volumes of blood,plasma, and red cells in dehydration, J. Appl. Physiol. 37 (1974) 247–248.

[24] U. Dreißigacker, M.T. Suchy, N. Maassen, D. Tsikas, Human plasmaconcentrations of malondialdehyde (MDA) and the F2-isoprostane 15(S)-8-iso-PGF2a may be markedly compromised by hemolysis: Evidence by GC-MS/MS and potential analytical and biological ramifications, Clin. Biochem. 43(2010) 159–167.

[25] D. Tsikas, J.M. Hohlfeld, On the importance of the use of proper approaches forcomparison of analytical methods for serum nitrate and for evaluation ofreference concentrations, Clin. Biochem. 42 (2009) 1200–1201.

[26] B. Sjodin, W.Y. Hellsten, F.S. Apple, Biochemical mechanisms for oxygen freeradical formation during exercise, Sports Med. 10 (1990) 236–254.

[27] S.U. Rajguru, G.S. Yeargans, N.W. Seidler, Exercise causes oxidative damage torat skeletal muscle microsomes while increasing cellular sulfhydryls, Life Sci.54 (1994) 149–157.

[28] T.A. Watson, R. Callister, R.D. Taylor, D.W. Sibbritt, L.K. MacDonald-Wicks, M.L.Garg, Antioxidant restriction and oxidative stress in short-duration exhaustiveexercise, Med. Sci. Sports Exer. 37 (2005) 63–71.

[29] I. Karamouzis, D. Grekas, M. Karamouzis, A. Triantos, S. Vasileva-Staikova, D.Kovachev, Enhanced oxidative stress with a gradient between plasma andmuscle interstitial fluid in patients with end stage renal failure onhemodialysis, Hormones (Athens) 7 (2008) 62–69.

[30] K. Margonis, I.G. Fatouros, A.Z. Jamurtas, M.G. Nikolaidis, L. Douroudos, A.Chatzinikolaou, Oxidative stress biomarkers responses to physicalovertraining: Implications for diagnosis, Free Radical Biol. Med. 43 (2007)901–910.

[31] C. Groussard, G. Machefer, F. Rannou, H. Faure, M. Chevanne, S. Vincent, Ashort term supramaximal anaerobic exercise induced changes in blood lipidperoxidation markers and enzymatic and non enzymatic antioxidant system,Free Radical Biol. Med. 33 (2002) S263–S264.

[32] M.T. Gladwin, D.B. Kim-Shapiro, The functional nitrite reductase activity of theheme-globins, Blood 112 (2008) 2636–2647.

[33] B.W. Allen, C.A. Piantadosi, How do red blood cells cause hypoxic vasodilation?The SNO-hemoglobin paradigm, Am. J. Physiol. Heart Circ. Physiol. 291 (2006)H1507–H1512.

[34] D.A. Vitturi, X. Teng, J.C. Toledo, S. Matalon, J.R. Lancaster Jr., R.P. Patel,Regulation of nitrite transport in red blood cells by hemoglobin oxygenfractional saturation, Am. J. Physiol. Heart Circ. Physiol. 296 (2009) H1398–H1407.

[35] T. Rassaf, T. Lauer, C. Heiss, J. Balzer, S. Mangold, T. Leyendecker, J. Rottler, C.Drexhage, C. Meyer, M. Kelm, Nitric oxide synthase-derived plasma nitritepredicts exercise capacity, Br. J. Sports Med. 41 (2007) 669–673.

[36] T.W. Balon, J.L. Nadler, Nitric oxide release is present from incubated skeletalmuscle preparations, J. Appl. Physiol. 77 (1994) 2519–2521.

[37] I. Kjellmer, Potassium ion as a vasodilator during muscular exercise, ActaPhysiol. Scand. 63 (1965) 460–466.

[38] S. Mellander, J. Lundvall, Role of tissue hyperosmolality in exercise hyperemia,Circ. Res. 28 (Suppl. 45) (1971).

[39] H. Drexler, A.M. Zeiher, H. Wollschlager, T. Meinertz, H. Just, T. Bonzel, Flow-dependent coronary-artery dilatation in humans, Circulation 80 (1989) 466–474.

[40] J.S. Stamler, G. Meissner, Physiology of nitric oxide in skeletal muscle, Physiol.Rev. 81 (2001) 209–237.

[41] S.J. Bradley, B.A. Kingwell, G.K. McConell, Nitric oxide synthase inhibitionreduces leg glucose uptake but not blood flow during dynamic exercise inhumans, Diabetes 48 (1999) 1815–1821.

[42] S. Mohr, J.S. Stamler, B. Brune, Posttranslational modification ofglyceraldehyde-3-phosphate dehydrogenase by S-nitrosylation andsubsequent NADH attachment, J. Biol. Chem. 271 (1996) 4209–4214.

[43] G. Marechal, P. Gailly, Effects of nitric oxide on the contraction of skeletalmuscle, Cell Mol. Life Sci. 55 (1999) 1088–1102.

[44] D. Tsikas, F.M. Gutzki, D.O. Stichtenoth, Circulating, excretory nitrite, nitrate asindicators of nitric oxide synthesis in humans: methods of analysis, Eur. J. Clin.Pharmacol. 62 (2006) 51–59.

[45] J. Wang, M.S. Wolin, T.H. Hintze, Chronic exercise enhances endothelium-mediated dilation of epicardial coronary artery in conscious dogs, Circ. Res. 73(1993) 829–838.

[46] V.A. Gomes, A. Casella-Filho, A.C. Chagas, J.E. Tanus-Santos, Enhancedconcentrations of relevant markers of nitric oxide formation after exercisetraining in patients with metabolic syndrome, Nitric Oxide 19 (2008) 345–350.

[47] R. Aamand, T. Dalsgaard, F.B. Jensen, U. Simonsen, A. Roepstorff, A. Fago,Generation of nitric oxide from nitrite by carbonic anhydrase: a possible linkbetween metabolic activity and vasodilation, Am. J. Physiol. Heart Circ. Physiol.297 (2009) H2068–H2074.