ORIGINAL ARTICLE

Differential effects on nitric oxide synthase, heat shock proteinsand glutathione in human endothelial cells exposed to heat stressand simulated diving

Lise Fismen • Astrid Hjelde • Asbjørn M. Svardal •

Rune Djurhuus

Received: 6 July 2011 / Accepted: 4 November 2011 / Published online: 24 November 2011

� Springer-Verlag 2011

Abstract Decompression sickness (DCS) may result from

damage to the endothelium caused by the gas bubbles

formed during decompression and may be related to nitric

oxide (NO) production by nitric oxide synthase (NOS).

Heat stress prior to diving has been shown to protect ani-

mals from DCS, and by simulating this treatment in human

endothelial cells (HUVEC) we have shown that a simulated

dive performed subsequent to a heat stress potentiated the

heat-induced expression of HSP70 and increased the level

of the antioxidant glutathione (GSH). Since operational

saturation diving is performed at an increased oxygen level,

HUVEC have been exposed to heat stress and simulated

diving at 40 kPa O2, comparing the response on HSP70,

HSP90 and GSH level to the effects previously observed at

20 kPa O2. In addition, we wanted to investigate the effect

on both endothelial NOS (eNOS) protein and enzymatic

activity. The present results showed that a heat stress (45�C,

1 h) decreased the NOS activity and the protein markedly.

Hyperoxia (40 kPa) alone or a dive either at 20 or 40 kPa

O2,had no effects on NOS activity or protein. At 40 kPa

O2 a simulated dive after heat stress potentiated the

HS-induced HSP70 response, whereas the heat-induced

HSP90 response decreased. GSH levels were found to be

inversely related to NOS activity and protein expression,

and might be explained by a possible post-translational

regulation by glutathionylation of eNOS protein. The

results add to the limited knowledge of these critical factors

in cellular defence mechanisms that can prevent injury

during decompression.

Keywords Endothelial cells � Diving � Decompression �Nitric oxide synthase � Heat shock protein � Glutathione

Introduction

Decompression sickness (DCS) may result from damage to

the endothelium caused by the gas bubbles formed during

decompression (Brubakk et al. 2007). For this reason, the

endothelium is thought to represent an important target for

preventing DCS. The biochemical mechanisms involved

are still unresolved; however, heavy physical exercise 20 h

before a dive has been shown to significantly reduce the

number of vascular bubbles and confer protection against

DCS in man and animals (Dujic et al. 2004; Wisloff et al.

2003). Heat stress (42�C) for 1 h, 24 h prior to a dive

conferred protection from decompression injury in rat and

was accompanied by an increase in the production of heat

shock protein 70 (HSP70; Kampinga et al. 2009) but did

not affect bubble formation (Medby et al. 2008). Huang

and colleagues (Huang et al. 2003) observed that heat stress

prior to a dive reduced air bubble-induced lung injury and

that occurrence of DCS was associated with increased

HSP70 expression in rats. We have recently shown that a

simulated dive at 250 msw performed subsequent to a heat

stress (45�C) had a potentiating effect on the heat-induced

expression of HSP70 in human endothelial cells (HUVEC)

Communicated by Guido Ferretti.

L. Fismen � R. Djurhuus (&)

NUI AS, 5848 Bergen, Norway

e-mail: rdj@nui.no

L. Fismen � A. M. Svardal

Institute of Medicine, University of Bergen,

5021 Bergen, Norway

A. Hjelde

Department of Circulation and Medical Imaging,

Norwegian University of Science and Technology (NTNU),

7491 Trondheim, Norway

123

Eur J Appl Physiol (2012) 112:2717–2725

DOI 10.1007/s00421-011-2241-4

(Djurhuus et al. 2010). Experiments on rats exposed to

rapid decompression indicated an association between DCS

and increased gene expression of small heat shock proteins

in brain and lung, but surprisingly HSP70 was not signif-

icantly affected (Montcalm-Smith et al. 2007).

Exercise is known to be an inducer of endothelial-derived

nitric oxide (NO) in the vascular system (Green et al. 2002;

Kingwell et al. 1997). NO is a potent vasodilator and the

smallest cell signalling molecule known. It has several

physiological functions including regulation of vascular

tone (Lefroy et al. 1993), blood pressure (Rees et al. 1989)

and inflammation (Gross and Wolin 1995). NO dilates the

blood vessels by stimulating soluble guanylyl cyclase,

which elevates cyclic GMP causing smooth muscles to

relax. The NADPH-dependent oxidation from L-arginine to

NO and L-citrulline is catalyzed by nitric oxide synthase

(NOS). NOSs occur in three isoforms: the neuronal (nNOS),

the endothelial (eNOS) and the inducible (iNOS) form.

All the three isoforms require a number of cofactors

including heme protein, the flavins FAD and FMN, NADPH

and tetrahydrobiopterin (BH4) to function normally.

nNOS and eNOS also require calcium and calmodulin

(CaM) (Pollock et al. 1991; Snyder and Bredt 1991).

Endothelial NOS is regulated by and binds directly to heat

shock protein 90 (HSP90; Kampinga et al. 2009), indicating

that HSP90 is involved in NO generation (Fontana et al.

2002; Garcia-Cardena et al. 1998). eNOS synthesis requires

tight regulation at multiple levels, including transcription

and post-translational modifications (Fulton et al. 2001).

eNOS may produce NO or generate the superoxide anion

radical O2-, a process known as NOS uncoupling (Pou et al.

1992). The dominating pathway is dependent upon avail-

ability of its substrate, L-arginine, and its cofactor, BH4

(Vasquez-Vivar et al. 1998; Wever et al. 1997). BH4 has

been shown to be highly susceptible to oxidation, and when

oxidized causes NOS-derived superoxide generation rather

than NO formation. The cellular effects on signalling and

function exerted by O2- are rather different and usually

opposite to those of NO (Cardounel et al. 2005). Inhibiting

NO synthesis resulted in increased bubble formation and

reduced survival in sedentary, but not exercised rats fol-

lowing decompression (Wisloff et al. 2003). Conversely,

administration of a NO-releasing chemical resulted in

reduced bubble formation and increased survival in rats

following an identical dive (Wisloff et al. 2004). These

findings indicate that NO may play a role in bubble for-

mation and endothelial dysfunction.

The cellular antioxidant glutathione (GSH), a major

determinant of intracellular redox state, might be linked to

NOS activity by serving as a necessary reducing cofactor

(Laursen et al. 2001). We have recently shown that heat

stress (45�C, 1 h) prior to a simulated dive at 250 msw

resulted in an approximately 62% increase in the level of

intracellular glutathione in HUVEC (Djurhuus et al. 2010).

In macrophages, glutathione has been shown to be required

for maximal activity by the inducible nitric oxide synthase

(iNOS) (Stuehr et al. 1990). Whereas some investigators

have indicated that glutathione does not seem to have

significant impact on NOS (Huang et al. 2001), others have

indicated a correlation between GSH level and NOS

activity (Laursen et al. 2001). Based on the observations

above, we hypothesize that there could be a link between

NOS, GSH, HSP70, HSP90 and the possible endothelial

injury that may follow decompression.

Endothelial cells line the inside of blood vessels forming

the endothelium, and are thus the first cell layer that comes

in contact with the vascular gas bubbles formed during

decompression. In our model system, isolated HUVEC are

exposed to both heat and simulated diving conditions in a

pressure chamber. Heat stress has in several aspects similar

effects in vitro as physical exercise has in vivo and is a

known inducer of defence mechanisms such as HSPs.

Different stressors, such as hyperthermia, hypoxia, hyper-

oxia and exercise are all known to induce HSPs (Kregel

2002). Exposing bovine aortic endothelial cells (BAECs)

for heat shock (42�C, 1 h) resulted in no change in HSP90

and HSP70 protein content 24 h later (Harris et al. 2003).

However, a more severe heat shock (45�C, 1 h) caused an

eight-fold increase in HSP70, indicating that 42�C treat-

ment was too mild to produce a heat shock response. In the

present study, heat stress at 45�C is used to imitate the heat

stress used in several animal experiments and in vitro

experiments as indicated above and to induce stress

responses similar to those induced by physical exercise in

humans.

A main goal of our research is to study biochemical

mechanisms that may prevent adverse effects of decom-

pression stress in saturation diving. These mechanisms are

most likely activated prior to the decompression stress

itself, as indicated from both the animal and human studies

cited above. Accordingly, we expose our model system to

simulated dives at pressures relevant for saturation diving.

However, decompression of divers in saturation at, for

example, 250 msw would last for approximately 12 days

(Djurhuus et al. 2006), making such a decompression

profile incompatible with a cell model system. Moreover,

an important point is to preserve the biochemical param-

eters of interest, so that they as far as possible reflect the

situation immediately prior to decompression. Conse-

quently, we use a very rapid decompression profile of only

5 min. Major changes in parameters like HSPs and eNOS

are not likely to be expressed during such a short time

implying that the changes observed are not due to the

decompression, but rather to the compression, the pressure

per SE or the composition of the gas during the simulated

dive. Moreover, pilot studies at our laboratory have shown

2718 Eur J Appl Physiol (2012) 112:2717–2725

123

that such rapid decompression had no apparent adverse

effects on the endothelial cells.

In a recent study, we exposed HUVEC to heat stress and

a subsequent simulated dive at 250 msw. The results

showed that the dive potentiated the heat stress induced

expression of HSP70 many times, while a dive alone did

not seem to have any effects on HSP70. In contrast, the

potentiating effect of a dive was not observed for HSP90.

Furthermore, the results showed that neither heat stress nor

dive had any effect on the GSH level, while a dive per-

formed after a heat stress increased the intracellular GSH

level approximately 62% (Djurhuus et al. 2010). In order to

distinguish between the effects of pressure itself and

increased partial pressure of O2, the experiments in that

study were conducted with a ‘‘normal’’ oxygen partial

pressure of 20 kPa. However, in operational saturation

diving the oxygen concentration is considerably higher,

usually from 35 to 80 kPa, depending on the different

phases of a dive. An elevated oxygen level may well-alter

redox status in the cells and affect the generation of NO,

either by altering the functionality of the NOS enzyme or

by changing the availability of essential cofactors like BH4

or molecular chaperones like HSP90.

It is important to investigate how the endothelial

NO-generating system responds to diving conditions at a

more realistic oxygen level. The present study is therefore

carried out by exposing HUVEC to heat stress and simulated

diving at 40 kPa O2, comparing the response on HSP70,

HSP90 and GSH level to the effects previously observed at

20 kPa O2. In addition, we wanted to investigate the effect on

both the eNOS protein and enzymatic activity after simulated

diving at both 20 and 40 kPa O2. The capability of NOS to

function after a dive has to our knowledge not been reported

previously.

Materials and methods

Cells and culture conditions

Human umbilical vein endothelial cells (HUVEC; Ameri-

can Type Culture Collection (ATCC) no CRL-1730) were

obtained from ATCC (Manassas, VA, USA). The cells were

grown without the addition of antibiotics in MCDB-131

medium (Gibco, Invitrogen Ltd., Paisley, UK) supple-

mented with heparin (50 lg/ml), endothelial cell growth

supplement (ECGS, 50 lg/ml; Millipore, Temecula, CA,

USA), and 20% heat-inactivated foetal bovine serum

(Biochrom AG, Berlin, Germany). Cells were seeded on

standard cell culture plastic flasks or dishes, all pre-treated

with a solution of 0.2% gelatine in Dulbecco’s phosphate-

buffered saline (DPBS) for 30 min at room temperature.

Stock cultures of cells were maintained at 37�C in an

atmosphere of 5% CO2 in air and a relative humidity of 95%

(standard conditions). For experiments, the cells were

exposed in pressure chambers as indicated below.

Exposure of cells to heat and simulated dive

Unless otherwise indicated, the experiments consisted of

five groups as indicated on Fig. 1. Cells were exposed to

heat and simulated dive in two identical, 15-l stainless steel

pressure chambers as described previously (Djurhuus et al.

2010). In brief, for heat stress the cell dishes were placed in

a pressure chamber that was flushed with helium to remove

air. The chamber was then flushed with He/O2/CO2 mixture

(75/20/5; mixed from 6.0, 5.5 and 5.0 qualities, respec-

tively, Yara Industrial AS, Bergen, Norway), isolated at

atmospheric pressure and 100% relative humidity at a

temperature of 45�C for 1 h. After exposure, the cell dishes

were returned to standard conditions in the incubator. For

the initial experiments with 20 kPa O2, the cell dishes were

left at the standard conditions for 48 h until harvesting. For

the experiments conducted with 40 kPa O2, the cell dishes

exposed to heat stress were subsequently left at standard

conditions for 24 h and then placed in a pressure chamber

that was flushed with a He/O2/CO2 mixture (55/40/5; mixed

from 6.0; 5.5 and 5.0 qualities, respectively, Yara Industrial

AS, Bergen, Norway), isolated at atmospheric pressure and

100% relative humidity at a temperature of 37�C for 24 h

until harvesting. For the simulated dive, the cell dishes were

placed in the other pressure chamber that was flushed with

He/O2/CO2 mixture (either (75/20/5 or 55/40/5). The

chamber was then pressurized with helium (6.0 quality,

Gardner Cryogenics A/S, Sandnes, Norway) to an absolute

pressure of 2.6 MPa and isolated with 100% relative

humidity at 37�C for 24 h. Following a rapid decompres-

sion (10 kPa/s), the cells were harvested immediately for

analysis as described below. An outline of the experimental

design is shown in Fig. 1.

Since the activity of NOS had not been determined at

similar conditions previously, we first exposed the cells to

simulated dive at 20 kPa O2, and accordingly group 2

(Fig. 1) was omitted.

Determination of NOS activity

Harvesting of cells

HUVEC were grown until confluence on 10 cm diameter

culture dishes. The medium was aspirated off and the cells

washed with pre-heated (37�C) PBS. The cells were

detached by trypsination (0.25% trypsin–0.5 mM EDTA)

for 5 min at room temperature (RT) before trypsin was

inactivated by the addition of basal Eagle’s medium

containing 10% foetal calf serum. Detached cells were

Eur J Appl Physiol (2012) 112:2717–2725 2719

123

centrifuged at 120g for 10 min before resuspended in cold

PBS and centrifuged again. PBS was carefully aspirated off

and the pellets snapfrozen in a mixture of dry ice and ethanol

before stored at -80�C.

Enzyme activity

The release of NO generated by NOS was measured by

monitoring the conversion of radio-labelled L-[14C]-argi-

nine to L-[14C]-citrulline following the procedures devel-

oped by Knowles and Salter (1998) and Weissman and

Gross (2001) with some modifications. The formation of

L-citrulline is stoichiometric with the synthesis of NO, hence

the production of L-citrulline was assumed to correspond to

NO synthesis. In brief, pellets containing approximately

3–4 9 106 cells were resuspended in 120 ll ice-cold

20 mM HEPES buffer pH 7.4 containing 1 mM EDTA,

1 mM DTT, 0.25% Brij35 (Sigma-Aldrich Corp., St.

Louis, MO, USA), 10 mM ascorbic acid, 25 lM BH4,

80 lM FeCl2 and a cocktail of protease inhibitors (Com-

plete Mini, Roche Diagnostics GmbH, Mannheim, Ger-

many). Cells were left to lyse on ice for about 30 min

before 20 ll aliquots of cell suspension were placed in

tubes and mixed together with 55 ll assay buffer, 20 ll

L-[14C]-arginine assay buffer and 5 ll lysis buffer to make up

a total reaction volume of 100 ll. The cell suspension was

not centrifuged, because the majority of NOS in endothelial

cells is membrane associated, and so each sample was

drawn immediately after vortexing. The reaction mixture

contained (final concentrations) 45 mM Tris-HEPES, pH

7.4, 1.25 mM DTT, 0.37 mM CaCl2, 1 mM NADPH,

56 lM BH4, 1 lM FAD, 1 lM FMN, 60 mM L-valine,

0.02 lg/ll calmodulin (bovine brain, Calbiochem, Merck

KGaA, Darmstadt, Germany) and 100 lM L-arginine

(0.1 lCi L-[14C (U)]-arginine; PerkinElmer, Inc., Boston,

MA, USA).

After incubation at 37�C for 30 min, the reactions were

terminated by the addition of 1 ml ice-cold stop buffer

(50 mM HEPES, pH 5.5; 5 mM EDTA). L-citrulline pro-

duced was separated from the L-arginine by cation-exchange

chromatography using 2 ml 50% slush of Dowex resin

(50WX8, 100–200 mesh, ion-exchange, Acros organics,

New Jersey, USA) converted to the Na? form and pre-

equilibrated in stop buffer. 3.4 ml of water were added and

each sample mixed, after 30 min the tubes were spun lightly

(30g for 2 min) in a centrifuge. 4.5 ml (82% of total sample)

of L-[14C]-citrulline supernatant was aspirated off, 5 ml

scintillation fluid was added and the samples counted in a

liquid scintillation counter. Background values obtained

from complete samples, but with cell extracts replaced by

lysis buffer were subtracted from all results.

Protein was determined by the assay of Lowry et al.

(1951) using the Bio-Rad DC protein assay kit and bovine

IgG as a standard. Each experiment consisted of three

separate dishes per group and samples in triplicate were

drawn from each dish for NOS measurements. The NOS

activity was expressed as pmol/min/mg protein and pre-

sented as activity relative to control.

Determination of intracellular HSP70, HSP90

and eNOS

Harvesting and determination of viable cells

HUVEC were grown until confluence on 6 cm culture

dishes. Cells were trypsinated as described previously and

Exposure

group 0 24 48

1: Control Harvest

2: Control 40 kPa O2 Harvest

3: HS HS 45˚C, 1 h Harvest

20 kPa O2

4: Dive Harvest

5: HS + Dive HS 45˚C, 1 h Harvest

20 kPa O2

20 or 40 kPa O2, 100 kPa 24 h

Dive 24 h, 2.6 MPa (250 msw), 20 or 40 kPa O2

Hours

Std. conditions, 37 ˚C, 100 kPa, 24 h

Std. conditions, 37 ˚C, 100 kPa (1 bar)

Std. conditions, 37 ˚C, 100 kPa Dive 24 h, 2.6 MPa (250 msw), 20 or 40 kPa O2

Std. conditions, 37 ˚C, 100 kPa 40 kPa O2, 100 kPa 24 h

Std. conditions, 37 ˚C, 100 kPa, 24 h

Fig. 1 Exposure design. Overview of exposure to heat stress (HS,

45�C, 1 h) and subsequent simulated diving at 37�C. Standard

conditions were in an incubator with 5% CO2 in air, 100% humidity

and 37�C. Where indicated a second control included exposure to

40 kPa (0.4 bar) O2 and 5 kPa (0.05 bar) CO2 in He at atmospheric

pressure in parallel with the dive exposure. Heat stress was performed

in an atmosphere of 20 kPa O2 and 5 kPa CO2 in He at atmospheric

pressure, and simulated diving were in an atmosphere of 20 kPa

(0.20 bar) or 40 kPa O2 and 5 kPa CO2 in He

2720 Eur J Appl Physiol (2012) 112:2717–2725

123

counted in an electronic cell counter (CASY model TT,

Innovatis AG, Reutlingen, Germany). The cells were cen-

trifuged at 120g for 10 min and washed by the addition of

1.5 ml DPBS at RT, gently resuspended, transferred to

microtubes and spun in a microcentrifuge at 300g for

10 min. The cell pellets were then snapfrozen and stored at

-20�C for later analysis.

Analysis of HSP70, HSP90 and eNOS

Cell pellets were lysed in the lysis buffer according to man-

ufacturer’s instructions. HSP70 (EKS-700B) and HSP90a

(EKS-895) ELISA kits were obtained from Stressgen Bior-

eagents (Victoria, Canada). Human eNOS Quantikine ELISA

kit was from R&D Systems Europe Ltd, Abingdon, England.

All ELISA kits were quantitative sandwich immunoassays

containing a monoclonal antibody against the actual protein

as the primary antibody and determine a coloured product

formed by a horseradish peroxidase conjugated to a second-

ary antibody. Colour formation was measured in a microplate

reader at 450 nm, and HSP or eNOS were quantified from a

standard curve generated with standard protein provided in

the kit. Concentrations were calculated in nanograms per 106

cells and presented as concentration relative to the control

group.

Determination of intracellular glutathione

HUVEC were grown until confluence on 6 cm culture

dishes. The analysis of glutathione was performed as

described previously (Djurhuus et al. 1991). In brief,

medium was removed, and the culture dishes were imme-

diately placed on ice and gently washed twice with 5 ml

ice-cold DPBS. The cells were extracted with 300 ll 5%

ice-cold sulphosalicylic acid containing 50 lM DTE using

a rubber policeman to scrape the cells off the dish, trans-

ferred to a microtube and frozen at -20�C. For analysis,

the samples were thawed, centrifuged and the supernatants

were used for determination of reduced and total glutathi-

one as described below.

Total free glutathione which represents the sum of

GSH, GSSG and soluble mixed disulphides (GSSR), and

reduced glutathione (GSH) were determined in the acid

extracts according to a modification (Mansoor et al. 1992)

of a chromatographic procedure described previously

(Svardal et al. 1990). Each experiment consisted of three

separate dishes per group and two parallel samples were

drawn from each dish for both reduced and total gluta-

thione measurements. Another three dishes in each group

were used for determination of viable cell number.

The results were expressed as glutathione equivalents in

nmol/106 cells, and presented as concentration relative to

control.

Statistical analysis

All results are calculated as the average ± SD of all indi-

vidual dishes seeded from the same stock culture and

undergoing the same exposure. The results were evaluated

by the analysis of variance (ANOVA) with the Tukey–

Kramer procedure for multiple comparisons. Effects were

considered significant when p \ 0.05.

Results

NOS enzyme activity and eNOS protein

Pilot experiments showed that the rate of NO production in

HUVEC was linear with regard to time for 30 min (data

not shown). The typical NOS activity for control cells

incubated at standard conditions was approximately

85 pmol/min/mg protein. A heat stress (45�C, 1 h) alone

decreased NOS activity 48%, while HS and a subsequent

exposure to a simulated dive decreased NOS activity 46%,

both compared to control at 20 kPa O2 (p \ 0.001, Fig. 2).

A dive alone at 20 kPa O2 had no significant effects on the

enzymatic activity. Increasing pO2 to 40 kPa had a slight

but not significant increasing effect on the NOS activity

(Fig. 3). Exposure to HS alone decreased the activity 54%,

while HS and a subsequent exposure to a simulated dive at

40 kPa O2 decreased NOS activity 51% (p \ 0.001 vs.

control 40 kPa O2). Exposing cells to 40 kPa O2 during the

dive did not affect enzyme activity. The results of NOS

activity at 20 and 40 kPa O2 showed similar patterns,

indicating that the elevated level of O2 during simulated

dive did not influence the enzyme’s capability for NO

production.

Determination of eNOS protein demonstrated a rather

similar pattern to the NOS activity. Increasing pO2 to

40 kPa had a slight but not significantly increasing effect

on the eNOS protein level (Fig. 3). HS decreased the

amount of eNOS protein 41% (p \ 0.01 vs. control 40 kPa

O2) (Fig. 3). Similarly, HS prior to simulated dive

decreased eNOS protein 52% (p \ 0.001 vs. control

40 kPa O2). No significant difference was observed

between eNOS protein in the control (40 kPa O2) and the

dive group.

Expression of HSP70 and HSP90

The results shown in Fig. 4 demonstrated that heat stress

increased HSP70 many fold (p \ 0.01 vs. control 40 kPa

O2) and that a dive performed after heat stress had a

potentiating effect on the HSP70 expression increasing its

concentration on average 79 times compared to control

at 40 kPa O2 (p \ 0.001; data from two independent

Eur J Appl Physiol (2012) 112:2717–2725 2721

123

experiments) and 4.8 times compared to heat stress alone

(p \ 0.001). A slight but non-significant increasing effect

of the dive itself was also observed.

The HSP90 level (Fig. 5) showed a slight but non-sig-

nificant reduction after exposure to 40 kPa O2. Heat stress

increased HSP90 60% compared to control at 40 kPa O2

(p \ 0.001). A slight increase in HSP90 was also observed

in cells dived at 40 kPa O2 (p \ 0.05 vs. control 40 kPa

O2). In contrast to HSP70, a subsequent simulated dive had

a decreasing effect on the heat-induced HSP90 expression

(p \ 0.001).

Glutathione levels

A simulated dive had no effect on the intracellular gluta-

thione level as shown in Fig. 6. However, the combination

of a heat stress followed by a simulated dive as well as the

heat stress alone increased the total glutathione 33%

(p \ 0.05 vs. control 40 kPa O2) and 36% (p \ 0.01 vs.

control 40 kPa O2), respectively. Similar responses were

observed for reduced glutathione.

0

0.2

0.4

0.6

0.8

1

1.2N

OS

act

ivity

, rel

ativ

e to

con

trol

a a

Fig. 2 NOS activity in HUVEC exposed to HS for 1 h at 45�C and a

subsequent simulated dive (2.6 MPa) at 20 kPa O2 for 24 h. Data are

from two separate experiments. The results are shown as NOS activity

relative to control and given as the average ± SD of all the dishes

within each treatment (n = 8). a p \ 0.001 versus 1,4

0

0.4

0.8

1.2

1.6

NO

S, r

elat

ive

to c

ontr

ol

NOS activity

eNOS protein

a a

bc

Fig. 3 NOS activity and eNOS protein expression in HUVEC

exposed to HS for 1 h at 45�C and a subsequent simulated dive

(2.6 MPa) at 40 kPa O2 for 24 h. Data for both NOS activity and

eNOS protein are from two separate experiments. The results are

shown as activity and concentration respectively, relative to control

and given as the average ± SD of all the dishes within each treatment

(n = 6). a p \ 0.001 versus 1, 2, 4; b p \ 0.01 versus 2; c p \ 0.001

versus 1, 2

0

20

40

60

80

100

120

HS

P70

, rel

ativ

e to

con

trol

a, b

c

Fig. 4 Expression of HSP70 in HUVEC exposed to HS for 1 h at

45�C and a subsequent simulated dive (2.6 MPa), at 40 kPa O2 for

24 h. Data are from two separate experiments. The results are shown

as concentration relative to control and given as the average ± SD of

all the dishes within each treatment (n = 6). a p \ 0.01 versus 2;

b p \ 0.05 versus 1, 4; c p \ 0.001 versus 1, 2, 3, 4

0

0.5

1

1.5

2H

SP

90, r

elat

ive

to c

ontro

la, b

c

Fig. 5 HSP90 protein in HUVEC exposed to HS for 1 h at 45�C and

a subsequent simulated dive (2.6 MPa) at 40 kPa O2 for 24 h. Data

are from two separate experiments. The results are shown as

concentration relative to control and given as the average ± SD of

all the dishes within each treatment (n = 6). a p \ 0.001 versus 2,5; b p \ 0.05 versus 1; c p \ 0.05 versus 2, 3, 5

2722 Eur J Appl Physiol (2012) 112:2717–2725

123

The reduced to total glutathione ratio did not change

except in the group exposed to heat stress and a subsequent

simulated dive. Statistical analyses on data from two

independent experiments showed a lower reduced/total

glutathione ratio in this group compared to control 40 kPa

O2 (p \ 0.001, Fig. 6).

Discussion

The results of the present study demonstrated that the

enzymatic capability of NOS in human endothelial cells in

vitro was not significantly affected by a simulated dive or

an increase in pO2 from 20 to 40 kPa. The fact that there

was no significant difference between the level of NOS

activity or protein at normoxic and hyperoxic conditions

under normobaria was surprising, and in contrast to what

was observed by Zhen and co-workers (Zhen et al. 2008).

In their study, human coronary artery endothelial cells

(HCAECs) were subjected to oxidative stress by H2O2,

xanthine–xanthine oxidase (X–XO) generated superoxide,

or inhibiting GSH synthesis by buthionine sulfoximine

causing upregulatation of eNOS mRNA, protein expression

and activity. This may indicate that the 24 h hyperoxia

experienced by HUVEC in the present study did not gen-

erate equivalent oxidative stress. Exposure of rats up to

280 kPa O2 for 45 min increased NO generation in aorta,

but did not alter the protein content of neither eNOS nor

nNOS (Thom et al. 2003). Recent investigations with rat

aortic endothelial cells showed that NOS activity increased

rapidly with increasing pO2 at low levels with a Km of

4.2 kPa O2, and reaching maximal rate at approximately

100 kPa O2 (Hink et al. 2006). These results show that

below, approximately, 40 kPa there was a very rapid

change in NOS activity. Consequently, the slight but non-

significant increase in NOS activity seen in our present

study may be more prominent with a pO2 above 40 kPa. A

more detailed investigation of the oxygen effect in this

respect is an obvious task for future studies. Furthermore,

the elevated level of O2 during simulated diving did not

influence the enzyme’s capability for NO production. The

hyperoxic conditions at 40 kPa O2 might increase the

oxidation of BH4, implying decreased production of NO by

NOS in intact cells. However, since BH4 is added in excess

to the reaction mixture when NOS activity is determined in

cell lysates, any decrease in NOS activity due to lack of

BH4 availability will be masked. Consequently, the present

results show that the enzyme’s capability of generating

NOS was not altered after a simulated dive. The reduction

of NOS activity after heat stress was followed by a similar

reduction of the amount of NOS protein, indicating that the

response to heat stress may be regulated at the gene

expression level.

We have previously shown that both HSP70 and HSP90

increased after heat stress and that HSP70 was potentiated

by a subsequent simulated dive at 20 kPa O2 (Djurhuus

et al. 2010). Also in the present study at 40 kPa O2, heat-

induced HSP70 expression was potentiated by a subsequent

simulated dive. The potentiating effect was even greater

than at 20 kPa; in the present experiments HSP70

increased 79 versus 54 times (Djurhuus et al. 2010) com-

pared to control, indicating an effect of the elevated oxygen

partial pressure (p \ 0.05 for HSP70 response at 20 and

40 kPa). On the other hand, a simulated dive performed

after a HS abolished the HS-induced expression of HSP90

at 40 kPa O2. A study by Harris and co-workers in bovine

aortic endothelial cells (Harris et al. 2003) found that

HSP70 increased after a heat stress for 1 h at 45�C in a

similar manner to what was observed in the present study,

but no effects on eNOS expression and NOS activity were

observed. However, a moderate heat stress (42�C, 1 h)

resulted in a significant increase in both eNOS expression

and NOS activity. In contrast, the present study showed a

decreasing effect of HS at 45�C on NOS. It should be noted

however, that while Harris and co-workers measured the

effects 24 h after HS, all parameters in the present study

were determined 48 h after HS. The time point for deter-

mination may be important as we have shown that the

HSP70 response to HS declined from 24 to 48 h after HS

(data not shown).

A role for the antioxidant glutathione (GSH) in the

mechanisms involved in the regulation of NO synthesis in

endothelial cells has not been established and the literature

is contradicting. Glutathione has been shown to participate

in NO synthesis by the inducible NOS in macrophages

0

0.5

1

1.5

2G

luta

thio

ne e

quiv

alen

ts, r

elat

ive

to c

ontr

ol

GSH total GSH red

a

a b, c

d

Fig. 6 Intracellular GSH levels in HUVEC exposed to HS for 1 h at

45�C and a subsequent simulated dive (2.6 MPa) at 40 kPa O2 for

24 h. Data are from two separate experiments. The results are

expressed as equivalents of GSH relative to control and given as the

average ± SD of all the dishes within each treatment (n = 6).

a p \ 0.01 versus 1, 2, 4; b p \ 0.01 versus 1, 4; c p \ 0.05 versus 2;

d p \ 0.05 versus 1

Eur J Appl Physiol (2012) 112:2717–2725 2723

123

(Stuehr et al. 1990). Decrease in GSH level has also been

found to impair the ability of cultured hepatocytes to

synthesize NO (Harbrecht et al. 1997). Laursen and co-

workers (2001) concluded that GSH is a necessary cofactor

for endothelial NO synthesis, and in conditions of low GSH

availability, such as during oxidative stress, GSH levels

may be a rate-limiting factor in the synthesis of NO in

endothelial cells from rat. In contrast, several investigators

have suggested that GSH-depleting agents had no effect on

NO synthesis in endothelial cells (Huang et al. 2001;

Murphy et al. 1991). In the present study GSH seemed to

be inversely related to the effect on NOS, demonstrating

approximately 35% increase in total GSH after heat stress

with or without a subsequent dive at 40 kPa O2. This is in

contrast to our previous results at 20 kPa O2 where only the

combination of heat and a subsequent dive gave an increase

in GSH level (Djurhuus et al. 2010). Elevated GSH content

might protect against oxidation of BH4, and thus help to

maintain the NO production. Hence, there could be a role

for GSH in maintaining BH4 in its reduced form to pre-

serve eNOS function. An interesting explanation may be

found in a recent work by Chen and co-workers (Chen et al.

2010) who established that oxidized glutathione (GSSG)

induced S-glutathionylation of human eNOS resulting in

markedly decreased NOS activity. In our experiments, the

cells exposed to heat and a subsequent dive demonstrated

both an increase in the total GSH level and a decrease in

the ratio red/tot GSH. This might be due to increased

GSSG level and a resulting S-glutathionylation of eNOS,

thus explaining the decreased NOS activity. In the heat-

exposed group we did not observe significant changes in

the redox status, however, the total GSH level was

increased in both HS-exposed groups implying also an

increase in the absolute amount of GSSG. It is therefore

tempting to suggest that an increased GSSG level induced

glutathionylation of eNOS responsible for the decreased

NOS-activity observed in both HS-exposed groups. The

decrease in eNOS protein observed after exposure to heat

stress, might be due to decreased immunoreactivity of

eNOS against the monoclonal antibody due to the S-glu-

tathionylation, and not to a lower content of the total eNOS

protein. If the latter is the case, the previous assumption

that the eNOS response to HS is regulated at the gene

expression level may not be valid; rather eNOS may be

regulated at least in part by the post-translational mecha-

nisms, like S-glutathionylation. eNOS protein is also

known to be regulated by other post-translational modifi-

cations such as acylation, phosphorylation and S-nitrosy-

lation (Michel and Vanhoutte 2010).

Further studies of eNOS and the role of the cofactor

BH4 during saturation diving are in progress. The impact

of the depth is also an important aspect that should be

addressed in the future.

Conclusions

In summary, a heat stress (45�C, 1 h) decreased the NOS

activity and eNOS protein markedly. A dive alone, either at

20 or 40 kPa O2, had no effects on NOS activity or eNOS

protein. Increasing O2 levels under normobaria from 20 to

40 kPa also did not significantly affect the NOS activity or

protein. At 40 kPa O2, a simulated dive after heat stress

potentiated the HS-induced HSP70 response, but decreased

the heat-induced HSP90 response. GSH levels were found

to be inversely related to NOS activity and protein

expression, and might be explained by a possible post-

translational regulation by glutathionylation of eNOS

protein.

Our findings may have importance for understanding the

mechanisms involved in endothelial dysfunction following

decompression (stress).

Acknowledgments The authors would like to thank Mrs. Torill

Sage and Mr. Harald A. Sundland, both NUI AS, and Mrs. Torunn

Eide at Institute of Medicine, University of Bergen, for excellent

technical assistance. The present study is an extension of previous

work initiated by Professor Alf O. Brubakk at Dept. Circulation and

Medical Imaging, Norwegian University of Science and Technology

(NTNU), Trondheim, Norway. The authors are grateful for his never

ending enthusiasm and scientific support. This work was financially

supported by the Norwegian Research Council (NFR), Statoil,

ExxonMobil Norway and Gassco.

Conflict of interest Lise Fismen and Rune Djurhuus are employed

by NUI AS, a company that is 100% owned by Statoil. The other

authors declare no conflict of interest.

References

Brubakk AO, Eftedal OS, Wisløff U (2007) Endothelium and diving.

In: Aird WC (ed) Endothelial biomedicine. Cambridge Univer-

sity Press, Cambridge, pp 497–505

Cardounel AJ, Xia Y, Zweier JL (2005) Endogenous methylarginines

modulate superoxide as well as nitric oxide generation from

neuronal nitric-oxide synthase: differences in the effects of

monomethyl- and dimethylarginines in the presence and absence

of tetrahydrobiopterin. J Biol Chem 280:7540–7549

Chen CA, Wang TY, Varadharaj S, Reyes LA, Hemann C, Talukder

MA, Chen YR, Druhan LJ, Zweier JL (2010) S-glutathionylation

uncouples eNOS and regulates its cellular and vascular function.

Nature 468:1115–1118

Djurhuus R, Svardal AM, Mansoor MA, Ueland PM (1991)

Modulation of glutathione content and the effect on methionine

auxotrophy and cellular distribution of homocysteine and

cysteine in mouse cell lines. Carcinogenesis 12:241–247

Djurhuus R, Segadal K, Svardal AM (2006) Glutathione in blood cells

decreases without DNA breaks after a simulated saturation dive

to 250 msw. Aviat Space Environ Med 77:597–604

Djurhuus R, Nossum V, Lundsett N, Hovin W, Svardal AM, Havnes

MB, Fismen L, Hjelde A, Brubakk AO (2010) Simulated diving

after heat stress potentiates the induction of heat shock protein

70 and elevates glutathione in human endothelial cells. Cell

Stress Chaperones 15:405–414

2724 Eur J Appl Physiol (2012) 112:2717–2725

123

Dujic Z, Duplancic D, Marinovic-Terzic I, Bakovic D, Ivancev V,

Valic Z, Eterovic D, Petri NM, Wisloff U, Brubakk AO (2004)

Aerobic exercise before diving reduces venous gas bubble

formation in humans. J Physiol 555:637–642

Fontana J, Fulton D, Chen Y, Fairchild TA, McCabe TJ, Fujita N,

Tsuruo T, Sessa WC (2002) Domain mapping studies reveal that

the M domain of hsp90 serves as a molecular scaffold to regulate

Akt-dependent phosphorylation of endothelial nitric oxide

synthase and NO release. Circ Res 90:866–873

Fulton D, Gratton JP, Sessa WC (2001) Post-translational control of

endothelial nitric oxide synthase: why isn’t calcium/calmodulin

enough? J Pharmacol Exp Ther 299:818–824

Garcia-Cardena G, Fan R, Shah V, Sorrentino R, Cirino G,

Papapetropoulos A, Sessa WC (1998) Dynamic activation of

endothelial nitric oxide synthase by Hsp90. Nature 392:821–824

Green D, Cheetham C, Mavaddat L, Watts K, Best M, Taylor R,

O’Driscoll G (2002) Effect of lower limb exercise on forearm

vascular function: contribution of nitric oxide. Am J Physiol

Heart Circ Physiol 283:H899–H907

Gross SS, Wolin MS (1995) Nitric oxide: pathophysiological

mechanisms. Annu Rev Physiol 57:737–769

Harbrecht BG, Di Silvio M, Chough V, Kim YM, Simmons RL,

Billiar TR (1997) Glutathione regulates nitric oxide synthase in

cultured hepatocytes. Ann Surg 225:76–87

Harris MB, Blackstone MA, Ju H, Venema VJ, Venema RC (2003)

Heat-induced increases in endothelial NO synthase expression

and activity and endothelial NO release. Am J Physiol Heart Circ

Physiol 285:H333–H340

Hink J, Thom SR, Simonsen U, Rubin I, Jansen E (2006) Vascular

reactivity and endothelial NOS activity in rat thoracic aorta

during and after hyperbaric oxygen exposure. Am J Physiol

Heart Circ Physiol 291:H1988–H1998

Huang A, Xiao H, Samii JM, Vita JA, Keaney JF Jr (2001)

Contrasting effects of thiol-modulating agents on endothelial NO

bioactivity. Am J Physiol Cell Physiol 281:C719–C725

Huang KL, Wu CP, Chen YL, Kang BH, Lin YC (2003) Heat stress

attenuates air bubble-induced acute lung injury: a novel mech-

anism of diving acclimatization. J Appl Physiol 94:1485–1490

Kampinga HH, Hageman J, Vos MJ, Kubota H, Tanguay RM,

Bruford EA, Cheetham ME, Chen B, Hightower LE (2009)

Guidelines for the nomenclature of the human heat shock

proteins. Cell Stress Chaperones 14:105–111

Kingwell BA, Sherrard B, Jennings GL, Dart AM (1997) Four weeks

of cycle training increases basal production of nitric oxide from

the forearm. Am J Physiol 272:H1070–H1077

Knowles R, Salter M (1998) Measurement of NOS activity by

conversion of radiolabeled arginine to citrulline using ion-

exchange separation nitric oxide protocols. Methods Mol Biol

100:67–73

Kregel KC (2002) Heat shock proteins: modifying factors in

physiological stress responses and acquired thermotolerance.

J Appl Physiol 92:2177–2186

Laursen JB, Boesgaard S, Trautner S, Rubin I, Poulsen HE,

Aldershvile J (2001) Endothelium-dependent vasorelaxation in

inhibited by in vivo depletion of vascular thiol levels: role of

endothelial nitric oxide synthase. Free Radic Res 35:387–394

Lefroy DC, Crake T, Uren NG, Davies GJ, Maseri A (1993) Effect of

inhibition of nitric oxide synthesis on epicardial coronary artery

caliber and coronary blood flow in humans. Circulation

88:43–54

Lowry O, Rosebrough J, Farr A, Randall R (1951) Protein measure-

ment with the folin phenol reagent. J Biol Chem 193:265–275

Mansoor MA, Svardal AM, Ueland PM (1992) Determination of the

in vivo redox status of cysteine, cysteinylglycine, homocysteine,

and glutathione in human plasma. Anal Biochem 200:218–229

Medby C, Bye A, Wisløff U, Brubakk AO (2008) Heat shock

increases survival in rats exposed to hyperbaric pressure. Diving

Hyperb Med 38:189–193

Michel T, Vanhoutte PM (2010) Cellular signaling and NO produc-

tion. Pflugers Arch 459:807–816

Montcalm-Smith E, Caviness J, Chen Y, McCarron RM (2007) Stress

biomarkers in a rat model of decompression sickness. Aviat

Space Environ Med 78:87–93

Murphy ME, Piper HM, Watanabe H, Sies H (1991) Nitric oxide

production by cultured aortic endothelial cells in response to thiol

depletion and replenishment. J Biol Chem 266:19378–19383

Pollock JS, Forstermann U, Mitchell JA, Warner TD, Schmidt HH,

Nakane M, Murad F (1991) Purification and characterization of

particulate endothelium-derived relaxing factor synthase from

cultured and native bovine aortic endothelial cells. Proc Natl

Acad Sci USA 88:10480–10484

Pou S, Pou WS, Bredt DS, Snyder SH, Rosen GM (1992) Generation

of superoxide by purified brain nitric oxide synthase. J Biol

Chem 267:24173–24176

Rees DD, Palmer RM, Moncada S (1989) Role of endothelium-

derived nitric oxide in the regulation of blood pressure. Proc Natl

Acad Sci USA 86:3375–3378

Snyder SH, Bredt DS (1991) Nitric oxide as a neuronal messenger.

Trends Pharmacol Sci 12:125–128

Stuehr DJ, Kwon NS, Nathan CF (1990) FAD and GSH participate in

macrophage synthesis of nitric oxide. Biochem Biophys Res

Commun 168:558–565

Svardal AM, Mansoor MA, Ueland PM (1990) Determination of

reduced, oxidized, and protein-bound glutathione in human

plasma with precolumn derivatization with monobromobimane

and liquid chromatography. Anal Biochem 184:338–346

Thom SR, Fisher D, Zhang J, Bhopale VM, Ohnishi ST, Kotake Y,

Ohnishi T, Buerk DG (2003) Stimulation of perivascular nitric

oxide synthesis by oxygen. Am J Physiol Heart Circ Physiol

284:H1230–H1239

Vasquez-Vivar J, Kalyanaraman B, Martasek P, Hogg N, Masters BS,

Karoui H, Tordo P, Pritchard KA Jr (1998) Superoxide

generation by endothelial nitric oxide synthase: the influence

of cofactors. Proc Natl Acad Sci USA 95:9220–9225

Weissman B, Gross S (2001) Measurement of NO and NO synthase.

Curr Protoc Neurosci May; Chapter 7:Unit 7.13

Wever RM, van Dam T, van Rijn HJ, de Groot F, Rabelink TJ (1997)

Tetrahydrobiopterin regulates superoxide and nitric oxide gen-

eration by recombinant endothelial nitric oxide synthase. Bio-

chem Biophys Res Commun 237:340–344

Wisloff U, Richardson RS, Brubakk AO (2003) NOS inhibition

increases bubble formation and reduces survival in sedentary but

not exercised rats. J Physiol 546:577–582

Wisloff U, Richardson RS, Brubakk AO (2004) Exercise and nitric

oxide prevent bubble formation: a novel approach to the

prevention of decompression sickness? J Physiol 555:825–829

Zhen J, Lu H, Wang XQ, Vaziri ND, Zhou XJ (2008) Upregulation of

endothelial and inducible nitric oxide synthase expression by

reactive oxygen species. Am J Hypertens 21:28–34

Eur J Appl Physiol (2012) 112:2717–2725 2725

123