Attenuation of cardiac mitochondrial dysfunction
by melatonin in septic mice
Germaine Escames
1
, Luis C. Lo
´
pez
1
, Francisco Ortiz
1
, Ana Lo
´
pez
1
, Jose
´
A. Garcı
´a
1
, Eduardo Ros
2
and Darı
´
o Acun
˜
a-Castroviejo
1,3
1 Instituto de Biotecnologı
´
a, Departamento de Fisiologı
´

(Received 4 December 2006, revised 9
February 2007, accepted 23 February 2007)
doi:10.1111/j.1742-4658.2007.05755.x
The existence of an inducible mitochondrial nitric oxide synthase has been
recently related to the nitrosative ⁄ oxidative damage and mitochondrial dys-
function that occurs during endotoxemia. Melatonin inhibits both inducible
nitric oxide synthase and inducible mitochondrial nitric oxide synthase
activities, a finding related to the antiseptic properties of the indoleamine.
Hence, we examined the changes in inducible nitric oxide synthase ⁄ indu-
cible mitochondrial nitric oxide synthase expression and activity, bioener-
getics and oxidative stress in heart mitochondria following cecal ligation
and puncture-induced sepsis in wild-type (iNOS
+ ⁄ +
) and inducible nitric
oxide synthase-deficient (iNOS
– ⁄ –
) mice. We also evaluated whether melato-
nin reduces the expression of inducible nitric oxide synthase ⁄ inducible
mitochondrial nitric oxide synthase, and whether this inhibition improves
mitochondrial function in this experimental paradigm. The results show
that cecal ligation and puncture induced an increase of inducible mito-
chondrial nitric oxide synthase in iNOS
+ ⁄ +
mice that was accompanied by
oxidative stress, respiratory chain impairment, and reduced ATP produc-
tion, although the ATPase activity remained unchanged. Real-time PCR
analysis showed that induction of inducible nitric oxide synthase during
sepsis was related to the increase of inducible mitochondrial nitric oxide syn-
thase activity, as both inducible nitric oxide synthase and inducible mito-
chondrial nitric oxide synthase were absent in iNOS

and activity of i-mtNOS, but not those of mtNOS,
increase during sepsis [9,11,12]. Other studies have
shown induction of mitochondrial NOS in the dia-
phragm and heart of septic rats, although these reports
did not distinguish between constitutive and inducible
forms [8,18].
Increasing evidence suggests that the nitric oxide
(NO) produced by i-mtNOS plays a role in mitochond-
rial dysfunction during sepsis [9,11,12]. Because iNOS
– ⁄ –
mice do not express i-mtNOS, and the mitochondria
of these mice were unaffected by sepsis, it was sugges-
ted that the overproduction of NO by i-mtNOS is the
main factor responsible for mitochondrial nitrosative ⁄
oxidative stress and impairment during endotoxemia
[9,12]. The induction of i-mtNOS after lipopoly-
saccharide administration leads to an increase in NO
and other reactive species, such as superoxide anion
(O
2
–
), hydrogen peroxide (H
2
O
2
) and peroxynitrite
(ONOO
–
), in heart and diaphragm mitochondria
[8,18]. Cecal ligation and puncture (CLP) also induces

have shown that melatonin (aMT) protects against
mitochondrial oxidative stress, due to its antioxidant
properties and its ability to enter mitochondria [24–
28]. In muscular tissues such as skeletal muscle and
diaphragm of septic mice, aMT administration inhi-
bited the activity of i-mtNOS, restoring the mito-
chondrial GSH pool and the ETC activity in these
animals [9,12,29].
Mitochondrial dysfunction is an important patho-
physiologic event related to heart failure during sepsis,
and i-mtNOS may be directly related to it. To address
this question, we induced sepsis by CLP in iNOS
+ ⁄ +
and iNOS
– ⁄ –
mice, and explored in heart mitochon-
dria: (a) the presence and source of i-mtNOS; (b) the
relationship between i-mtNOS induction, ETC dys-
function, and oxidative phosphorylation activity; (c)
the steady-state energy and ATP production; and (d)
the protective effect of aMT against mitochondrial
damage produced during sepsis.
Results
Mitochondrial NOS activities
Figure 1 shows that heart mitochondria from
iNOS
+ ⁄ +
mice contain two mitochondrial NOS iso-
forms: a constitutive, Ca
2+

tion with cytosolic NOS was assessed by the absence
of any detectable NOS activity and nitrite levels in the
supernatant of the final centrifugation step (data not
shown). These data confirm the purity of the mito-
chondria used in our experiments, and guarantee the
mitochondrial origin of the NOS activity reported
NOS and heart mitochondrial dysfunction in sepsis G. Escames et al.
2136 FEBS Journal 274 (2007) 2135–2147 ª 2007 The Authors Journal compilation ª 2007 FEBS
here. Moreover, the method used for NOS measure-
ment specifically detects mtNOS activity, and the addi-
tion of NG-monomethyl- l-arginine (l-NMMA) (300 lm)
to the reaction mixture of mitochondrial samples from
septic mice blocked the transformation of l-arginine to
l-citrulline, due to mtNOS inhibition (14.67 ± 3.09
versus 1.09 ± 0.87 pmol citrullineÆmin
)1
Æmg
)1
protein,
CLP and CLP + l-NMMA, respectively) [9,11,12].
iNOS
+ ⁄ +
mice exhibited a slight increase in nitrite
level after sepsis, which was counteracted by aMT
treatment (Table 1). Interestingly, iNOS
– ⁄ –
mice showed
a significant decrease in nitrite level during sepsis, coin-
ciding with the mtNOS activity inhibition, that was
partially counteracted by aMT.

2+
-dependent,
and inducible, Ca
2+
-independent, compo-
nents in iNOS
+ ⁄ +
mice (A). Deficient iNOS
mice, however, show only the constitutive,
Ca
2+
-dependent component (B). In both
cases, mice were subjected to CLP to
induce sepsis, and killed 24 h later. Pure
mitochondrial preparations were used to
determine NOS activity with
L-[
3
H]arginine
as substrate. Data represent the means
± SE of six experiments per group. C, con-
trol; S, sepsis; S + aMT, sepsis + aMT.
*P<0.05, **P<0.01, ***P<0.001 versus
C;
#
P<0.05,
###
P<0.001 versus S.
Table 1. Effects of sepsis and aMT treatment on the mitochondrial nitrite and LPO levels, and on the activity of the mitochondrial ATPase in
wild-type and iNOS knockout mice. C, control; S, sepsis; S + aMT, sepsis + aMT. Sepsis was induced by CLP, and the animals were killed

sepsis, whereas aMT treatment increased it to above
control values (Fig. 3B). Heart mitochondria from
iNOS
– ⁄ –
mice did not show changes in GPx and GRd
activities with any treatment (Fig. 3). Basal GPx activ-
ity was lower in iNOS
– ⁄ –
than in iNOS
+ ⁄ +
mice
(Fig. 3A).
The mitochondrial level of GSH decreased and that
of GSSG increased in hearts from iNOS
+ ⁄ +
mice after
CLP (Fig. 4A,B), raising the GSSG ⁄ GSH ratio
(Fig. 4C). Sepsis also reduced total glutathione levels in
iNOS
+ ⁄ +
mice (Fig. 4D). Treatment with aMT
increased GSH levels and reduced GSSG levels in
iNOS
+ ⁄ +
mice, normalizing the GSSG ⁄ GSH ratio
(Fig. 4A–C). aMT also increased the total glutathione
pool in this mouse strain (Fig. 4D). No changes in glu-
tathione levels were found in heart mitochondria
of iNOS
– ⁄ –

+ ⁄ +
and iNOS
– ⁄ –
mice. Data represent the means ± SE of six
experiments per group. C, control; S, sep-
sis; S + aMT, sepsis + aMT. *P<0.05 and
**P<0.01 versus C;
##
P<0.005 versus S;
+
P<0.05 versus iNOS
+ ⁄ +
mice.
NOS and heart mitochondrial dysfunction in sepsis G. Escames et al.
2138 FEBS Journal 274 (2007) 2135–2147 ª 2007 The Authors Journal compilation ª 2007 FEBS
than in iNOS
+ ⁄ +
mice (Fig. 5A–D). The activity of
the ETC complexes was also unchanged by aMT treat-
ment in iNOS
– ⁄ –
mice. No changes in ATPase activity
were observed in any mouse strain under sepsis, and
aMT treatment only slightly decreased ATPase activity
in iNOS
+ ⁄ +
mice (Table 1).
Fig. 4. GSH level (A), GSSG level (B),
GSSG ⁄ GSH ratio (C) and GSH + GSSG level
(D) in heart mitochondria of iNOS

+ ⁄ +
mice.
G. Escames et al. NOS and heart mitochondrial dysfunction in sepsis
FEBS Journal 274 (2007) 2135–2147 ª 2007 The Authors Journal compilation ª 2007 FEBS 2139
Mitochondrial ATP production
To assess whether sepsis modifies the bioenergetic sta-
tus of heart mitochondria, ATP production was deter-
mined. ATP production was significantly reduced in
iNOS
+ ⁄ +
but not in iNOS
– ⁄ –
mice during sepsis,
whereas aMT administration restored the ability of
mitochondria to produce ATP in the former (Fig. 6).
After the ATP production assay, the amount of AMP
in the samples was less that 3% of the total nucleo-
tides, discounting extramitochondrial ATP production
by adenylate kinase in our assays. The experimental
procedure used here allowed us to detect ATP inside
(pellet, fraction p2) and outside (supernatant, frac-
tion s1) the mitochondria. The results indicated that
92–98% of the ATP produced was detected outside the
mitochondria.
Animal survival
To determine the mortality of CLP-induced sepsis in
our experimental paradigm, and to assess whether the
improvement in mitochondrial function after aMT
treatment was followed by a reduction in mortality,
mice survival was analyzed. Figure 7 shows the survi-

inhibition of the ETC complexes, leading to a reduction
in ATP. Because heart mitochondria from iNOS
– ⁄ –
Fig. 6. Changes in mitochondrial ATP production in heart of
iNOS
+ ⁄ +
and iNOS
– ⁄ –
mice, using succinate as substrate. Data rep-
resent the mean ± SE of six experiments per group. C, control;
S, sepsis; S + aMT, sepsis + aMT. *P < 0.05 and **P<0.01
versus C;
##
P<0.01 versus S.
Fig. 7. Survival curves obtained from untreated and aMT-treated
septic iNOS
+ ⁄ +
and iNOS
– ⁄ –
mice. The total number of animals
used in this study was 20 in each group.
NOS and heart mitochondrial dysfunction in sepsis G. Escames et al.
2140 FEBS Journal 274 (2007) 2135–2147 ª 2007 The Authors Journal compilation ª 2007 FEBS
mice were unaffected by endotoxemia and they do not
express i-mtNOS, mitochondrial impairment during
sepsis was probably related to i-mtNOS induction in
iNOS
+ ⁄ +
mice. aMT treatment counteracted sepsis-
induced iNOS mRNA expression, a finding related to

in heart mitochondria from rats [13]. Normally, the
induction of i-mtNOS produces a significant increase
in NO and nitrite [9,11,12]. The lack of a significant
increase in nitrite in iNOS
+ ⁄ +
mice after sepsis repor-
ted here could be explained by two main mechanisms.
In mitochondria, the major oxidative decay pathway
of NO is its reaction with O
2
–
to form ONOO
–
[20]. In
turn, ONOO
–
reacts with a variety of biomolecules
[35]. Moreover, ONOO
–
can react with H4-biopterin
(BH), a cofactor necessary for NO synthesis by NOS,
leading to formation of the BH
3
radical [36], and caus-
ing NOS inactivation [37,38]. An alternative explan-
ation for the lack of changes in nitrite under sepsis is
the presence of an NOS-independent NO source in
mitochondria. Alterations in the redox state of the
ETC lead to the formation of reactive nitrogen species,
including NO and ONOO

opathy [41]. In these phases, the innate inflammatory
response corresponds with iNOS induction and a subse-
quent increase in NO. These results suggest that ATPase
is more resistant to oxidative ⁄ nitrosative stress than
ETC complexes, probably because ETC complexes,
unlike ATPase, have redox centers such as Fe–S that are
very sensitive to NO [42].
ETC coupled with oxidative phosphorylation is
responsible for the production of 90–95% of the total
ATP synthesized in the cell [26]. Thus, ETC damage
may alter the synthesis of ATP without any effect on
ATPase. Our results show a reduced ability of the
mitochondria to produce ATP during sepsis, which
may reduce cardiomyocyte contractility [43,44]. Be-
cause the activity of the respiratory complexes was not
affected in septic iNOS
– ⁄ –
mice, ATP production by
heart mitochondria was not altered in this mouse
strain. Thus, the reduction in the production of ATP
found in diaphragm and heart after endotoxin admin-
istration [5,22,45] probably reflects mitochondrial
impairment due to i-mtNOS induction by the toxin.
Heart mitochondria from iNOS
– ⁄ –
mice show lower
complex I and II activities and higher complex III and
IV activities than iNOS
+ ⁄ +
mice, a finding also des-

2
is superoxide dismutase activity [47], whereas HO
can be derived from H
2
O
2
and ONOO
–
decomposition,
although the latter is a minor process [35]. The small
effect on complex I compared with the strong inhibition
of complex III produced during sepsis in iNOS
+ ⁄ +
mice suggests that the latter was the most important
source of free radicals in our experimental model. NO
and O
2
–
react to produce ONOO
–
in mitochondria
[7,19,20], increasing ETC damage [19,20] and LPO
activity [48]. Besides causing direct oxidative damage,
ONOO
–
can produce nitration, and to a lesser extent ni-
trosation, of mitochondrial components indirectly [35].
The free radical pathways of ONOO
–
are mainly initi-

nitrogen dioxide and HO
•
radicals. Carbonate radicals,
however, are poor direct initiators of LPO, due to their
negative charge, which limits their diffusion to the
hydrophobic domains of membrane phospholipids [35].
These mechanisms explain the LPO increase found in
heart mitochondria from septic iNOS
+ ⁄ +
mice, a find-
ing related to the i-mtNOS induction, because iNOS
– ⁄ –
mice did not show changes in LPO levels.
Mitochondrial GPx activity increased in iNOS
+ ⁄ +
mice, reflecting a compensatory mechanism to reduce
oxidative stress during sepsis. However, the
GSSG ⁄ GSH ratio remains elevated under these condi-
tions, because the reduced activity of GRd, probably
due to oxidative damage [53], prevents the recovery of
GSH from GSSG. Moreover, total glutathione levels
in these mitochondria were also reduced, probably
reflecting inhibition of GSH transport into the mito-
chondria [49]. The lack of mitochondrial oxidative
stress and the presence of a normal GSH pool in septic
iNOS
– ⁄ –
mice further support the idea that i-mtNOS
induction is the main event related to oxidative stress
during sepsis in heart mitochondria.

energy for muscle contraction, avoiding myocardial
dysfunction, and probably heart failure, in sepsis.
The significant increase in survival of mice treated
with aMT further supports this observation. It is inte-
resting that aMT had minor effects on heart mito-
chondria from iNOS
– ⁄ –
mice. As reported in other
pathophysiologic conditions, it seems that aMT can
upregulate mitochondrial function when it is impaired
[9,11,12,24,49,53,55], but the indoleamine had minor
effects under normal conditions. The half-life and
maximum survival time of septic iNOS
– ⁄ –
mice were
significantly higher than those of wild-type animals.
Moreover, aMT treatment significantly increased the
half-life and maximum survival time of mutant mice.
Because iNOS
– ⁄ –
mice with sepsis did not show signifi-
cant mitochondrial damage, they probably died by a
mechanism different from iNOS-dependent dysfunc-
tion, such as cyclooxygenase-2 induction. Because
aMT treatment had only minor effects on mitochon-
dria of iNOS
– ⁄ –
mice, the inhibitory effect of aMT on
cyclooxygenase-2 expression may explain the signifi-
cant improvement in survival of iNOS

to produce ATP, and increasing mice survival. These
properties, together with the prevention of endotoxin-
induced circulatory failure in rats [58,59], and the mor-
tality reduction in septic newborns after aMT therapy
[60], suggest that the use of the indoleamine in septic
patients should be seriously considered.
Experimental procedures
Chemicals
l-[2,3,4,5-
3
H]arginine monohydrochloride (58 CiÆmmol
)1
)
was obtained from Amersham Biosciences Europe GmBH
(Barcelona, Spain). Liquid scintillation cocktail (Ecolume)
was purchased from ICN (Madrid, Spain). All other
chemicals, of the purest available grade, were obtained from
Sigma-Aldrich (Madrid, Spain) unless otherwise specified.
Experimental animals
All procedures involving animals were carried out under an
approved protocol and in accordance with the Spanish
Government Guide and the European Community Guide
for animal care. iNOS knockout B6.129P2-Nos2
tm1Lau
mice
(iNOS
– ⁄ –
) and their respective wild-type control C57 ⁄ Bl ⁄ 6
mice (iNOS
+ ⁄ +

)1
) for 30 s, washed
with buffer A (250 mm mannitol, 0.5 mm EGTA, 5 mm He-
pes, 0.1% BSA, pH 7.4, 4 °C), and homogenized (1 : 10,
w ⁄ v) in buffer A at 800 r.p.m. at 4 °C with a Teflon pestle.
The homogenate was aliquoted, and centrifuged at 600 g
for 5 min at 4 °C (twice) (rotor type F34-6-38 Eppendorf
5810R centrifuge), and the supernatants were centrifuged at
10 300 g for 10 min at 4 °C (rotor type F1255 Beckman
TL-100 centrifuge). Then, the mitochondrial pellets were
suspended in 0.5 mL of buffer A, and placed in ultracentri-
fuge tubes containing 1.4 mL of buffer B (225 mm manni-
tol, 1 mm EGTA, 25 mm Hepes, 0.1% BSA, pH 7.4, 4 °C)
and 0.6 mL of Percoll. The mixture was centrifuged at
105 000 g for 30 min at 4 °C (rotor type F1255 Beckman
TL-100 centrifuge). The fraction corresponding to a pure
mitochondrial fraction was collected, washed twice with
buffer A at 10 300 g for 10 min at 4 °C (rotor type F1255
Beckman TL-100 centrifuge) to remove the Percoll, and
washed again with a high ionic strength solution of KCl
(150 mm) to yield a highly pure mitochondrial preparation
without contaminating organelles and broken mitochondria
[28,50]. Aliquots of these pure mitochondrial fractions were
frozen to ) 80 °C. The purity of the mitochondrial prepara-
tions was assessed as described elsewhere [9,11].
Mitochondrial NOS activity measurement
Measurement of constitutive, Ca
2+
-dependent, and indu-
cible, Ca

0.5 mm inosine, 0.5 mgÆmL
)1
BSA, 0.1 mm CaCl
2
,10lm
l-arginine, and 40 nml-[
3
H]arginine (pH 7.6). The reaction
was started by the addition of 10 l L of NADPH (0.75 mm
final concentration), and continued for 30 min at 37 °C. To
determinate the Ca
2+
-independent activity of NOS
(i-mtNOS), 10 mm EDTA was added to the buffer before
the reaction was started. Control incubations were per-
formed in the absence of NADPH. The reaction was
stopped by adding 400 lL of cold 0.1 m Hepes buffer con-
taining 10 mm EGTA and 1 mml-citrulline (pH 5.5). The
mixture was decanted onto a 2 mL column packed with
Dowex-50W ion exchange resin (Na
+
form), and eluted
with 1.2 mL of water. l-[
3
H]Citrulline was quantified
by liquid scintillation spectroscopy. The retention
of l-[
3
H]arginine by the column was greater than 98%.
Enzyme activity was determined as pmoL l-[

used as the final expression value. A negative control with-
out RNA template was run. The PCR program was initi-
ated by 10 min at 95 °C before 40 thermal cycles, each of
30 s at 95 °C and 1 min at 55 °C. Data were analyzed
according to the relative standard curve method, construc-
ted with triplicate serial dilutions (50, 5, 0.5 and 0.05 ng),
and were normalized by b-actin expression.
Nitrite determination
Mitochondrial fractions were thawed and suspended in ice-
cold distilled water, and immediately sonicated to break the
mitochondrial membranes. Aliquots of these samples were
used to calculate nitrite levels following the Griess reaction
[64], and expressed in nmoL nitriteÆmg protein
)1
.
Determination of mitochondrial function
The activities of the four respiratory complexes were deter-
mined as previously described [65,66], with slight modifica-
tions [9,12], and expressed as nmoLÆmin
)1
Æmg protein
)1
.
Complex V (ATP synthase) activity was measured by
following the rate of hydrolysis of ATP to ADP + P
i
.
Ferrous sulfate ⁄ ammonium molybdate reagent was utilized
to determinate P
i

125 nmol of ADP. After 45 s, the sample was centrifuged
at 13 000 g for 3 min at 2 °C (rotor type F1255 Beckman
TL-100 centrifuge) [68,69], and the ATP content in the pel-
let (p2) and supernatant (s1) was measured. Ice-cold 0.5 m
perchloric acid was rapidly added to the p1, p2 and s1 frac-
tions, mixed for 2 min in a vortex mixer, and centrifuged at
25 000 g for 15 min at 2 °C (rotor type F1255 Beckman
TL-100 centrifuge) to precipitate proteins. The pellets were
frozen to ) 80 °C for protein determination [63]; the sup-
ernatants were mixed with 8 lLof5m potassium carbon-
ate to neutralize the pH, and centrifuged at 12 000 g for
10 min at 2 °C (rotor type F1255 Beckman TL-100 centri-
fuge). ATP was measured in the resultant supernatants by
HPLC with a 4 · 250 mm ProPac PA1 column (Dionex,
Barcelona, Spain) [70]. After stabilization of the column
with the mobile phase, samples (20 lL) were injected onto
the HPLC system. The mobile phase consisted of water
(phase A) and 0.3 m ammonium carbonate (pH 8.9) (pha-
se B), and the following time schedule for the binary gradi-
ent (flow rate 1 mLÆmin
)1
) was used: 5 min, 50% A and
50% B; 5 min, 50% to 100% B, and then 100% B for
NOS and heart mitochondrial dysfunction in sepsis G. Escames et al.
2144 FEBS Journal 274 (2007) 2135–2147 ª 2007 The Authors Journal compilation ª 2007 FEBS
25 min; 5 min, 100% to 50% B, and then another 5 min
with 50% B [71]. Water was used for calibration purposes.
A standard curve was constructed with 3.125 lgÆmL
)1
,

ation, aliquots of the mitochondrial fraction were suspen-
ded in 200 lLof50mm potassium phosphate buffer
containing 1 mm EDTA-K
2
(pH 7.4), and the oxidation of
NADPH was spectrophotometrically measured for 3 min at
340 nm [74]. The activity of GPx and GRd was expressed
in nmolÆmin
)1
Æmg protein
)1
.
Statistical analysis
Data are expressed as means ± SEM. Significance was
determined using two-way ANOVA followed by Dunnet’s
post hoc test, when appropriate. The level of statistical sig-
nificance was taken as P < 0.05.
Acknowledgements
This study was partially supported by grants FIS01⁄
1076, PI03 ⁄ 0817 and G03 ⁄ 137 from the Instituto de
Salud Carlos III, and Consejerı
´
a de Educacio
´
n, Junta
de Andalucı
´
a (CTS-101). L. C. Lo
´
pez is an FPI fellow

endotoxemia. FASEB J 13, 1637–1646.
8 Boveris A, A
´
lvarez S & Navarro A (2002) The role
of mitochondrial nitric oxide synthase in inflammation
and septic shock. Free Radic Biol Med 33, 1186–1193.
9 Escames G, Lo
´
pez LC, Tapias V, Utrilla P, Reiter RJ,
Hitos AB, Leo
´
n J, Rodrı
´
guez MI & Acun
˜
a-Castroviejo
D (2006) Melatonin counteracts inducible mitochondrial
nitric oxide synthase-dependent mitochondrial dysfunc-
tion in skeletal muscle of septic mice. J Pineal Res 40,
71–78.
10 Massion PB, Feron O, Dessy C & Balligand JJ (2003)
Nitric oxide and cardiac function: ten years after, and
continuing. Circ Res 93, 388–398.
11 Escames G, Leo
´
n J, Macı
´
as M, Khaldy H & Acun
˜
a-

Aubier M, Boczkowski J & Poderoso JJ (2004) The
mitochondrial interplay of ubiquinol and nitric oxide in
endotoxemia. Methods Enzymol 382, 67–81.
16 Kanai AJ, Pearce LL, Clemens PR, Birder LA,
VanBibber MM, Choi SY, de Groat WC & Peterson
J (2001) Identification of a neuronal nitric oxide
synthase in isolated cardiac mitochondria using
G. Escames et al. NOS and heart mitochondrial dysfunction in sepsis
FEBS Journal 274 (2007) 2135–2147 ª 2007 The Authors Journal compilation ª 2007 FEBS 2145
electrochemical detection. Proc Natl Acad Sci USA
98, 14126–11431.
17 Tatoyan A & Giulivi C (1998) Purification and charac-
terization of a nitric oxide synthase from rat liver mito-
chondria. J Biol Chem 273, 11044–11048.
18 A
´
lvarez S & Bo
´
veris A (2004) Mitochondrial nitric
oxide metabolism in rat muscle during endotoxemia.
Free Radic Biol Med 37, 1472–1478.
19 Brown GC (2001) Regulation of mitochondrial respir-
ation by nitric oxide inhibition of cytochrome c oxidase.
Biochim Biophys Acta 1504, 46–57.
20 Cadenas E, Poderoso JJ, Antunes F & Boveris A (2000)
Analysis of the pathways of nitric oxide utilization in
mitochondria. Free Radic Res 33, 747–756.
21 Piantadosi CA, Tatro LG & Whorton AR (2002) Nitric
oxide and differential effects of ATP on mitochondrial
permeability transition. Nitric Oxide 6 , 45–60.

n M, Macı
´
as M, Escames
G, Leo
´
n J, Khaldy H & Reitrer RJ (2001) Melatonin,
mitochondria and cellular bioenergetics. J Pineal Res
30, 65–74.
27 Leo
´
n J, Acun
˜
a-Castroviejo D, Escames G, Tan DX &
Reiter RJ (2005) Melatonin mitigates mitochondrial
malfunction. J Pineal Res 38, 1–9.
28 Leo
´
n J, Acun
˜
a-Castroviejo D, Sainz RM, Mayo JC,
Tan DX & Reiter RJ (2004) Melatonin and mitochon-
drial function. Life Sci 75, 765–790.
29 Crespo E, Macias M, Pozo D, Escames G, Martin M,
Vives F, Guerreo JM & Acun
˜
a-Castroviejo D (1999)
Melatonin inhibits expression of the inducible NO
synthase II in liver and lung and prevents endotoxemia
in lipopolysaccharide-induced multiple organ dysfunc-
tion syndrome in rats. FASEB J 13, 1537–1546.

Kunimoto M (2004) Neuronal nitric oxide synthase
(nNOS) catalyzes one-electron reduction of 2,4,6-trini-
trotoluene, resulting in decreased nitric oxide production
and increased nNOS gene expression: implications for
oxidative stress. Free Radic Biol Med 37, 350–357.
39 Lacza Z, Kozlov AV, Pankotai E, Csorda
´
s A, Wolf G,
Redl H, Kollai M, Szabo C, Busija DW & Horn TF
(2006) Mitochondria produce reactive nitrogen species
via an arginine-independent pathway. Free Radic Res
40, 369–378.
40 Nohl H, Staniek K & Kozlov AV (2005) The existence
and significance of a mitochondrial nitrite reductase.
Redox Rep 10 , 281–286.
41 Vyatkina G, Bhatia V, Gerstner A, Papaconstantinou J
& Garg N (2004) Impaired mitochondrial respiratory
chain and bioenergetics during chagasic cardiomyopathy
development. Biochim Biophys Acta 1689, 162–173.
42 Pearce LL, Epperly MW, Greenberger JS, Pitt BR &
Peterson J (2001) Identification of respiratory complexes
I and III as mitochondrial sites of damage following
exposure to ionizing radiation and nitric oxide. Nitric
Oxide 5, 128–136.
43 Ataullakhanov FI & Vitvitsky VM (2002) What deter-
mines the intracellular ATP concentration. Biosci Rep
22, 501–511.
44 Saks V, Dzeja P, Schlattner U, Vendelin M, Terzic A &
Wallimann T (2006) Cardiac system bioenergetics: meta-
bolic basis of the Frank–Starling law. J Physiol 571,

Mene
´
ndez-Pela
´
ez A (1996) Neurohormone melatonin
prevents cell damage: effect on gene expression for anti-
oxidant enzymes. FASEB J 10, 882–890.
51 Mayo JC, Sainz RM, Tan DX, Hardeland R, Leo
´
nJ,
Rodrı
´
guez C & Reiter RJ (2005) Anti-inflammatory
actions of melatonin and its metabolites, N
1
-acetyl-
N
2
-formyl-5-methoxykynuramine (aFMK) and (N
1
-
acetyl-5-methxykynuramine (AMK), in macrophages.
J Neuroimmunol 165 , 139–149.
52 Tan DX, Manchester LC, Reiter RJ, Qi WB, Karbow-
nik M & Calvo JR (2000) Significance of melatonin in
antioxidative defense system: reactions and products.
Biol Signals Recep 9, 137–159.
53 Martı
´
n M, Macı

Mitochondrial permeability transition and oxidative
stress. FEBS Lett 495, 12–15.
58 Wu CC, Chiao CW, Hsiao G, Chen A & Yen MH
(2001) Melatonin prevents endotoxin-induced circula-
tory failure in rats. J Pineal Res 30 , 147–156.
59 Escames G, Acun
˜
a-Castroviejo D, Lo
´
pez LC, Tan DX,
Maldonado MD, Sa
´
nchez-Hidalgo M, Leo
´
n J & Reiter
RJ (2006) The pharmacological utility of melatonin in
the treatment of septic shock. J Pharm Pharmacol 58,
1153–1165.
60 Gitto E, Karbownik M, Reiter RJ, Tan DT, Cuzzocrea
S, Chiurazzi P, Cordaro S, Corona G, Troimarchi G &
Barberi I (2001) Effect of melatonin treatment in septic
newborns. Pediatr Res 50, 756–760.
61 Wichterman KA, Baue AE & Chaudry IH (1980) Sepsis
and septic shock ) a review of laboratory models and a
proposal. J Surg Res 29, 189–201.
62 Bredt DS & Snyder SH (1989) NO mediates glutamate-
linked enhancement of cGMP levels in the cerebellum.
Proc Natl Acad Sci USA 86, 9030–9033.
63 Lowry OH, Rosenbrough NJ, Farr AL & Randall RJ
(1951) Protein measurement with Folin phenol reagent.

Part 2. Separations on polymeric supports. J High
Resolut Chromatogr 20, 693–696.
72 Esterbauer H & Cheeseman KH (1990) Determination
of aldehydic lipid peroxidation products: malonalde-
hyde and 4-hydroxynonenal. Methods Enzymol 186,
407–421.
73 Hissin PJ & Hilf R (1976) A fluorometric method for
determination of oxidized and reduced glutathione in
tissues. Anal Biochem 74, 214–216.
74 Jaskot RH, Charlet EG, Grose EC, Grady MA &
Roycroft JH (1983) An automatic analysis of glu-
tathione peroxidase, glutathione-S-transferase, and
reductase activity in animal tissue. J Anal Toxicol 7,
86–88.
G. Escames et al. NOS and heart mitochondrial dysfunction in sepsis
FEBS Journal 274 (2007) 2135–2147 ª 2007 The Authors Journal compilation ª 2007 FEBS 2147