Biochem. J. (2009) 424, 221–231 (Printed in Great Britain)
221
doi:10.1042/BJ20091144
Myosin is reversibly inhibited by S-nitrosylation
Leonardo NOGUEIRA*1,2 , Cicero FIGUEIREDO-FREITAS*2 , Gustavo CASIMIRO-LOPES*, Margaret H. MAGDESIAN*,
Jamil ASSREUY† and Martha M. SORENSON*3
*Instituto de Bioquı́mica Médica, Centro de Ciências da Saúde, Universidade Federal do Rio de Janeiro, Cidade Universitária, Rio de Janeiro, RJ 21941-590, Brazil, and †Departamento
de Farmacologia, Centro de Ciências Biológicas, Universidade Federal de Santa Catarina, Florianópolis, SC 88015-420, Brazil
Nitric oxide (NO• ) is synthesized in skeletal muscle and its
production increases during contractile activity. Although myosin
is the most abundant protein in muscle, it is not known whether
myosin is a target of NO• or NO• derivatives. In the present
study, we have shown that exercise increases protein S-nitrosylation in muscle, and, among contractile proteins, myosin
is the principal target of exogenous SNOs (S-nitrosothiols) in
both skinned skeletal muscle fibres and differentiated myotubes.
The reaction of isolated myosin with S-nitrosoglutathione results
in S-nitrosylation at multiple cysteine thiols and produces two
populations of protein-bound SNOs with different stabilities.
The less-stable population inhibits the physiological ATPase
activity, without affecting the affinity of myosin for actin.
However, myosin is neither inhibited nor S-nitrosylated by the
NO• donor diethylamine NONOate, indicating a requirement for
transnitrosylation between low-mass SNO and myosin cysteine
thiols rather than a direct reaction of myosin with NO• or its autooxidation products. Interestingly, alkylation of the most reactive
thiols of myosin by N-ethylmaleimide does not inhibit formation
of a stable population of protein-SNOs, suggesting that these sites
are located in less accessible regions of the protein than those
that affect activity. The present study reveals a new link between
exercise and S-nitrosylation of skeletal muscle contractile
proteins that may be important under (patho)physiological
conditions.
INTRODUCTION
The formation of RSNOs, including protein-SNO, in biological
systems may occur by a number of different mechanisms,
including the direct reaction of protein thiol with NO• or its
auto-oxidation products or transnitrosylative transfer of NO+
from protein-bound or low-mass SNO to protein thiol [12].
The importance of transnitrosylation is evident by an increase
in constitutive SNOs in cells and tissues devoid of the GSNO
(S-nitrosoglutathione)-metabolizing enzyme, GSNO reductase.
GSNO, the naturally occurring S-nitrosylated derivative of
glutathione, is also widely used as a nitroso donor in cells.
In skeletal muscle, the concentration of GSNO is unknown;
however, glutathione is the most abundant low-molecular-mass
thiol, between 1 and 3 mM [13].
In the present study, we have shown for the first time
that skeletal muscle promotes formation of protein-SNO
in vivo after acute exercise. We show too that myosin is a site
for SNO formation in cultured myotubes, and, in myofibrils,
both myosin and actin are targets. S-nitrosylation promotes a
significant reduction in the Mg2+ -ATPase activity of myosin and
actomyosin, but does not change the actin–myosin affinity. When
DEANO (diethylamine NONOate) is used as an NO• donor,
neither myosin-SNO formation nor Mg2+ -ATPase inhibition
occurs, suggesting that transnitrosylation is the main pathway for
myosin-SNO formation. The inhibition of Mg2+ -ATPase activity
accompanies the formation of around four unstable SNO groups in
myosin, suggesting the possibility that a transient S-nitrosylation
in vivo might contribute to the regulation of actomyosin during
exercise or nitrosative stress.
Myosin (M r ∼
= 500 000) is the most abundant protein in skeletal
muscle, where it accounts for nearly 60 % of the total myofibrillar
proteins. It has a large impact on the demand for ATP during
muscle contraction due to its ability to convert chemical energy
derived from ATP hydrolysis into mechanical work. Myosin’s 42
cysteine residues, most of which are reduced in the native protein
[1], are distributed among two heavy chains and four light chains.
In adult skeletal muscle, the low turnover rate of myosin (halflife ∼ 30 days [2]) makes it a potential target for functionally
significant post-translational modifications, including protein
S-nitrosylation, the reaction of protein cysteine thiols with
nitrosonium ion (NO+ ) equivalents to form protein-SNOs (protein
S-nitrosothiols).
Skeletal muscle is capable of producing NO• (for a review,
see [3]), and the NO• formed during contractile activity is
not significantly metabolized by myoglobin [4]. Several authors
have postulated that either NO• or low-molecular-mass SNOs
(S-nitrosothiols) inhibit muscle force production and may lead
to S-nitrosylation of contractile proteins [5–8]. In fact, among
sarcomeric contractile and regulatory proteins, only actin has been
examined for functional changes caused by SNO formation [9].
Oxidizing or alkylating agents such as peroxynitrite and NEM (Nethylmaleimide) can alter myosin ATPase activity and its affinity
for actin in vitro [10,11]. However, it is not known whether
myosin is capable of redox-based regulation by NO• or low-mass
SNOs.
Key words: fast-twitch muscle, myosin ATPase, myotube, skinned
fibre, transnitrosylation.
Abbreviations used: biotin-HPDP, N -[6-(biotinamido)hexyl]-3 -(2 -pyridyldithio)-propionamide; CysNO, S -nitrosocysteine; DEANO, diethylamine
NONOate; DMEM, Dulbecco’s modified Eagle’s medium; DTNB, 5,5 -dithiobis-(2-nitrobenzoic acid); DTT, dithiothreitol; EDL, extensor digitorum longus;
FBS, fetal bovine serum; GdmCl, guanidinium chloride; GSNO, S -nitrosoglutathione; MHC, myosin heavy chain; MMTS, methylmethanethiosulfonate;
NEM, N -ethylmaleimide; protein-SNO, protein S -nitrosothiol; SNO, S-nitrosothiol; SNO-RAC, protein-SNO detection by resin-assisted capture; TNB− ,
2-nitro-5-thiobenzoate.
1
Present address: Division of Physiology, Department of Medicine, University of California San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0623,
U.S.A.
2
These authors contributed equally to this work.
3
To whom correspondence should be addressed (email [email protected]).
c The Authors Journal compilation c 2009 Biochemical Society
222
L. Nogueira and others
EXPERIMENTAL
Materials
Glutathione, cysteine, NEM, Sephadex G-25, HgCl2 , DTNB
[5,5 -dithiobis-(2-nitrobenzoic acid)], ATP (Grade I), EDTA,
EGTA, imidazole, Hepes, Tris, DTT (dithiothreitol), MgCl2 ,
CaCl2 and BSA were purchased from Sigma–Aldrich. GdmCl
(guanidinium chloride) was from Acros Organics, urea was
from Merck, MMTS (methylmethanethiosulfonate) was from
Calbiochem, biotin-HPDP {N-[6-(biotinamido)hexyl]-3 -(2 pyridyldithio)-propionamide} was from Pierce, DEANO sodium
salt was from Molecular Probes and thiopropyl-Sepharose 6B was
from GE Healthcare. Other salts were from Brazilian suppliers.
Mouse monoclonal antibody IgG1 against MHC (myosin heavy
chain) (MA1-16774) was purchased from Affinity Bioreagents
and anti-mouse secondary antibody IgG was from Promega.
Synthesis of SNOs
GSNO and CysNO (S-nitrosocysteine) were freshly prepared
before each experiment by mixing glutathione or cysteine
(acidified in 0.1 M HCl) with equimolar concentrations of NaNO2
[9]. After 1–2 min, the reaction was stopped by addition of 50 mM
imidazole containing 0.1 mM EDTA and titrated with KOH to pH
7.0. The concentration was determined by measuring absorbance
at 335 nm (ε335 = 900 M−1 · cm−1 for both GSNO and CysNO).
Purification of myosin, actin and skinned muscle fibres
Skeletal muscle myosin was obtained from rabbit back muscles
as described previously [14] and stored at −20 ◦C in 50 % (v/v)
glycerol at high ionic strength (0.6 M KCl and 50 mM imidazole,
pH 7). Before use, the glycerol was removed, and the protein
concentration was determined using the Biuret method, using
BSA as a standard.
Chicken skeletal muscle actin was purified [15] and
polymerized [16] essentially as described previously, but in the
absence of DTT. There was no difference in actin-activated
myosin ATPase activity in the presence or absence of DTT (results
not shown).
Skinned skeletal muscle fibres were prepared from rabbit
psoas muscle and stored at −20 ◦C in 50 % (v/v) glycerol
containing relaxing solution (152 mM potassium propionate,
20 mM imidazole propionate, pH 7.0, 6.4 mM magnesium
acetate, 5 mM EGTA and 4.4 mM ATP) and a cocktail of protease
inhibitors [17]. Glycerol was removed before use.
Cell culture
Mouse-derived C2C12 myoblasts were maintained in DMEM
(Dulbecco’s modified Eagle’s medium) supplemented with 10 %
(v/v) FBS (fetal bovine serum), 100 units/ml penicillin G and
100 μg/ml streptomycin and incubated at 37 ◦C in ambient air
with 5 % CO2 . To differentiate the cells to myotubes, DMEM with
10 % FBS was replaced by DMEM with 2 % (v/v) horse serum,
and cells were allowed to differentiate for 10 days, changing the
medium every 2 days [18].
Total protein-SNO in vivo and in myofibrils
Male Sprague–Dawley rats (300–350 g body weight) were
exercised intensely by swimming against a load (15 % of the total
body weight, attached to the tail) for 4.7 min (14 cycles of 20 s
with 10 s rest). Non-exercised rats were handled and placed in the
tank, but had no load and were not forced to swim. After exercise,
animals were killed using sodium pentobarbital (30 mg/kg of body
c The Authors Journal compilation c 2009 Biochemical Society
weight) and the EDL (extensor digitorum longus) muscles from
both hindlimbs were removed, weighed wet and homogenized
for 15 s in the dark at 0 ◦C using a Polytron PTA10 probe and
2 ml of buffer/250 mg of wet weight (1–2 % Triton X-100,
20 mM Tris/HCl, pH 7.7, 0.1 mM neocuproine, 1 mM EGTA,
1 mM EDTA, 1 mM NaVO4 , 2.5 mM sodium pyrophosphate
and 1 mM PMSF), followed by centrifugation at 500 g for
10 min. In some experiments, sodium pyrophosphate/NaVO4 was
replaced with Mg-ATP and 0.6 M KCl [19]. The supernatant was
used to measure NO2 − and protein-SNO by Saville’s method (as
described below for myosin-SNO) [20] and total protein using the
Biuret method. The experiments were approved by the Committee
for Ethics in Animal Experimentation of the Federal University
of Rio de Janeiro (Protocol # IBQM 024/031), Brazil.
Differentiated myotubes (in 100-mm-diameter×20-mm-high
dishes) were treated with 500 μM CysNO for 0–60 min at 37 ◦C in
differentiation medium. The medium was removed, and cells were
washed twice with 2 ml of PBS supplemented with 0.1 mM EDTA
(PBS-EDTA), followed by lysis and the SNO-RAC (protein-SNO
detection by resin-assisted capture) assay described below.
Small bundles of skinned muscle fibres (50 mg of wet weight)
were treated with 0.5 ml of GSNO (0.01–1 mM) for 30 min at
30 ◦C in relaxing solution. Unreacted GSNO was removed by
washing three times with 0.5 ml of relaxing solution, and proteins
were solubilized in 1 ml of buffer A (25 mM Hepes, pH 7.7,
4 mM EGTA, 4 mM MgCl2 , 2 mM ATP and 0.1 mM neocuproine)
containing 0.3 M KCl. The biotin-switch assay was carried out as
described below.
Quantification of myosin-SNO
Myosin S-nitrosylation was carried out using fresh GSNO at
30 ◦C in the ionic condition that promotes the classical Mg2+ ATPase activity of myosin [21], but without ATP. This Mg2+ buffer
contained 50 mM imidazole, pH 7, 50 mM KCl, 0.2 mM EDTA
and 5 mM MgCl2 . To start the reaction, GSNO (final concentration
1 μM–8 mM) was added to the Mg2+ buffer containing 4 mg of
purified myosin in a final volume of 2 ml. After 15–120 min,
5 ml of cold water was added, followed by centrifugation at 2700 g
for 15 min, and the pellet (0.1 ml) was solubilized in 1 ml of
high-ionic-strength buffer (0.6 M KCl and 50 mM imidazole, pH
7; one wash). The washing was repeated again (two washes)
or up to five times to remove unreacted GSNO. After two
to five washes, the solubilized pellet was covered with foil to
diminish SNO decomposition by light [22]. Myosin-SNO was
determined colorimetrically on 40 μl (∼ 40 μg) of myosin in a
final volume of 0.5 ml. NO2 − was measured using the Griess
reagent with and without 3 mM HgCl2 , to decompose SNO [20],
followed by centrifugation at 2000 g for 10 min to avoid light
scattering from precipitated protein. The difference between the
absorbance of the supernatant with and without HgCl2 addition
was taken to represent the amount of protein-SNO. Standard
curves were constructed using NaNO2 , and protein concentration
was determined using the method of Lowry et al. [23].
Detection of protein-SNO by the biotin-switch assay and by the
SNO-RAC method
Detection of protein-SNO was carried out in myofibrillar
proteins from rat EDL muscle (as described above), C2C12
myotubes (2 mg of protein), skinned skeletal muscle fibres (50 mg
of wet weight) and in purified myosin (2 mg/ml) using the
methods described in [22,24] with some modifications. When
an exogenous RSNO donor was used, the excess was removed
after the S-nitrosylation step by washing with PBS-EDTA (for
Inhibition of myosin by S-nitrosylation
223
myotubes) or with relaxing buffer (for skinned muscle fibres) or
with buffer B (20–100 mM Tris/HCl, pH 7.7, 1 mM EDTA and
0.1 mM neocuproine) followed by centrifugation at 2000 g for
10 min (for purified myosin). After the washing procedure, either
the SNO-RAC (for myotubes) or the biotin-switch assay (for EDL
muscles, skinned fibres and myosin) was carried out as described
below.
To detect myosin-SNO in C2C12 myotubes, cells were lysed
in ice using 400 μl of RIPA buffer (50 mM Tris/HCl, pH 8.0,
150 mM NaCl, 1 % Nonidet P40, 0.5 % sodium deoxycholate
and 0.1 % SDS) supplemented with 0.5 mM PMSF and a cocktail
of protease inhibitors followed by centrifugation at 12 000 g for
10 min at 4 ◦C. The supernatant was allowed to react with 1.6 ml
of blocking buffer (buffer B plus 2.5 % SDS and 20 mM MMTS)
for 20 min at 50 ◦C with frequent vortex-mixing followed by 6 ml
of pre-chilled acetone and precipitation for 20 min at −20 ◦C.
Some samples were photolysed in 1.6 ml of buffer B plus 2.5 %
SDS for 10 min using a mercury vapour lamp as described in [22]
and allowed to react with 20 mM MMTS, followed by acetone
precipitation as above. After centrifugation at 2000 g for 10 min
at 4 ◦C, the pellet was washed with acetone and resuspended in
0.5 ml of buffer B plus 1 % SDS, 20 mM ascorbate and 40 μl
of thiopropyl-Sepharose 6B and allowed to react for 2 h at room
temperature (25 ◦C) under rotation to capture the SNO proteins by
SNO-RAC [24]. All of the steps were carried out in the absence
of light. The resin was washed five times with 1 ml of buffer B
plus 1 % SDS and twice with buffer B diluted 1:10 with 1 % SDS
(buffer B/10 SDS) followed by elution with buffer B/10 SDS plus
2 % 2-mercaptoethanol for 1 h at room temperature. The proteins
were separated by SDS/PAGE (4–15 % gels) and transferred on
to a nitrocellulose membrane or stained with Coomassie Blue
R. Myosin input or the pull-down was immunoblotted using the
monoclonal antibody for MHC (1:2000) followed by secondary
anti-mouse IgG (1:10 000), and MHC was detected using ECL®
(enhanced chemiluminescence) Plus (GE Healthcare).
For the myofibrillar proteins from EDL muscle (see above)
or skinned fibres, 2.5 % SDS and 0.1 M MMTS were added to
the supernatant and then proteins were precipitated with acetone
after 1 h. In both cases, the precipitated proteins were solubilized
in buffer B plus 1 % SDS, and SNO sites were switched to
biotin tags by incubating with ascorbate and biotin-HPDP for
2 h at room temperature. For purified myosin, solubilization was
in blocking buffer containing 2.5 % SDS and 40 mM MMTS. In
these cases, pull-down was not performed and the proteinSNO was detected in the blot by total thiobiotinylation using
streptavidin–horseradish peroxidase (1:8000 dilution) followed
by the ECL® reaction as described above and comparison with
the total protein in the Coomassie Blue R staining using Image
Quant 5.2 (Molecular Dynamics).
at 3000 g for 25 min to remove the charcoal. The supernatants
were analysed colorimetrically to determine Pi release. The time
course of hydrolysis was linear in all cases. To investigate the
effects of NO• on Mg2+ -ATPase activity, myosin was treated with
DEANO (50 μM–2 mM) for 30 min in Mg2+ buffer, followed by
dilution (1:10) in Mg2+ buffer containing 3 mM ATP. A control
experiment showed that the half-time for decay of 1 mM DEANO
at pH 7 was 4 min, so that 30 min was adequate for conversion of
NO• into the S-nitrosylating species NO2 • and N2 O3 .
Actin-activated myosin ATPase activity was measured after
myosin S-nitrosylation followed by two washes and resuspension.
Hydrolysis was started by adding 50 μl of modified myosin (final
concentration 0.1–0.2 mg/ml) to ATPase medium in the presence
of actin (60 mM KCl, 50 mM imidazole, pH 7, 5 mM MgCl2 ,
1 mM EGTA, 3 mM ATP and 0–0.5 mg/ml actin). After 10 min,
the reaction was interrupted as described above but with 1 M
perchloric acid, and Pi was measured as before. Actin-activation
of myosin was calculated by subtracting the specific activity of
myosin from the total ATPase activity in the presence of actin.
Kinetic parameters (K m and V max ) were determined by fitting the
data with a rectangular hyperbola.
Myosin and actomyosin ATPase activity
SNOmyo = SNOmyoo + (SNOmyomax
After treating myosin with GSNO followed by two to five washes
and resuspension as described above, 0.1 ml (∼ 0.1 mg of protein)
was used to measure myosin ATPase activity in a final volume
of 1 ml in the presence of 3 mM ATP at 30 ◦C. This procedure
ensured a dilution of the original GSNO by a factor of ∼ 13 000,
after two washes. The samples were pre-incubated at 30 ◦C for
2 min at a final concentration of 0.05–0.2 mg/ml myosin either
in Mg2+ buffer (for Mg2+ -ATPase) or in 0.6 M KCl, 5 mM
EDTA and 50 mM imidazole, pH 7, for K+ -EDTA-ATPase, or
in 50 mM KCl, 50 mM imidazole, pH 7, 0.2 mM EDTA and
5 mM CaCl2 for Ca2+ -ATPase activity. Hydrolysis was started
by adding 3 mM ATP and interrupted by activated charcoal
suspended in 0.1 M perchloric acid followed by centrifugation
Detection of myosin thiol groups
The number of free thiol groups in native and unfolded myosin was
measured in the same buffers as those used for S-nitrosylation (see
above), but without GSNO. Myosin (0.25 mg/ml) was incubated
with 500 μM DTNB for 2 h at pH 7 and 25–30 ◦C, and the time
course of TNB− (2-nitro-5-thiobenzoate) release was analysed
at 412 nm (ε412 = 13600 M−1 · cm−1 ) [25]. The free thiol groups
of myosin-SNO could not be analysed with DTNB, since TNB−
decomposes RSNO, overestimating the real number of free thiols
[26].
Alkylation of myosin
In the experiments of Figure 5(D)–5(E), to block the most reactive
thiol groups, myosin (5 mg/ml) was treated with NEM (1 mM) for
30 min at 25 ◦C in Mg2+ buffer, and the reaction was stopped
with 10 mM DTT, followed by extensive dialysis to remove
excess NEM and DTT [10]. The remaining number of accessible
thiols was determined using DTNB in the presence (native) and
absence of GdmCl (unfolded myosin). During NEM treatment,
the ATPase activity in the presence of Mg2+ buffer showed the
expected activation at short times followed by inhibition at longer
times, and, in the presence of K+ EDTA buffer, only inhibition was
observed. This pattern is characteristic of alkylation of the more
reactive thiols in each head, including Cys707 and Cys697 [10].
In Figure 5(D), the GSNO concentration-dependence of Snitrosylation in non-alkylated and alkylated myosin was fitted
using a rectangular hyperbola (eqn 1):
× [GSNO])/(K 0.5 + [GSNO])
(1)
where SNOmyo and SNOmyoo represent S-nitrosylation
determined by Saville’s method in myosin treated or not with
GSNO, SNOmyomax is the maximum myosin-SNO formation; and
K 0.5 is the GSNO concentration for half-maximum myosin-SNO
formation.
Statistical analyses
The experimental results are presented as means +
− S.E.M. For
comparison between two groups, Student’s t test was used, paired
or unpaired as required. For multiple comparisons, a one-way
c The Authors Journal compilation c 2009 Biochemical Society
224
L. Nogueira and others
rats (0.05 +
− 0.03 nmol of RSNO/mg of protein compared with
0.30 +
− 0.16 nmol of RSNO/mg of protein respectively; P < 0.05,
n = 8; Figure 1B). Collectively, these data point to an increase
in NO production and concomitant S-nitrosylation in fast-twitch
skeletal muscle as a result of intense exercise. In a second
experiment, we utilized the biotin-switch technique to assess the
speciation of protein SNOs. As above, rats were subjected to
a session of intense exercise or not, and the EDL muscles were
processed to obtain a fraction enriched in myofibrillar proteins. Snitrosylation of a ∼ 200 kDa band, prominent among myofibrillar
proteins, increased in high-intensity exercise (Figures 1C and 1D).
Myosin is the major target of S-nitrosylation in myotubes and
in skinned muscle fibres
Figure 1 High-intensity exercise and protein-SNO content in myofibrillar
proteins from rat EDL muscle
Myofibrillar proteins (EDL muscle) from rats that were or were not acutely exercised at high
intensity were removed, purified and analysed by Saville’s method (see the Experimental section)
for (A) free NO2 − and (B) total SNOs (Hg2+ -displaceable NO2 − ) quantification. Results are
means + S.E.M.; n = 8–9. *P < 0.05 (unpaired Student’s t test). (C) Non-reducing SDS/PAGE
(8 % gels; Coomassie Blue staining) for total protein content and Western blotting with anti-biotin
for total biotinylated proteins after the biotin-switch technique on samples from a second
experiment with exercised and non-exercised rats. Molecular masses are indicated in kDa.
(D) Densitometric analysis of the ∼ 200 kDa band of biotinylated proteins from Western blot
development and Coomassie Blue staining shown in (C). Results are means + S.E.M.; n = 4.
*P < 0.05 (unpaired Student’s t test).
ANOVA followed by the Tukey test or a two-way ANOVA was
used, as indicated. Analyses were carried out using GraphPad
Prism 4 software and P < 0.05 was considered to represent a
significant difference.
The majority of proteins of skeletal muscle are those associated
with the thick filaments (e.g. myosin) and thin filaments (e.g.
actin, tropomyosin and troponin complex). Among these proteins,
we speculated that MHC was the identifiable target of exerciseinduced S-nitrosylation (Figure 1D). To positively identify
myosin as a target of S-nitrosylation, differentiated C2C12
myotubes were treated with CysNO and were subsequently
assayed by SNO-RAC, a newly-described assay for S-nitrosylated
proteins that, compared with the biotin-switch assay, detects
more sensitively high-molecular-mass SNO-proteins [24]. The
reaction of myotubes with CysNO resulted in S-nitrosylation of
a ∼ 200 kDa protein (Figure 2A, lane 2; arrow indicates the
MHC that was prominent among total cellular proteins), and it
disappeared when CysNO was omitted or when pre-photolysis
was applied. Western blotting identified S-nitrosylated MHC
(myosin-SNO) in pull-downs from CysNO-treated C2C12 cells
(Figure 2B); the specificity of the assay was confirmed by both
ascorbate-dependence and UV pre-photolysis (which results in
homolysis of the S-NO bond [22]). A time course revealed that
myosin-SNO persisted for at least 60 min following addition of
CysNO, suggesting that myosin-SNO may be more stably Snitrosylated than the majority of intracellular protein-SNOs [24].
We also investigated S-nitrosylation of skinned muscle fibres,
which contain intact myofibrillar proteins, but lack many cytosolic
constituents. These permeabilized fibres should be accessible to
GSNO, which, unlike CysNO, does not readily cross the cell
membrane and did not promote myosin-SNO formation in intact
C2C12 cells (results not shown). On the basis of the well-known
distribution of proteins within these fibres [27], MHC and actin
were identified as the primary targets of this physiologically
relevant low-mass SNO (Figure 2D). An examination of total
thiol biotinylation revealed that free cysteine residues in many
other proteins were not reactive towards GSNO (Figure 1C).
Collectively, these data demonstrate that MHC is a major cellular
target of S-nitrosylation.
RESULTS
High-intensity exercise increases protein S-nitrosylation in EDL
muscle
S-nitrosylation regulates the Mg2+ - but not K+ - or Ca2+ -ATPase
activities of myosin
As an initial investigation, we subjected rats to a session of
intense exercise in order to determine whether these conditions
would promote protein S-nitrosylation within fast-twitch EDL
muscle. We first measured NO2 − and RSNO in a fraction enriched
with myofibrillar proteins. Exercise resulted in an approx. 2fold increase in NO2 − from levels found at rest (1.07 +
− 0.20
nmol of NO2 − /mg of protein compared with 0.52 +
0.06
nmol
−
of NO2 − /mg of protein respectively; P < 0.05, n = 9; Figure 1A),
suggestive of exercise-induced NO• production. SNOs, detected
as Hg2+ -displaceable NO2 − , were barely detectable in resting
muscle, but were significantly increased in muscle from exercised
There are two recent reports of S-nitrosylation on cardiac myosin
light chains in vitro [28,29], and, in another study, the cardiac
MHC was S-nitrosylated in intact hearts following ischaemic
perfusion [30]. There are no reports, however, that correlate
skeletal muscle myosin S-nitrosylation with a change in function.
Myosin exhibits pronounced differences in activity under
various ionic conditions, presumably reflecting conformational
changes associated with the binding of K+ , Mg2+ or Ca2+ to
the motor domain [31] as well as the stability of the cationATP complex at the catalytic site. Using these three classical
ionic conditions, we first explored the effects of GSNO (15 min of
c The Authors Journal compilation c 2009 Biochemical Society
Inhibition of myosin by S-nitrosylation
Table 1
225
GSNO inhibits only Mg2+ -ATPase activity of myosin
Myosin was incubated with or without GSNO in each of the specific ion buffers for 15 min,
followed by removing reaction products and excess GSNO by one wash (see the Experimental
section). Pellets were solubilized in high-ionic-strength solution (0.6 M KCl and 50 mM
imidazole, pH 7). ATPase activity was measured by adding modified myosin to Mg2+ , Ca2+ or
K+ -EDTA buffer (see the Experimental section), followed by 3 mM ATP. *P < 0.05 compared
with unmodified myosin (paired Student’s t test). Results are means +
− S.E.M. for the number of
preparations shown in parentheses in the first column. Data obtained with lower concentrations
2+
of GSNO in Mg buffer (60 min) are shown in Figure 2(A).
ATPase activities (nmol of Pi · mg−1 · min−1 )
Figure 2 Targets of S-nitrosylation in skinned muscle fibres and in C2C12
myotubes
(A) Reducing SDS/PAGE (4–12 % gels) stained with Coomassie Blue shows the C2C12
myotubes lysate after the SNO-RAC assay. Arrow indicates the MHC band. Molecular masses
are indicated in kDa. (B) Representative immunoblot using monoclonal antibody for MHC
of C2C12 myotubes lysate after the SNO-RAC assay. (C) Time course of the reaction between
C2C12 myotubes and 500 μM CysNO followed by the SNO-RAC assay. Immunoblot shows
monoclonal antibody for MHC. (D) Non-reducing SDS/PAGE (10 % gels) of skinned muscle
fibre proteins after the biotin-switch technique as described in the Experimental section,
either stained with Coomassie Blue or after Western blotting biotinylated proteins following
chemiluminescence development for 5 min (PC) or 15 min (Negative control and 0–1 mM
GSNO). Protein identification was based on molecular masses of SigmaMarkerTM Wide Range
(6.5–205 kDa) correlated with muscle proteins in skinned fibres [27] and with the mobility
of purified myosin, S1, TnI (troponin I) and TnC (troponin C). MLC 2, myosin light chain 2;
Negative control when proteins were reduced by DTT treatment followed by MMTS and the
biotin-switch technique; PC, positive control for biotinylation when MMTS was replaced with
DTT followed by the biotin-switch protocol; TM, tropomyosin; TnT, troponin T.
incubation followed by a single wash) on myosin ATPase
activity in vitro (Table 1). Interestingly, the Mg2+ -ATPase activity
of myosin was inhibited by up to ∼ 25 % upon treatment with
GSNO (Table 1 and Figure 3A, open circles), whereas neither
Condition
0 mM GSNO
3 mM GSNO
Activity after GSNO treatment
( % of native myosin)
K+ -EDTA (3)
Ca2+ (3)
Mg2+ (6)
1054 +
− 80
860 +
− 91
40 +
−6
1089 +
− 31
860 +
− 93
29 +
− 3.5*
103.7 +
− 4.9
100.1 +
− 3.0
75.5 +
− 9.3*
the Ca2+ - nor the K+ -EDTA-ATPase activity was altered (Table
1). Myosin was also dose-dependently S-nitrosylated by GSNO
under identical conditions (Figure 3B and Supplementary Figure
S1A at http://www.BiochemJ.org/bj/424/bj4240221add.htm),
consistent with the idea that modifications of cysteine thiol(s)
on myosin were responsible for its inhibition. Quantification
of myosin S-nitrosylation using the Saville assay revealed that
maximal inhibition of Mg2+ -ATPase activity correlated with
S-nitrosylation of four to six cysteine thiols per myosin (Figure
3D).
We also found that a longer incubation with GSNO (60 min)
greatly enhanced both its inhibition of Mg2+ -ATPase activity
(Figure 3A, inset) and myosin S-nitrosylation (Figure 3C and
Supplementary Figure S1B). In fact, under these conditions,
significant inhibition was observed with as little as 5 μM GSNO,
which approaches physiological levels of this endogenous SNO.
However, the nitric oxide donor DEANO neither inhibited
nor markedly S-nitrosylated myosin (Figures 3A and 3B and
Supplementary Figure S1A). These data suggest that myosin
regulatory thiol(s) react relatively slowly with GSNO and that
transnitrosylation between GSNO and myosin, but not the
reaction of NO• (a possible product of GSNO decomposition),
was responsible for myosin S-nitrosylation in vitro. A similar
result was reported for the cardiac ryanodine receptor RyR2,
which, unlike the skeletal muscle protein RyR1, is also
S-nitrosylated by GSNO, but not by NO• [32]. GSNO appears
to be a potential physiologically relevant mediator of myosin
S-nitrosylation, although these data cannot rule out the possibility
that oxidative reactions between NO• and myosin, which are
apparently disfavoured even under aerobic conditions in vitro,
also govern endogenous myosin S-nitrosylation (Figures 1C
and 1D).
Actin–myosin affinity is not affected by GSNO
Myosin’s ability to bind to actin during the cross-bridge
cycle, increasing the ATPase activity, is fundamental to its
chemomechanical role in muscle. To determine whether the
maximal ATPase activation or actin–myosin affinity is affected
by GSNO, myosin was allowed to react with 0.1–4 mM GSNO
in Mg2+ buffer for 15 min at 30 ◦C followed by two washes. The
actin-activation of modified myosin was induced by adding it to
Mg2+ buffer containing ATP and various concentrations of actin.
Figure 4(A) shows an example of the actin-activation of
myosin prepared as described in Figure 3(A) (two washes,
open circles). On average, the maximal activity decreased in a
c The Authors Journal compilation c 2009 Biochemical Society
226
Figure 3
L. Nogueira and others
Inhibition and S-nitrosylation of myosin
(A) Myosin (2 mg/ml) was incubated with GSNO for 15 min (䊉, 䊊; n = 6) or 60 min (inset; n = 5) or with DEANO for 30 min (䉱; n = 3) in Mg2+ buffer at 30 ◦C. Mg2+ -ATPase activity was then
measured by diluting myosin (1:10) in Mg2+ buffer containing 3 mM ATP (䊉, 䉱, 䊏), or after removing unreacted GSNO by two washes (䊊, see the Experimental section). (B) Myosin treatment
with either GSNO or DEANO for 15 and 30 min respectively, followed by the biotin-switch assay (see the Experimental section). (C) Time-course of myosin-SNO formation from GSNO. Myosin was
allowed to react with GSNO (100 μM) up to 120 min, followed by the biotin-switch assay and myosin-SNO detection as in (B). (D) Stoichiometry of myosin S-nitrosylation by GSNO compared with
the ATPase effect. The concentrations of GSNO and conditions of the reaction (15 min, removal of the unreacted GSNO by two washes) were the same as in (A, 䊊). Results are means +
− S.E.M.;
n = 6.
concentration-dependent manner, approaching a plateau value
(≈ 20 % inhibition) after treatment with 2 mM GSNO (Figure 4B). There was no effect on myosin’s affinity for actin
(Figure 4A), and thus the effect of GSNO on the actomyosin
ATPase can be attributed to its effect on the catalytic cycle of
myosin alone.
Stability of myosin-SNO groups and inhibition of the ATPase
A number of different proteins that can be S-nitrosylated in vitro
have been shown to lose the SNO groups subsequently, at different
rates and by different mechanisms, including release of NO• to
the medium and transfer of the nitroso group to other proteins.
Although it was not our objective to explore possible routes of
myosin-SNO decomposition, we did compare the stability of the
S-nitrosylated cysteine residues with the inhibition of ATPase
activity, using for convenience a relatively short incubation time
(15 min) and a wide range of GSNO concentrations. As described
c The Authors Journal compilation c 2009 Biochemical Society
below, these same experiments were useful for eliminating the
possibility of overestimating SNO groups because of contaminant
GSNO trapped in the myosin pellet.
Figure 5(A) shows that the Mg2+ -ATPase activity is inhibited
by GSNO to the same extent after simple dilution (1:10) and after
one and two washes (P>0.05). During the washing procedure,
the GSNO dilution in the medium is significantly increased
from 10-fold (no washing), to 440-fold (one wash) and approx.
13 000-fold (two washes). Thus it is unlikely that contaminant
GSNO is responsible for inhibition of the ATPase. However, after
five washes, the inhibitory effect of GSNO is lost completely
(Figure 5A).
To understand why the inhibitory effect is lost after extensive
washing, we quantified myosin-SNO groups after two and five
washes, and found a marked reduction after five washes (Figure
5B, P < 0.05). Thus with 1 mM GSNO, for example, 5.0 +
− 0.3
mol of SNO/mol of myosin (two washes) is decreased to 1.0 +
− 0.1
mol/mol after five washes, and there is a corresponding loss of the
Inhibition of myosin by S-nitrosylation
227
of SNO groups: a more labile one (about four SNO groups per
myosin), which includes some groups that inhibit the ATPase; and
a population that does not affect the activity, but is more stable.
These results reinforce and extend the conclusion drawn from
Figure 3(D): inhibition arises from S-nitrosylation of no more
than four to six labile cysteine residues, and possibly fewer.
The more stable SNO groups are located in less accessible regions
Figure 4
S-nitrosylation of myosin: effect on actomyosin
(A) Actin activation by myosin reacted with 0–4 mM GSNO for 15 min followed by two
washes. Activation of ATP hydrolysis was measured by adding modified or non-modified
myosin (0.1 mg/ml) to Mg2+ buffer (see the Experimental section) containing 3 mM ATP
and actin (0–0.5 mg/ml). The activity of myosin alone was subtracted from each point. The
apparent affinity for actin (K m , see the Experimental section) was not modified by GSNO
(P > 0.05, one-way ANOVA; n = 3). Results are from a typical experiment. (B) GSNO treatment
significantly decreases maximal actin activation (V max ) from the control value (423 +
− 83 nmol
of Pi · mg−1 · min−1 ). Results are means + S.E.M.; n = 3. *P < 0.05 (one-way ANOVA).
inhibitory effect on activity (Figure 5A). For both sets of data, the
number of SNO groups remaining after washing is significantly
different from the untreated sample (i.e. 0 mM GSNO, Figure 5B,
P < 0.05). This observation suggests that the loss of myosin-SNO
groups after washing is related primarily to the lability of some
of these groups, and not merely to the removal of contaminant
GSNO. We infer from the data in Figures 5(A) and 5(B) that
S-nitrosylation of myosin by GSNO produces two populations
The stability of protein-SNO groups is associated with the
environment of the SNO group in the molecule. The nature and
proximity of neighbouring amino acid residues can increase or
decrease its rate of decomposition [33]. To characterize further
the more stable myosin-SNO groups formed after reaction with
GSNO, NEM was used to block the more reactive thiols and Snitrosylation was measured after reaction with GSNO followed
by five washes, which eliminates the more labile SNO groups
(Figures 5C and 5D).
As can be seen in Figure 5(D), the treatment with 1 mM NEM
significantly reduced the stable myosin-SNO groups formed at
high GSNO concentrations (from 4.8 +
− 0.1 to 2.5 +
− 0.5 mol/mol
in 8 mM GSNO for native and NEM–myosin respectively;
P < 0.05), but NEM had no effect at low GSNO concentrations
(1.7 +
− 1.0 mol/mol in 1 mM GSNO for native and
− 0.2 and 2.3 +
NEM-myosin respectively). These data show that between two
and three cysteine residues (filled squares in Figure 5D) that
contribute to the more stable group of SNO were not alkylated,
suggesting that these cysteine residues are reactive to GSNO, but
not to NEM. The two cysteine residues lost to S-nitrosylation by
their reaction with NEM (open squares at 8 mM GSNO in Figure
5D) do not affect the ATPase activity, since the alkylated myosin
was washed thoroughly beforehand. Since S-nitrosylation with
GSNO followed by two washes produces four to six myosinSNO groups (Figure 5B), and two to three of them belong to the
stable group, one or both of the other two labile SNO groups must
promote the inhibitory effect of S-nitrosylation.
To relate the stable SNO groups with the reactivity and
availability of thiols in myosin, DTNB was used after the alkylation step to measure the number of free thiol groups. The
protein was alkylated in the folded state with NEM, and free
thiols were determined in the folded (without GdmCl) and
unfolded (with GdmCl) protein (Figures 5C and 5E). Under
both conditions, NEM-myosin had a decreased number of free
thiol groups relative to non-alkylated myosin (Figure 5E). The
difference in free thiol content of native myosin and NEM–myosin
was 10 +
− 2 mol of thiol/mol of myosin when measured in the
folded state and 11 +
− 2 mol of thiol/mol of myosin when measured
in the unfolded state (Figure 5E; P > 0.05). This result shows
that alkylation reaches the more accessible thiols in myosin, since
the alkylation step in both cases was carried out in the native
state. Thus it seems likely that, among these ten or eleven alkylated
cysteine residues, located in regions more exposed to the solution,
one or two can form stable SNO groups and the eight to ten
remaining thiols can form more labile SNO groups. We infer that
some of these latter more exposed residues are responsible for the
inhibition of ATPase activity by S-nitrosylation.
DISCUSSION
The principal findings of the present study are that myosin can
be extensively S-nitrosylated both in vivo and in vitro and the
physiological ATPase activity can be affected by a few labile
SNO groups that inhibit Mg2+ -ATP hydrolysis, but do not alter
the affinity of myosin for actin. Evidence is presented to show
that the mechanism is one of transnitrosylation rather than a direct
effect of NO• .
c The Authors Journal compilation c 2009 Biochemical Society
228
Figure 5
L. Nogueira and others
Stability of myosin SNO groups with washing, alkylation and GdmCl
(A) Loss of inhibitory effect on Mg2+ -ATPase activity by extensive washing. GSNO treatment (15 min in Mg2+ buffer) was followed by 1:10 dilution without washing (䊉; n = 3) and one (; n = 6),
two (䊊; n = 3) or five (䊐; n = 3) washes. Results are means +
− S.E.M. Only the curve with five washes is significantly different from the others (P < 0.05, two-way ANOVA test). (B) Myosin was
allowed to react with GSNO as in (A), followed by two or five washes. Myosin-SNO content was determined by Saville’s method. *P < 0.05 compared with 0 mM GSNO; **P < 0.05 compared with
five washes (one-way ANOVA). (C) Scheme showing the alkylation of the most reactive thiol groups in myosin by NEM (see the Experimental section). Alkylated (䊏) and non-alkylated (䊐) myosins
were purified by extensive dialysis and, in (D), treated with GSNO (0.25–8 mM) for 15 min at 30 ◦C followed by five washes, and myosin-SNO content was measured by Saville’s method. Results are
means +
− S.E.M.; n = 3. *P < 0.05 compared with non-alkylated myosin (paired Student’s t test). A rectangular hyperbola was used to fit the results (eqn 1). There is a significant difference in the
+
maximum number of myosin-SNO bonds in alkylated (6.5 +
− 0.7 mol of SNO/mol of myosin) compared with non-alkylated myosin (2.4 − 0.1 mol of SNO/mol of myosin; P < 0.05, paired Student’s
t test). In (E), alkylated and non-alkylated myosins were treated with DTNB (0.5 mM) for 2 h in the absence (native, folded protein) or presence of 2 M GdmCl (unfolded protein), and free thiol groups
were measured by absorption at 412 nm. Results are means +
− S.E.M.; n = 3. *P < 0.05 (paired Student’s t-test) compared with control or control + GdmCl.
We did not undertake a detailed study of S-nitrosylation in vivo,
but there was a clear increase in total RSNO and protein-SNO
levels of a band corresponding to the molecular mass of MHC in
rat fast-twitch muscle following intense exercise (Figure 1). An increase in protein-SNO levels with exercise is consistent with previous reports showing that stimulation increases intracellular NO•
[4,34] as well as NO• release [35] from skeletal muscle by activation of the Ca2+ -dependent nitric oxide synthases. Similar results
have been obtained in experiments with cardiac myocytes [36].
Even though many muscle proteins have their function changed
by S-nitrosylation in vitro, it has never been demonstrated that
S-nitrosylation occurs in myofibrils. In the present study, we
show that in both C2C12 myotubes, a well-known differentiated
cell type that expresses myofibrillar proteins [18], and skinned
fibres, where the sarcolemma is disrupted by EGTA and there
is extensive loss of cytosolic proteins [37], a pharmacological
c The Authors Journal compilation c 2009 Biochemical Society
challenge by CysNO or GSNO leads to S-nitrosylation primarily
of myosin and actin. The data show that many other proteins
are susceptible to thiol modification, but are not S-nitrosylated
(Figure 2). The evident S-nitrosylation of actin in skinned muscle
fibres is consistent with reports showing actin-SNO in non-muscle
cells [38] and in homogenates of heart muscle [30], suggesting
that this protein may be a ubiquitous site for S-nitrosylation of
the cytoskeleton. However, other proteins such as tropomyosin
and troponin (C+T+I), which have cysteine residues, are not
visibly affected up to 0.5 mM GSNO (Fig. 2D). Together with
the observation that myosin can be readily S-nitrosylated in
skinned fibres and in C2C12 myotubes, these experiments raise
the possibility of a (patho)physiological role for myosin Snitrosylation in vivo.
One way in which this could occur would be via a direct
reaction of NO• or its auto-oxidation products with myosin
Inhibition of myosin by S-nitrosylation
thiols. This route appears to be unlikely, since myosin-SNO did
not form when NO• was provided by DEANO decomposition
(Figures 3A and 3B). Our experiments with GSNO and CysNO,
however, show that myosin is susceptible to transnitrosylation.
Considering that muscle fibres contain an ample supply of
glutathione, an important physiological anti-oxidant and a GSNO
precursor, an increase in NO• produced by activation of nNOS
(neuronal nitric oxide synthase) or eNOS (endothelial nitric
oxide synthase) in vivo, could lead to an increase in GSNO
followed by transnitrosylation of susceptible myosin thiols. Under
pathophysiological conditions, such as endotoxaemia or obesity,
an additional source of NO• would be provided by induction of
iNOS (inducible nitric oxide synthase) [39]. The concentration
of GSNO in muscle is not known, but estimates of intracellular
GSNO concentration in other cells vary widely, from ∼ 2 μM
(in cerebellum [40]) to 50 μM (in aorta [41]) and even 200 μM (in
mitochondria [42]). However, GSNO is generally thought to
be compartmentalized, meaning that higher values may be
encountered locally, and homogenization will certainly disrupt
the normal distribution [43,44]. A clue to its physiological
concentration in the cytosol is provided by the K m of GSNO
reductase (4.8 μM in humans, 20 μM in mice), which is important
in GSNO metabolism [45,46].
A legitimate concern in evaluating this proposal is the question
of whether NO• can ever reach significant concentrations in
muscle fibres, given the presence of oxymyoglobin, a putative
NO• scavenger. Recent experiments with intracellular fluorescent
probes for NO• show clearly that it increases substantially in
stimulated skeletal and cardiac cells [4,34,36], and so does
protein-SNO [30,39]. Thus physiological levels of oxymyoglobin
do not eliminate these responses, and in fact the idea that
myoglobin will always serve as a sink for NO• has been questioned
[47].
A second concern that is relevant to the possible significance of
myosin S-nitrosylation in vivo is that of O2 concentration. In some
proteins, such as the skeletal muscle ryanodine receptor RyR1
[33], regulation of a critical thiol by S-nitrosylation occurs only
at low O2 concentrations characteristic of the in vivo situation,
whereas in other proteins (e.g. the cardiac receptor RyR2 [32]), the
O2 tension has no effect on SNO formation. All of our experiments
in vitro were carried out in solutions equilibrated with atmospheric
O2 , well above the level expected in living muscle. Thus it will be
important to examine the dependence of myosin S-nitrosylation
on O2 tension before concluding that it has a role in vivo.
The thiol groups of protein cysteine residues have different
reactivities depending on thiol pK a and three-dimensional
structure, which can increase or decrease the tendency for Snitrosylation [33] and denitrosylation [48]. One of our goals
was to determine whether S-nitrosylation reveals different classes
of thiols in myosin, and we started by evaluating its effects on
myosin ATPase activities. Myosin contains 42 cysteine residues,
and different populations of thiol groups have been described,
based on their reactivity to thiol-group-specific reagents and
the resulting changes in ATPase activity [1,10,31]. In our
experiments, maximal inhibition of the Mg2+ -ATPase in 15 min
was attained after S-nitrosylation of four to six cysteine residues
(Figure 3D), apparently with no contribution from the other
SNO groups formed by transnitrosylation at higher GSNO
concentrations. When the same S-nitrosylation was carried out
in Ca2+ or K+ -EDTA buffer, there was no effect whatsoever on
myosin activity in these buffers (Table 1 and data not shown).
The so-called ‘essential’ thiols Cys707 (SH1) and Cys697 (SH2)
are among the first to react with classical thiol-group-specific
reagents, and provide the basis for these reagents’ inhibiting the
K+ -EDTA ATPase activity as well as for first activating and then
229
inhibiting the Ca2+ and Mg2+ -ATPase activities [10]. The fact that
this profile does not appear when myosin is S-nitrosylated may be
an indication that these two cysteine residues are not susceptible
to attack by GSNO.
Our experiments with actomyosin provide additional evidence
that S-nitrosylation may not affect SH1 or SH2. Alkylation
of myosin with NEM enhances the affinity between actin and
myosin, even in the presence of Mg2+ -ATP, and reduces the
maximal actin activation [49]. Once again, the profile does
not match: after S-nitrosylation, maximum actin activation is
decreased, but actomyosin affinity is maintained (Figures 4A
and 4B). We infer that the time- and concentration-dependent
inhibition of the Mg2+ -ATPase may be a consequence of S-nitrosylation of some cysteine residue(s) other than SH1 or SH2. It
is also possible that SH1 and SH2 are S-nitrosylated without
functional impairment. In this regard, it is of interest that
Sakamoto et al. [50] have described a thiol-group-specific reagent
that reacts with a single residue in each myosin head, without
modifying either SH1 or SH2. The effect on ATPase activities
is very different from the pattern observed when SH1 or
SH2 is modified. Ultimately, the fate of SH1 and SH2 during
S-nitrosylation will have to be settled using mass spectrometry.
We have tried repeatedly to identify S-nitrosylated cysteine
residues of myosin peptides using this method, but so far have
been unable to achieve satisfactory coverage of the MHC.
Our second approach to assignment of myosin-SNO to different
thiol populations was based on their different stabilities to washing
with deionized water in the absence of chelating agents. This
simple procedure leads to loss of a more labile population of
SNO groups, some of which affect the Mg2+ -ATPase activity.
First, it is clear that some fraction of the four to six SNO groups
that survive after two washes promotes the inhibitory effect on
the Mg2+ -ATPase activity (Figures 5A and 5B). However, after
five washes, the inhibitory effect is abolished completely, and
the myosin S-nitrosylation is drastically reduced to 2–3 mol/mol.
This suggests that there are two groups of SNOs in myosin: one
more labile, which affects ATPase, and the other more stable, with
no effect. The thiols that are transiently S-nitrosylated have the
potential to modulate the demand for ATP in situations where
the production of nitric oxide increases during muscle contraction.
As an additional strategy to investigate the relationship between
different thiol populations and the SNO groups we have found in
myosin, we concentrated on the more stable S-nitrosylation.
In these experiments, we first alkylated the more reactive and
accessible thiol groups of the native protein with NEM. In this
condition, in agreement with Watterson et al. [51], ten cysteine
residues are blocked (Figure 5E). After this procedure, the protein
was exposed to GSNO, followed by five washes, to study the
more stable SNO groups. The SNO content of myosin alkylated
and then S-nitrosylated in this way was decreased only in high
GSNO concentrations, from ∼ 5 to ∼ 3 mol/mol for native and
alkylated myosin respectively. This means that, at most, only
20 % (two out of ten) of the total number of thiols accessible
to alkylation can form stable SNO groups, whereas the other
alkylated cysteine residues undergo labile S-nitrosylation. Since
unfolding the protein with GdmCl before the alkylation step does
not increase the number of cysteine residues blocked by NEM,
we infer that all of the alkylated thiols are located on residues that
are more accessible to the solvent in the native protein, including
those (six out of ten) that form the more labile SNO groups;
their geometry might well account for the higher rate of RSNO
decomposition [33]. Conversely, as with other proteins [52],
stable SNO bonds may be associated with limited accessibility
of the solvent to the interior of the protein. Recently, Paige
et al. [48] presented evidence from a range of different proteins
c The Authors Journal compilation c 2009 Biochemical Society
230
L. Nogueira and others
that stable SNO groups may form when S-nitrosylation induces
a conformational change that blocks access of glutathione to the
reactive thiol, even one on the surface. Our data with myosin
suggest only that the SNO sites belong to two classes of thiols:
those in more exposed regions, which bind transiently to nitroso
groups and affect Mg2+ -ATPase activity, and those in more buried
regions on myosin, which form stable S-nitrosylation sites, but do
not inhibit ATPase activity.
Formation of 3-nitrotyrosine and protein disulfide bonds can
also occur with GSNO, and would not be detected by the Saville
assay, where specificity is conferred by the reaction of Hg(II)
with sulfur in the RSNO [53]. If these other bonds are formed,
it seems unlikely that they are responsible for inhibition of the
ATPase in our experiments, since they are inherently more stable
than SNO modifications and are not easily lost simply by washing
the protein [11].
The force production in skeletal muscle fibres is the result of
Ca2+ release from the sarcoplasmic reticulum, activation of the
thin filament by Ca2+ and power stroke production by myosin
on the thin filament. Significant decreases in maximal force and
unloaded shortening velocity have been obtained using SNO
donors either in skinned fibres [6,54] or in intact fibres [7,30].
Inhibition of contraction by SNOs in intact fibres appears to
involve Ca2+ release from the sarcoplasmic reticulum [7,30,55]. In
skinned fibres, however, this component can easily be eliminated,
leaving only cross-bridges and Ca2+ regulation of the thin filament
as potential targets. A somewhat larger effect on Ca2+ sensitivity
suggests additional sites of action on the interacting filaments
[6,8]. Inhibition of the Mg2+ -ATPase activity of myosin and
actomyosin by GSNO provides a plausible explanation for these
functional effects of SNO donors in muscle fibres.
During skeletal muscle activation, Ca2+ released by the
sarcoplasmic reticulum promotes both muscle contraction and
activation of nitric oxide synthase. This transient increase in NO•
production could form reversible SNO bonds in proteins only
during muscle contraction. Different authors have proposed that a
reversible S-nitrosylation reaction would serve to protect cysteine
residues from irreversible oxidation during ischaemia [30,33,44].
In skeletal muscle, myosin is present in such large amounts that
its labile SNO groups could act as a buffer against a transient
increase of low-molecular-mass RSNO in muscle fibres during
nitrosative stress or contractile activity. It is interesting that only
the Mg2+ -ATPase activity is inhibited by GSNO. Further study
will be required to establish whether S-nitrosylation of a specific
residue is responsible for this effect.
AUTHOR CONTRIBUTION
Leonardo Nogueira, Cicero Figueiredo-Freitas and Martha M. Sorenson designed
research; Leonardo Nogueira and Cicero Figueiredo-Freitas performed research and
analysed data; Gustavo Casimiro-Lopes exercised the rats and removed the muscles; Jamil
Assreuy and Margaret H. Magdesian contributed new reagents/analytic tools; Leonardo
Nogueira and Martha M. Sorenson wrote the paper.
ACKNOWLEDGEMENTS
We thank Ms D. M. Araujo for the skilful preparation of actin and myosin, Mr D. Ramos
Filho for assistance with the exercised rats, Dr M. W. Foster (Department of Medicine,
Duke University, Durham, NC, U.S.A.), Mr D. P. Reynaldo, Dr H. M. Scofano and Dr P. L.
Oliveira (Instituto Bioquı́mica Médica, UFRJ) for critical reading of the manuscript. We are
grateful to Dr J. Stamler (Department of Medicine, Duke University) for making it possible
for L. N. to carry out the experiment with C2C12 cells.
FUNDING
This work was supported financially by the Brazilian agencies Conselho Nacional de
Desenvolvimento Cientı́fico e Tecnológico (CNPq), Fundação de Amparo à Pesquisa
c The Authors Journal compilation c 2009 Biochemical Society
Carlos Chagas Filho do Estado do Rio de Janeiro (FAPERJ), Fundação de Apoio à
Pesquisa Cientı́fica e Tecnológica do Estado de Santa Catarina (FAPESC/SC), Programa
de Apoio ao Desenvolvimento Cientı́fico e Tecnológico (PADCT) and Programa de Núcleos
de Excelência (PRONEX). L. N. and C. F. F. were the recipients of graduate fellowships from
CNPq and Coordenação de Aperfeiçoamento de Pessoal de Nı́vel Superior (CAPES)
respectively. J. A. and M. M. S. hold research fellowships from CNPq, and G. C. L. has a
postdoctoral fellowship from FAPERJ.
REFERENCES
1 Buttkus, H. (1971) The sulfhydryl content of rabbit and trout myosins in relation to protein
stability. Can. J. Biochem. 49, 97–107
2 Kay, J. (1978) Intracellular protein degradation. Biochem. Soc. Trans. 6, 789–797
3 Stamler, J. S. and Meissner, G. (2001) Physiology of nitric oxide in skeletal muscle.
Physiol. Rev. 81, 209–237
4 Pye, D., Palomero, J., Kabayo, T. and Jackson, M. J. (2007) Real-time measurement of
nitric oxide in single mature mouse skeletal muscle fibres during contractions. J. Physiol.
581, 309–318
5 Kobzik, L., Reid, M. B., Bredt, D. S. and Stamler, J. S. (1994) Nitric oxide in skeletal
muscle. Nature 372, 546–548
6 Perkins, W. J., Han, Y. S. and Sieck, G. C. (1997) Skeletal muscle force and
actomyosin ATPase activity reduced by nitric oxide donor. J. Appl. Physiol. 83,
1326–1332
7 Andrade, F. H., Reid, M. B., Allen, D. G. and Westerblad, H. (1998) Effect of nitric oxide on
single skeletal muscle fibres from the mouse. J. Physiol. 509, 577–586
8 Spencer, T. and Posterino, G. S. (2009) Sequential effects of GSNO and H2 O2 on the
Ca2+ -sensitivity of the contractile apparatus of fast- and slow-twitch skeletal muscle
fibers from the rat. Am. J. Physiol. Cell Physiol. 296, C1015–C1023
9 Dalle-Donne, I., Milzani, A., Giustarini, D., Di Simplicio, P., Colombo, R. and Rossi, R.
(2000) S-NO-actin: S-nitrosylation kinetics and the effect on isolated vascular smooth
muscle. J. Muscle Res. Cell Motil. 21, 171–181
10 Reisler, E., Burke, M. and Harrington, W. F. (1974) Cooperative role of two sulfhydryl
groups in myosin adenosine triphosphatase. Biochemistry 13, 2014–2022
11 Tiago, T., Simao, S., Aureliano, M., Martin-Romero, F. J. and Gutierrez-Merino, C. (2006)
Inhibition of skeletal muscle S1-myosin ATPase by peroxynitrite. Biochemistry 45,
3794–3804
12 Foster, M. W., McMahon, T. J. and Stamler, J. S. (2003) S-nitrosylation in health and
disease. Trends Mol. Med. 9, 160–168
13 Leeuwenburgh, C., Fiebig, R., Chandwaney, R. and Ji, L. L. (1994) Aging and exercise
training in skeletal muscle: responses of glutathione and antioxidant enzyme systems.
Am. J. Physiol. 267, R439–R445
14 Bremel, R. D. and Weber, A. (1975) Calcium binding to rabbit skeletal myosin under
physiological conditions. Biochim. Biophys. Acta 376, 366–374
15 Pardee, J. D. and Spudich, J. A. (1982) Purification of muscle actin. Methods Enzymol.
85, 164–181
16 Mariano, A. C., Alexandre, G. M., Silva, L. C., Romeiro, A., Cameron, L. C., Chen, Y.,
Chase, P. B. and Sorenson, M. M. (2001) Dimethyl sulphoxide enhances the effects of Pi
in myofibrils and inhibits the activity of rabbit skeletal muscle contractile proteins.
Biochem. J. 358, 627–636
17 Sousa, V. P., Pinto, J. R. and Sorenson, M. M. (2006) Ionic interventions that alter the
association of troponin C C-domain with the thin filaments of vertebrate striated muscle.
Biochim. Biophys. Acta 1760, 272–282
18 Wang, Y., Xu, L., Duan, H., Pasek, D. A, Eu, J. P. and Meissner, G. (2006) Knocking down
type 2 but not type 1 calsequestrin reduces calcium sequestration and release in C2C12
skeletal muscle myotubes. J. Biol. Chem. 281, 15572–15581
19 Barany, M. and Close, R. I. (1971) The transformation of myosin in cross-innervated rat
muscles. J. Physiol. 213, 455–474
20 Saville, B. (1958) A scheme for the colorimetric determination of microgram amounts of
thiols. Analyst 83, 670–672
21 Lymn, R. W. and Taylor, E. W. (1970) Transient state phosphate production in the
hydrolysis of nucleoside triphosphates by myosin. Biochemistry 9, 2975–2983
22 Forrester, M. T., Foster, M. W. and Stamler, J.S. (2007) Assessment and application of the
biotin switch technique for examining protein S -nitrosylation under conditions of
pharmacologically induced oxidative stress. J. Biol. Chem. 282, 13977–13983
23 Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J. (1951) Protein
measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265–275
24 Forrester, M. T., Thompson, J. W., Foster, M. W., Nogueira, L., Moseley, M. A. and
Stamler, J. S. (2009) Proteomic analysis of S -nitrosylation and denitrosylation by
resin-assisted capture. Nat. Biotechnol. 27, 557–559
25 Ellman, G. L. (1959) Tissue sulfhydryl groups. Arch. Biochem. Biophys. 82, 70–77
Inhibition of myosin by S-nitrosylation
26 Gergel, D. and Cederbaum, A. I. (1997) Interaction of nitric oxide with
2-thio-5-nitrobenzoic acid: implications for the determination of free sulfhydryl groups by
Ellman’s reagent. Arch. Biochem. Biophys. 347, 282–288
27 Porzio, M. A. and Pearson, A. M. (1977) Improved resolution of myofibrillar proteins with
sodium dodecyl sulfate–polyacrylamide gel electrophoresis. Biochim. Biophys. Acta 490,
27–34
28 Reinartz, M., Ding, Z., Flogel, U., Godecke, A. and Schrader, J. (2008) Nitrosative stress
leads to protein glutathiolation, increased S -nitrosation, and up-regulation of
peroxiredoxins in the heart. J. Biol. Chem. 283, 17440–17449
29 Shi, Q., Feng, J. H., Qu, H. B. and Cheng, Y.-Y. (2008) A proteomic study of
S -nitrosylation in the rat cardiac proteins in vitro . Biol. Pharm. Bull. 31, 1536–1540
30 Sun, J., Morgan, M., Shen, R. F., Steenbergen, C. and Murphy, E. (2007) Preconditioning
results in S -nitrosylation of proteins involved in regulation of mitochondrial energetics
and calcium transport. Circ. Res. 101, 1155–1163
31 Schaub, M. C., Watterson, J. G. and Waser, P. G. (1975) Conformational differences in
myosin, IV. [1–3] Radioactive labeling of specific thiol groups as influenced by ligand
binding. Hoppe Seylers Z. Physiol. Chem. 356, 325–339
32 Sun, J., Yamaguchi, N., Xu, L., Eu, J. P., Stamler, J. S. and Meissner, G. (2008) Regulation
of the cardiac muscle ryanodine receptor by O2 tension and S -nitrosoglutathione.
Biochemistry 47, 13985–13990
33 Hess, D. T., Matsumoto, A., Kim, S. O., Marshall, H. E. and Stamler, J. S. (2005)
Protein S -nitrosylation: purview and parameters. Nat. Rev. Mol. Cell Biol. 6,
150–166
34 Silveira, L. R., Pereira-da-Silva, L., Juel, C. and Hellsten, Y. (2003) Formation of
hydrogen peroxide and nitric oxide in rat skeletal muscle cells during contractions. Free
Radical Biol. Med. 35, 455–464
35 Balon, T. W. and Nadler, J. L. (1994) Nitric oxide release is present from incubated
skeletal muscle preparations. J. Appl. Physiol. 77, 2519–2521
36 Dedkova, E. N., Wang, Y. G., Ji1, X., Blatter, L. A., Samarel, A. M. and Lipsius, S. L.
(2007) Signalling mechanisms in contraction-mediated stimulation of intracellular NO
production in cat ventricular myocytes. J. Physiol. 580, 327–345
37 Eastwood, A. B., Wood, D. S., Bock, K. L. and Sorenson, M. M. (1979) Chemically
skinned mammalian skeletal muscle. I. The structure of skinned rabbit psoas. Tissue Cell
11, 553–566
38 Jaffrey, S. R., Erdjument-Bromage, H., Ferris, C. D., Tempst, P. and Snyder, S. H. (2001)
Protein S -nitrosylation: a physiological signal for neuronal nitric oxide. Nat. Cell Biol. 3,
193–197
39 Carvalho-Filho, M. A., Ueno, M., Carvalheira, J. B. C., Velloso, L. A. and Saad, M. J.
(2006) Targeted disruption of iNOS prevents LPS-induced S -nitrosation of IRβ/IRS-1
and Akt and insulin resistance in muscle of mice. Am. J. Physiol. Endocrinol. Metab.
291, E476–E482
231
40 Kluge, I., Gutteck-Amsler, U., Zollinger, M. and Do, K. Q. (1997) S -nitrosoglutathione in
rat cerebellum: identification and quantification by liquid chromatography-mass
spectrometry. J. Neurochem. 69, 2599–2607
41 Ng, E. S., Cheng, Z. J., Ellis, A., Ding, H., Jiang, Y., Li, Y., Hollenberg, M. D. and Triggle,
C. R. (2007) Nitrosothiol stores in vascular tissue: modulation by ultraviolet light,
acetylcholine and ionomycin. Eur. J. Pharmacol. 560, 183–192
42 Steffen, M., Sarkela, T. M., Gybina, A. A., Steele, T. W., Trasseth, N. J., Kuehl, D. and
Giulivi, C. (2001) Metabolism of S -nitrosoglutathione in intact mitochondria. Biochem. J.
356, 395–402
43 Bryan, N. S., Rassaf, T., Maloney, R. E., Rodriguez, C. M., Saijo, F., Rodriguez, J. R. and
Feelisch, M. (2004) Cellular targets and mechanisms of nitros(yl)ation: an insight into
their nature and kinetics in vivo . Proc. Natl. Acad. Sci. U.S.A. 101, 4308–4313
44 Hare, J. M. and Stamler, J. S. (2005) NO/redox disequilibrium in the failing heart and
cardiovascular system. J. Clin. Invest. 115, 509–517
45 Liu, L., Hausladen, A., Zeng, M., Que, L., Heitman, J. and Stamler, J. S. (2001) A metabolic
enzyme for S -nitrosothiol conserved from bacteria to humans. Nature 410, 490–494
46 Sanghani, P. C., Davis, W. I., Zhai, L. M. and Robinson, H. (2006) Structure–function
relationships in human glutathione-dependent formaldehyde dehydrogenase: role of
Glu-67 and Arg-368 in the catalytic mechanism. Biochemistry 45, 4819–4830
47 Pearce, L. L., Kanai, A. J., Birder, L. A., Pitt, B. R. and Peterson, J. (2002) The catabolic
fate of nitric oxide. J. Biol. Chem. 277, 13556–13562
48 Paige, J. S., Xu, G., Stancevic, B. and Jaffrey, S. R. (2008) Nitrosothiol reactivity profiling
identifies S -nitrosylated proteins with unexpected stability. Chem. Biol. 15, 1307–1316
49 Meeusen, R. L. and Cande, W. Z. (1979) N -ethylmaleimide-modified heavy meromyosin:
a probe for actomyosin interactions. J. Cell Biol. 82, 57–65
50 Sakamoto, H., Furukawa, K., Matsunaga, K., Nakamura, H. and Ohizumi, Y. (1995)
Xestoquinone activates skeletal muscle actomyosin ATPase by modification of the specific
sulfhydryl group in the myosin head probably distinct from sulfhydryl groups SH1 and
SH2. Biochemistry 34, 12570–12575
51 Watterson, J. G., Schaub, M. C., Locher, R., Di Pierri, S. and Kutzer, M. (1975)
Temperature-induced transitions in the conformation of intermediates in the hydrolytic
cycle of myosin. Eur. J. Biochem. 56, 79–90
52 Weichsel, A., Brailey, J. L. and Montfort, W. R. (2007) Buried S -nitrosocysteine revealed
in crystal structures of human thioredoxin. Biochemistry 46, 1219–1227
53 Swift, H. R. and Williams, D. L. H. (1997) Decomposition of S -nitrosothiols by
mercury(II) and silver salts, J. Chem. Soc. Perkin Trans. 2, 1933–1935
54 Galler, S., Hilber, K. and Gobesberger, A. (1997) Effects of nitric oxide on force-generating
proteins of skeletal muscle. Pflügers Arch. 434, 242–245
55 Heunks, L. M., Machiels, H. A., Dekhuijzen, P. N., Prakash, Y. S. and Sieck, G. C. (2001)
Nitric oxide affects sarcoplasmic calcium release in skeletal myotubes. J. Appl. Physiol.
91, 2117–2124
Received 30 July 2009/3 September 2009; accepted 11 September 2009
Published as BJ Immediate Publication 11 September 2009, doi:10.1042/BJ20091144
c The Authors Journal compilation c 2009 Biochemical Society
Biochem. J. (2009) 424, 221–231 (Printed in Great Britain)
doi:10.1042/BJ20091144
SUPPLEMENTARY ONLINE DATA
Myosin is reversibly inhibited by S-nitrosylation
Leonardo NOGUEIRA*1,2 , Cicero FIGUEIREDO-FREITAS*2 , Gustavo CASIMIRO-LOPES*, Margaret H. MAGDESIAN*,
Jamil ASSREUY† and Martha M. SORENSON*3
*Instituto de Bioquı́mica Médica, Centro de Ciências da Saúde, Universidade Federal do Rio de Janeiro, Cidade Universitária, Rio de Janeiro, RJ 21941-590, Brazil, and †Departamento
de Farmacologia, Centro de Ciências Biológicas, Universidade Federal de Santa Catarina, Florianópolis, SC 88015-420, Brazil
Figure S1
Reaction of purified myosin with GSNO or DEANO followed by the biotin-switch technique
(A) Densitometric analysis of the myosin-SNO signal divided by the MHC input (220 kDa; Coomassie Blue stain; 4 +
− 1 μg) following incubation for 30 min with different concentrations of GSNO
or DEANO. Results are means +
− S.E.M.; n = 4. *P < 0.05 compared with 0 mM GSNO (one-way ANOVA). There is no significant effect for DEANO (n = 2). In the positive control (PC), MMTS
was replaced by DTT, followed by the biotin-switch protocol. (B) Densitometry of the time-dependent S -nitrosylation of myosin by 100 μM GSNO followed by the biotin-switch technique was used
to quantify the relationship between myosin-SNO and protein content. Results are means +
− S.E.M.; n = 3. *P < 0.05 compared with 0, 15 and 30 min of reaction with 100 μM GSNO (one-way
ANOVA).
Received 30 July 2009/3 September 2009; accepted 11 September 2009
Published as BJ Immediate Publication 11 September 2009, doi:10.1042/BJ20091144
1
Present address: Division of Physiology, Department of Medicine, University of California San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0623,
U.S.A.
2
These authors contributed equally to this work.
3
To whom correspondence should be addressed (email [email protected]).
c The Authors Journal compilation c 2009 Biochemical Society