Download Sphingosine 1-phosphate protects mouse extensor digitorum longus

Am J Physiol Cell Physiol 288: C1367–C1373, 2005.
First published January 19, 2005; doi:10.1152/ajpcell.00246.2004.
Sphingosine 1-phosphate protects mouse extensor digitorum longus skeletal
muscle during fatigue
Daniela Danieli-Betto,1 Elena Germinario,1 Alessandra Esposito,1 Aram Megighian,1 Menotti Midrio,1
Barbara Ravara,2 Ernesto Damiani,2 Luciano Dalla Libera,3 Roger A. Sabbadini,4 and Romeo Betto3
1
Department of Human Anatomy and Physiology and 2Department of Biomedical and Experimental
Sciences, University of Padua, Padua; 3Muscle Biology and Physiopathology Unit, Consiglio
Nazionale delle Ricerche Institute of Neuroscience, Padua, Italy; and 4Department
of Biology and Heart Institute, San Diego State University, San Diego, California
Submitted 17 May 2004; accepted in final form 18 January 2005
THE SPHINGOMYELIN SIGNALING PATHWAY exerts important biological actions in different cell types (25–27, 37, 43). The receptor-mediated activation of sphingomyelinase sequentially generates different biologically active compounds, namely, ceramide, sphingosine (SPH), and sphingosine 1-phosphate (S1P).
Ceramide activates pathways that regulate stress responses,
cell senescence, and apoptosis (21). The enzyme ceramidase
converts ceramide to SPH, which can be phosphorylated to
S1P by a specific kinase: sphingosine kinase (SK) (35). SPH
regulates the activity of intracellular targets (20), while S1P
shows both intracellular and extracellular actions, the latter by
stimulating specific cell surface G protein-coupled receptors
(25, 27, 37, 43). SPH usually inhibits cell proliferation and
promotes apoptosis, while S1P stimulates growth and protects
against apoptosis (2, 37, 43). Important to their actions is the
considerable serum level of both SPH and S1P (9, 10, 13, 27,
47). Because of the different and sometimes opposite roles of
these lipids, great importance is attributed to the expression
level and presence of the converting enzymes ceramidase and
SK, whose activity may control the “sphingolipid rheostat”
wherein the ratio of sphingolipids with opposing actions becomes important in determining cell fate (27, 37).
Recent evidence shows that many of the signaling components of the sphingomyelin cascade are present in skeletal
muscle (19, 40). Data from rabbit, rat, and pig show that SPH
is endogenous to skeletal muscle (6, 14, 39). In fact, T-tubular
membranes are rich in both SPH and the SPH precursor
sphingomyelin (15, 39), and T tubules express high levels of
sphingomyelinase activity (19, 40). Ceramide is elevated in
fast-twitch glycolytic rat skeletal muscles and low in oxidative
muscles (14). Importantly, during prolonged exercise, sphingomyelinase activity of skeletal muscle increases, ceramide
level decreases, and SPH level increases (15), suggesting that
the whole sphingomyelin pathway is activated in exercised
muscles. SPH is reported to act on the ryanodine receptor, the
sarcoplasmic reticulum (SR) Ca2⫹ release channel (5, 6, 34,
38, 42), and the T tubule dihydropyridine receptor (DHPR)
voltage-dependent Ca2⫹ channel, i.e., the protein complex
responsible for the L-type Ca2⫹ current in cardiac muscle cells
(29, 44). SPH is also known to regulate the activity of protein
kinase C and related downstream pathways (20); thus it also
may act in skeletal muscle, especially considering that this
kinase is localized to the T-tubular membranes (41). The level
of S1P in resting skeletal muscle is estimated to be in the low
nanomolar range (47), but its possible role in contractility and
fatigue has not been evaluated. S1P mobilizes Ca2⫹ in C2C12
myoblasts, apparently by activating membrane voltage-sensitive Ca2⫹ channels and intracellular Ca2⫹ release channels (18,
30). The extracellular action of S1P occurs by activation of
specific G protein-coupled cell surface receptors (25, 37, 43),
which have been described in skeletal muscle (4). SK is also
expressed in skeletal muscle extracellular surfaces (30), suggesting that some of the SPH effects could be explained by
Address for reprint requests and other correspondence: D. Danieli-Betto,
Dept. of Human Anatomy and Physiology, Univ. of Padua, Via Marzolo 3,
35131 Padua, Italy (E-mail: [email protected]).
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
sphingosine kinase; action potential; sphingosine
http://www.ajpcell.org
0363-6143/05 $8.00 Copyright © 2005 the American Physiological Society
C1367
Downloaded from ajpcell.physiology.org on July 1, 2009
Danieli-Betto, Daniela, Elena Germinario, Alessandra Esposito, Aram Megighian, Menotti Midrio, Barbara Ravara,
Ernesto Damiani, Luciano Dalla Libera, Roger A. Sabbadini, and
Romeo Betto. Sphingosine 1-phosphate protects mouse extensor
digitorum longus skeletal muscle during fatigue. Am J Physiol Cell
Physiol 288: C1367–C1373, 2005. First published January 19, 2005;
doi:10.1152/ajpcell.00246.2004.—Sphingomyelin derivatives exert
various second messenger actions in numerous tissues. Sphingosine
(SPH) and sphingosine 1-phosphate (S1P) are two major sphingomyelin derivatives present at high levels in blood. The aim of the present
work was to investigate whether S1P and SPH exert relevant actions
in mouse skeletal muscle contractility and fatigue. Exogenous S1P
and SPH administration caused a significant reduction of tension
decline during fatigue of extensor digitorum longus muscle. Final
tension after the fatiguing protocol was 40% higher than in untreated
muscle. Interestingly, N,N-dimethylsphingosine, an inhibitor of SPH
kinase (SK), abolished the effect of supplemented SPH but not that of
S1P, suggesting that SPH acts through its conversion to S1P. Moreover, SPH was not effective in Ca2⫹-free solutions, in agreement with
the hypothesis that SPH action is dependent on its conversion to S1P
by the Ca2⫹-requiring enzyme SK. In contrast to SPH, S1P produced
its positive effects on fatigue in Ca2⫹-free conditions, indicating that
S1P action does not require Ca2⫹ entry and most likely is receptor
mediated. The effects of S1P could be ascribed in part to its ability to
prevent the reduction (⫺20 mV) of action potential amplitude caused
by fatigue. In conclusion, these results indicate that extracellular S1P
has protective effects during the development of muscle fatigue and
that the extracellular conversion of SPH to S1P may represent a
rheostat mechanism to protect skeletal muscle from possible cytotoxic
actions of SPH.
C1368
SPHINGOSINE 1-PHOSPHATE AND SKELETAL MUSCLE FATIGUE
SPH’s enzymatic conversion to S1P. Consistent with this view
is our recent demonstration that SPH released from cardiomyocytes is rapidly converted to S1P (7).
Because both SPH and S1P are present at very high levels in
the blood (9, 10, 13, 45), muscle fibers are continuously bathed
in elevated levels of the two lipids. Thus we decided to
investigate the possible extracellular actions of SPH and S1P in
skeletal muscle contractility and fatigue. Our results indicate
that exogenous SPH and S1P administration makes the muscle
more resistant to fatigue. Moreover, because pretreating muscle with SK inhibitors abolished the protective effect of SPH,
it appears that SPH acts through its conversion to S1P and that
S1P is responsible for the observed effects of SPH.
MATERIALS AND METHODS
AJP-Cell Physiol • VOL
288 • JUNE 2005 •
www.ajpcell.org
Downloaded from ajpcell.physiology.org on July 1, 2009
All the experiments were conducted according to the Declaration of
Helsinki regarding the treatment of animals during experimentation.
The study was approved by the Ethics Committee of the Medical
Faculty of the University of Padua. All efforts were made to minimize
animal suffering and to use only the number of mice necessary to
obtain reliable data.
Tension recording and fatigue protocols. Adult 3-month-old CD1
mice were used for these studies. Extensor digitorum longus (EDL)
muscles were removed and placed in a bath containing Ringer physiological saline solution maintained at room temperature and containing 120 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl2, 3.15 mM MgCl2, 1.3
mM NaH2PO4, 25 mM NaHCO3, 11 mM glucose, and 30 ␮M
D-tubocurarine, pH 7.2–7.4. The solution was continuously bubbled
with 95% O2-5% CO2 until used. For the experimental test, muscles
were mounted in a vertical bath filled with 12 ml of the above
physiological solution bubbled with 95% O2-5% CO2 at 30°C. Muscles were electrically stimulated with supramaximal pulses (0.5-ms
duration, 0.01 s⫺1) delivered using a Grass S44 electronic stimulator
(Grass Instrument, Quincy, MA) through a stimulus isolation unit
(Grass SIU5), and isometric force was recorded with a force transducer (Grass FT03). Muscle responses were recorded via an AT-MIO
16 AD card, and data were analyzed using a virtual instrument created
with the LabView software program (National Instruments, Austin,
TX) (11). Muscles were stretched to the optimal length for twitch
tension and equilibrated for 15 min at 30°C in the medium, to which
either 5 ␮M fatty acid-free bovine serum albumin (BSA) was added
in control experiments; otherwise, the indicated compounds were
added.
Before and after the incubation period, twitch and tetanic (150 Hz,
200 ms) tension levels were measured. Muscles were then subjected
to a fatigue protocol consisting of a train of short tetani (150 Hz, 200
ms, 0.2/s) under the conditions described below. Recovery of muscle
tetanic tension after administration of the fatigue protocol was followed for 30 min, with a single tetanus (150 Hz, 200 ms) delivered to
muscles every 5 min. Recovery was recorded as the percentage of
initial tension.
It is worth noting that we analyzed fatigue development only in
selected muscles having about the same initial tetanic tension (N/g) to
avoid the effects produced by different accumulation and depletion of
metabolites during fatigue (17, 50). The maximum tetanic tension, the
final force (expressed as percentage of the initial force) and the time
necessary for 50% tension decay (t1/2) were measured.
Treatments. The fatigue experiments were performed according to
the following protocols: 1) control condition, Ringer solution containing 5 ␮M fatty acid-free BSA; 2) 1, 10, and 100 ␮M SPH (Sigma); 3)
1, 10, and 20 ␮M S1P (Avanti Polar Lipids); 4) SPH ⫹ N,Ndimethylsphingosine (DMS, 20 ␮M; Avanti Polar Lipids), an inhibitor of SPH kinase (46) that converts SPH to S1P; 5) 10 or 20 ␮M
DMS; 6) 0.25 ␮M bis-indolylmaleimide (bis-IM; Sigma), a cellpermeable inhibitor of protein kinase C; 7) SPH ⫹ bis-IM, 10 ␮M
SPH, and 0.25 ␮M bis-IM; 8) S1P ⫹ bis-IM, 10 ␮M S1P, and 0.25
␮M bis-IM; 9) SPH/Ca2⫹-free Ringer solution, 10 ␮M SPH in a
Ca2⫹-free solution; and 10) S1P/Ca2⫹-free Ringer solution, 10 ␮M
S1P in a Ca2⫹-free solution.
SPH, S1P, DMS, and bis-IM were dispersed by performing sonication on a solution of 5 mM fatty acid-free BSA and then storing it
at ⫺80°C until used. Before addition to the bath, SPH, DMS, and S1P
were briefly sonicated and delivered to the physiological medium so
that the final concentration of albumin was 5 ␮M.
Electrophysiological analysis. Skeletal muscle fiber resting membrane potential (RMP) and directly evoked action potential (AP) were
studied in control and fatigued fibers in the absence or presence of 10
␮M SPH or S1P using intracellular glass microelectrodes filled with
3 M KCl. EDL muscles were incubated for 30 min with SPH or S1P
before the experiment was begun at 21 ⫾ 1°C. Direct extracellular
stimulation was performed using two-needle Ag/AgCl electrodes with
supramaximal electrical stimuli (130 –150 V, 0.1-ms duration) as
described previously (32). Intracellular recorded signals were sent to
a preamplifier (WPI 705; World Precision Instruments) and then to a
signal conditioner (CyberAmp; Axon Instruments-Molecular Devices,
Union City, CA). The output signal from the conditioner was then fed
to a digital oscilloscope (Tektronix 5022N; Tektronix, Richardson,
TX). No more than a single AP was generally recorded from the same
fiber, because muscle membrane was damaged during each stimulation by the relative movements of the recording electrode with respect
to the contracting muscle. The maximal rate of AP rise was determined by calculating the first derivative of the digitized spike.
Determination of SPH levels by reverse-phase HPLC. SPH was
determined using HPLC according to a previously described method
(9). Briefly, sphingolipids were extracted with butanol and derivatized
with o-phthaldehyde (Molecular Probes, Eugene, OR). Samples were
separated on a Hi-Pore Reversed Phase Column RP-318 (4.6 mm ⫻
25 cm; Bio-Rad, Hercules, CA) run on a Bio-Rad HPLC system
equipped with a PerkinElmer fluorescence detector (excitation,
340 nm; emission, 455 nm). The solvent system was composed of
methanol, glacial acetic acid, 1 M tetrabutylammonium dihydrogenphosphate, and nanopure water (82.9:1.5:0.9:14.7, vol/vol) run at 1.0
ml 䡠 min⫺1. It is worth mentioning that the assay measures SPH
content of the whole muscle and that it is not possible to evaluate the
contribution of SPH derived from blood, smooth muscle, or endothelial cells. As a consequence, even though the enzymatic apparatus for
SPH production has been demonstrated in skeletal muscle T tubules
(15, 19, 39), some of the muscle SPH could be generated by nonmuscle cells.
Membrane preparation and [3H]ryanodine binding assay. Heavy
membrane fractions were purified from mouse hindlimb muscles as
previously described (28). Briefly, muscle homogenates were prepared in 10 mM HEPES, pH 7.4, and 20 mM KCl buffer, supplemented with the Complete protease inhibitor cocktail (Roche), and
centrifuged at 650 g for 10 min. The myofibrillar sediment was
reextracted twice in the same buffer. The three supernatants were then
pooled and centrifuged at 120,000 g for 90 min. The total membrane
fraction was resuspended in 0.3 M sucrose and 10 mM Tris 䡠 HCl, pH
7.4, and stored at ⫺80°C.
SR-enriched membranes (at a protein concentration of 0.1
mg 䡠 ml⫺1) were incubated with 50 nM [3H]ryanodine in 0.5 ml of 0.1
M KCl, 50 ␮M CaCl2, and 20 mM HEPES, pH 7.4, for 30 min at 37°C
in the presence of various concentrations of SPH (5, 6, 38). The
sphingoid base was added to the medium assay as a DMSO solution,
and the solute did not exceed 1% of total volume. Control experiments
demonstrated that the vehicle had no effect on ryanodine binding.
Bound [3H]ryanodine was measured by filtration through Millipore
nitrocellulose filters (0.45 ␮m) and liquid scintillation counting.
Specific [3H]ryanodine binding was determined by subtracting the
nonspecific binding obtained in the presence of 10 ␮M unlabeled
ryanodine.
C1369
SPHINGOSINE 1-PHOSPHATE AND SKELETAL MUSCLE FATIGUE
Statistical analysis. The force recorded during the fatigue protocol
was normalized to the initial tension. Individual values are presented
as means ⫾ SE. Means were compared using ANOVA, and significance was determined using the Kruskal-Wallis post hoc test. The
statistical significance of SPH levels was determined using Student’s
t-test for unpaired samples. Differences were considered significant at
P ⬍ 0.05.
RESULTS
Effects of the different treatments on twitch and tetanic
tension levels. All treatments did not affect twitch properties
and tetanic tension in the resting state (data not shown).
Effects of S1P and SPH on the development of fatigue.
Because SPH and S1P are normally present at high levels in the
blood of experimental animals and humans (9, 10, 13, 45), we
hypothesized that they might affect contractility and/or fatigue.
Thus we performed fatigue protocols in the presence of exogenously applied SPH and S1P. Normal development of fatigue
Table 1. Effects of sphingolipids on development of fatigue in mouse EDL muscle
Final Tension Output
Control
SPH, 1 ␮M
SPH, 10 ␮M
SPH, 100 ␮M
S1P, 1 ␮M
S1P, 10 ␮M
S1P, 20 ␮M
SPH ⫹ DMS, 10 ␮M ⫹ 20 ␮M
DMS, 10 or 20 ␮M
Bis-IM, 0.25 ␮M
SPH ⫹ bis-IM, 10 ␮M ⫹ 0.25 ␮M
S1P ⫹ bis-IM, 10 ␮M ⫹ 0.25 ␮M
SPH, 10 ␮M/Ca2⫹ free
S1P, 10 ␮M/Ca2⫹ free
t1/2
%Initial tension
⌬%
Min
⌬%
22.8⫾2.2
34.7⫾2.9
31.9⫾2.7
26.7⫾2.0
32.0⫾3.1
32.5⫾3.1
29.3⫾3.4
23.8⫾2.4
22.2⫾1.5
27.4⫾1.4
27.4⫾3.4
31.6⫾3.4
23.2⫾2.2
40.0⫾4.1
⫹52.2†
⫹39.9*
⫹17.1
⫹40.4*
⫹42.5*
⫹28.4
⫹4.4
⫺2.6
⫹20.2
⫹20.2
⫹38.6*
⫹1.8
⫹75.4*
2.35⫾0.17
3.61⫾0.22
3.21⫾0.14
2.87⫾0.20
3.38⫾0.46
3.29⫾0.39
3.01⫾0.34
2.46⫾0.23
2.28⫾0.03
2.58⫾0.12
2.63⫾0.14
3.04⫾0.27
2.44⫾0.26
4.27⫾0.70
⫹53.6*
⫹36.6*
⫹22.1
⫹43.8*
⫹40.0*
⫹28.1
⫹4.7
⫺3.0
⫹9.8
⫹11.9
⫹29.4*
⫹3.8
⫹81.7*
Values are means ⫾ SE. EDL, extensor digitorum longus; t1/2, time necessary for 50% tension decay; SPH, sphingosine; S1P, sphingosine 1-phosphate; DMS,
N,N-dimethylsphingosine; bis-IM, bis-indolylmaleimide. At least four experiments were performed for each condition. The final tension output refers to 6-min
fatigue protocol. *P ⬍ 0.05, †P ⬍ 0.01 vs. control values.
AJP-Cell Physiol • VOL
288 • JUNE 2005 •
www.ajpcell.org
Downloaded from ajpcell.physiology.org on July 1, 2009
Fig. 1. Sphingosine (SPH) reduces fatigue development of extensor digitorum
longus (EDL) muscle. EDL mouse muscles were bathed in Ringer solution at
30°C and subjected to trains of short tetani (150 Hz, 200 ms, 0.2/s) for 6 min,
and tension development was recorded. Recovery was then estimated by
stimulating the muscle every 5 min with a single tetanus. Mean tetanic tension
recordings from control vehicle-treated [5 ␮M bovine serum albumin (BSA)]
EDL muscles (open circles; n ⫽ 4) and muscles preincubated with 10 ␮M SPH
(closed circles; n ⫽ 4) are shown. *P ⬍ 0.05 vs. control.
induced by repeated brief contractions at 150 Hz in the EDL
muscle occurred as a progressive decline of force output during
the 6-min stimulation protocol (Fig. 1), reaching 22.8 ⫾ 2.2%
of initial tension value, whereas t1/2 was 2.35 ⫾ 0.17 min
(Table 1). First, three concentrations of exogenous SPH were
tested. In the presence of 1 and 10 ␮M SPH, a marked slowing
of force decline was observed (Fig. 1 and Table 1). In contrast,
the reduction of fatigue produced by 100 ␮M SPH was not
significant. Muscle recovery in the presence of SPH was
similar to that of control EDL (Fig. 1).
We then evaluated the effects of S1P on fatigue development. Similar to the findings with its metabolic precursor, SPH,
S1P at 1 and 10 ␮M caused a significant reduction of fatigue
development (Fig. 2) as indicated by the substantial increase of
both final force and t1/2 with respect to controls (Table 1). As
also observed with SPH, higher S1P levels (20 ␮M) slightly
but not significantly reduced fatigue development (Table 1).
The recovery of the muscles treated with S1P did not differ
from that of controls.
Because SK activity has been detected on the extracellular
side of cells (48), including skeletal muscle cells (30, 33), we
hypothesized that SPH could act indirectly on muscle fatigue
development through its phosphorylation to S1P. To verify this
hypothesis, the action of the SK inhibitor, DMS (46), was
evaluated. The expected beneficial effects of SPH on fatigue
parameters shown in Fig. 1 were completely abolished in the
presence of DMS (Fig. 2). In contrast, DMS (10 or 20 ␮M) was
without evident effects on muscle fatigue when it was delivered alone (Table 1), suggesting that SPH produced during
fatigue is not liberated into the extracellular medium. The
recovery of the muscles treated with both SPH and DMS
appeared to be slightly delayed (Fig. 2).
Effects of bis-IM on the development of fatigue. To evaluate
whether SPH administration could produce, at least in part, an
intracellular action through its well-known inhibitory action on
protein kinase C activity (20), we tested the effect of another
potent inhibitor of the kinase, bis-IM, on the development of
EDL muscle fatigue. Bis-IM produced only a moderate but not
significant effect on EDL fatigue development (Fig. 3 and
Table 1). The simultaneous presence of bis-IM and SPH did
C1370
SPHINGOSINE 1-PHOSPHATE AND SKELETAL MUSCLE FATIGUE
not change the effect produced by bis-IM alone, suggesting
that SPH action on EDL fatigue is not mediated by protein
kinase C. In contrast, bis-IM was not able to alter the positive
effect of S1P (Table 1).
Effects of S1P and SPH on the development of fatigue in
Ca2⫹-free solution. Several workers have demonstrated that
S1P not only mobilizes Ca2⫹ from intracellular stores but also
can facilitate entry of Ca2⫹ from extracellular medium (24,
43). Accordingly, we tested the action of this sphingolipid on
EDL muscle fatigue produced in Ca2⫹-free Ringer solution to
evaluate whether S1P’s actions on fatigability were dependent
on extracellular Ca2⫹. As Fig. 4 shows, S1P maintained its
beneficial effect on fatigue development in Ca2⫹-free solutions. The final force output and t1/2 values were similar to
those obtained in the presence of S1P in normal Ringer
solution and significantly higher (P ⬍ 0.05) than those produced by fatigued control muscles (Table 1). Surprisingly, the
beneficial effects of SPH were completely abolished by the
Fig. 3. Effects of bis-indolylmaleimide (bis-IM) on fatigue development in
EDL mouse muscle. Fatigue was produced and recovery was examined as
described in the Fig. 1 legend. Mean tetanic tension recordings from control
vehicle-treated (5 ␮M BSA) EDL muscles (open circles; n ⫽ 4) and muscles
preincubated with 0.25 ␮M bis-IM, a cell-permeable inhibitor of protein kinase
C (closed circles; n ⫽ 4) are shown.
AJP-Cell Physiol • VOL
Fig. 4. Effects of SPH and S1P on EDL fatigue development in the absence of
external Ca2⫹. Fatigue was produced and recovery was examined as described
in Fig. 1. Mean tetanic tension recordings from control vehicle-treated (5 ␮M
BSA) EDL muscles (open circles; n ⫽ 4) and muscles bathed in a Ca2⫹-free
Ringer solution and preincubated either with 10 ␮M SPH (closed circles; n ⫽
4) or with 10 ␮M S1P (closed triangles; n ⫽ 4) are shown. *P ⬍ 0.05 with
respect to control.
absence of extracellular Ca2⫹ (Fig. 4). The recovery was not
significantly different from control values.
Effects of S1P and SPH on electrophysiological properties.
We next evaluated whether S1P and SPH effects on fatigue
development could be explained by the effects of sphingolipid
mediators on the skeletal muscle AP. SPH is known to induce
depolarization of cardiomyocytes, probably by acting on the
DHPR (29, 44). S1P also appears to influence skeletal muscle
membrane potential (4), an action that is in part receptor
mediated. These data thus suggest another possible site of
action of the sphingolipids during EDL muscle fatigue. Accordingly, skeletal muscle fiber RMP and directly evoked APs
were studied in control and fatigued fibers in the absence or
presence of either S1P or SPH (10 ␮M). As previously reported by others (for review, see Ref. 17), fatigue caused a
significant fall in RMP and a decrease in AP amplitude and
maximum rate of rise (Table 2). For muscles at rest, the
addition of S1P or SPH did not affect the RMP or the AP
amplitude, while the maximum rate of rise of the AP was
decreased. Interestingly, the presence of S1P or SPH prevented
the fatigue-induced decrease in AP amplitude, while it did not
alter the maximum rate of AP rise (Table 2).
Endogenous SPH levels of mouse EDL. Next, we determined
the concentration of SPH using HPLC in mouse EDL muscles
at rest and after fatigue. The resting level of SPH was 13.2 ⫾
4.2 nmol/g fresh tissue (n ⫽ 4), while fatigue caused a
significant increase of endogenous SPH concentration to
28.0 ⫾ 3.2 nmol/g fresh tissue (n ⫽ 4; P ⬍ 0.03), confirming
previous observations that this lipid mediator is increased in
fatigued rat muscles (14). The skeletal muscle endogenous
level of S1P at rest is reported to be at least one order of
magnitude lower than that of SPH (47).
Effects of SPH and S1P on [3H]ryanodine binding. SPH is
known to affect the ryanodine binding of both the cardiac and
skeletal muscle isoforms of the ryanodine receptor Ca2⫹ release channel from rabbit, dog, and pig SR preparations (5, 6,
12, 34, 38, 42). However, it is not known whether SPH affects
the ryanodine receptor from SR mouse membranes as it does in
other species. Figure 5 shows that SPH is able to produce a
dose-dependent inhibition of [3H]ryanodine binding to total
288 • JUNE 2005 •
www.ajpcell.org
Downloaded from ajpcell.physiology.org on July 1, 2009
Fig. 2. Sphingosine 1-phosphate (S1P) reduces fatigue development of EDL
muscle, while N,N-dimethylsphingosine (DMS) abolishes the effects of SPH.
Fatigue was produced and recovery was examined as described in Fig. 1. Mean
tetanic tension recordings from control vehicle-treated (5 ␮M BSA) EDL
muscles (open circles; n ⫽ 4); muscles preincubated with 10 ␮M S1P (closed
triangles; n ⫽ 7); and muscles preincubated with 20 ␮M DMS, the inhibitor of
SPH kinase, and 10 ␮M SPH (closed circles; n ⫽ 4) are shown. *P ⬍ 0.05 vs.
control.
C1371
SPHINGOSINE 1-PHOSPHATE AND SKELETAL MUSCLE FATIGUE
Table 2. Effects of SPH and S1P on electrophysiological properties of mouse EDL muscle
RMP, mV
Amplitude, mV
MRR, V/s
Control
(n ⫽ 138 and 3)
Control ⫹ SPH
(n ⫽ 86 and 3)
Control ⫹ S1P
(n ⫽ 34 and 3)
Fatigued control
(n ⫽ 50 and 3)
Fatigued ⫹ SPH
(n ⫽ 34 and 3)
Fatigued ⫹ S1P
(n ⫽ 29 and 3)
⫺80.5⫾0.8
96.2⫾1.1
299.9⫾7.6
⫺78.6⫾0.9
93.9⫾1.3
262.2⫾8.6*
⫺77.0⫾1.6
96.1⫾1.1
260.6⫾4.3*
⫺76.5⫾1.3*
76.2⫾2.2*
239.2⫾17.3*
⫺77.7⫾1.5
95.3⫾2.2†
235.7⫾12.9
⫺75.2⫾1.6
94.9⫾1.0†
231.5⫾7.4
Values are means ⫾ SE. Numbers in parentheses refer to number of fibers and animals, respectively. Resting membrane potential (RMP), amplitude, and
maximum rate of rise (MRR) of action potentials of single fibers in control and after 30 min of incubation in either 10 ␮M SPH or 10 ␮M S1P. *P ⬍ 0.05 vs.
resting control values; †P ⬍ 0.05 vs. fatigued muscle control values.
membrane preparations isolated from mouse skeletal muscles.
Because the binding of [3H]ryanodine is dependent on the open
state of the Ca2⫹ release channel (31), these data indicate that
SPH acts as an inhibitor of the channel. The SPH concentration
able to inhibit 50% of ryanodine binding (IC50) was ⬃18 ␮M.
As also reported by others (42), S1P was without appreciable
effects on [3H]ryanodine binding (Fig. 5).
The present work shows that the exogenous administration
of S1P and SPH individually caused the significant slowing of
fatigue. However, our data show that SPH action depends on
its conversion to S1P. Hence, this report is the first to suggest
that S1P may exert beneficial effects on the promotion of
fatigue resistance. Our data also show that both S1P and SPH
exert positive actions by affecting membrane electrophysiological properties of muscle fibers in preventing the fatiguedependent drop in AP amplitude.
Because both SPH and S1P demonstrated positive effects on
mouse muscle fatigue, we asked whether SPH’s beneficial
effects could be indirect and mediated by its conversion into
S1P. This hypothesis was suggested by recent findings showing that SK activity could be detected on the extracellular
surfaces of cells (48), including skeletal muscle cells (30). We
Fig. 5. Dose dependence of SPH effect on [3H]ryanodine binding to mouse
microsomes enriched in junctional sarcoplasmic reticulum. Ryanodine binding
was measured using 100 ␮g of microsomes isolated from mouse hindlimb
skeletal muscles and incubated for 30 min at 37°C in 0.5 ml of 0.1 M KCl, 50
␮M CaCl2, and 20 mM HEPES, pH 7.4, and 50 nM [3H]ryanodine. Bound
[3H]ryanodine was determined using Millipore filtration and liquid scintillation
counting. Nonspecific binding was measured in the presence of an excess (10
␮M) of unlabeled ryanodine. Data are means ⫾ SE of three experiments
performed in triplicate. A pilot experiment performed in the presence of 30 ␮M
S1P showed that the lipid was ineffective on [3H]ryanodine binding as
demonstrated previously by others (42).
AJP-Cell Physiol • VOL
288 • JUNE 2005 •
www.ajpcell.org
Downloaded from ajpcell.physiology.org on July 1, 2009
DISCUSSION
recently reported that a substantial amount of SPH produced by
cardiomyocytes is released into the extracellular medium and
rapidly phosphorylated to S1P (7). In the current study, we
discovered that when muscle fatigue was produced in the
presence of the SK inhibitor, DMS, the effects produced by
SPH were completely abolished. Additional support for a role
of SK is the finding that SPH is not effective in retarding
fatigue when muscles do not have Ca2⫹ in the extracellular
fluid. It is known that SK activity is dependent on Ca2⫹ (27,
35). We suggest that a “rheostat” mechanism operates in
skeletal muscle cells as it does in other cell types (27, 37)
whereby S1P is generated from SPH directly at the extracellular surfaces by a Ca2⫹-dependent SK localized in the muscle
cell basement membrane as suggested by observations in
C2C12 myoblasts (30).
An interesting result is that exogenously applied S1P had the
same effect in either the presence or absence of external Ca2⫹,
suggesting that S1P does not act by mobilizing extracellular
Ca2⫹. It has consistently been reported that in skeletal muscle,
exogenous S1P induces Ca2⫹ transients independent of L-type
channels (4). Thus S1P appears to act on the extracellular side
by activating cell surface receptors and downstream signaling
pathways, leading to an overall beneficial effect on decreasing
the development of fatigue. Normally, S1P produces important
intracellular actions by activating specific cell surface G protein-coupled receptors (25, 43), and it is important to note that
skeletal muscle possesses these receptors (4). SK is activated
by several signals (35, 37, 43), among which is the depletion of
endoplasmic reticulum Ca2⫹ stores (24) and Ca2⫹ mobilization
(48), events that are known to occur during sustained lowfrequency fatigue stimulation (1, 3, 17, 50).
We also have investigated whether SPH may exert an
intracellular action by modulating the activity of protein kinase
C, a well-established target for SPH (20). However, SPH does
not seem to produce the positive effects on fatigue through
inhibition of protein kinase C, because the moderate, nonsignificant action produced by the cell-permeable inhibitor of the
kinase, bis-IM, was not augmented by the addition of SPH. The
data presented herein do not exclude the possibility that exogenously added SPH may affect the closure of the ryanodine
receptor, a very potent effect of SPH that is shown in Fig. 5.
Finally, bis-IM does not influence the effects of S1P on fatigue,
suggesting that the signaling pathway activated by S1P does
not include protein kinase C.
A possible target of S1P and SPH action could be found in
the effects of the lipids on the AP. In fact, our results show that
S1P and SPH were able to counteract the fatigue-dependent
decrease in AP amplitude. The prolonged duration of AP
produced by the two lipids likely results in an increase in the
C1372
SPHINGOSINE 1-PHOSPHATE AND SKELETAL MUSCLE FATIGUE
GRANTS
The financial support of Telethon Italy Grants 620, 1286, and GGP02077
(to D. Danieli-Betto, R. Betto, and L. Dalla Libera, respectively) is gratefully
acknowledged. This work also was supported by grants from Programma di
AJP-Cell Physiol • VOL
Ricerca Scientifica di Rilevante Interesse Nazionale Ministero dell’Istruzione
dell’Università della Ricerca Project MIUR-PRIN 2003 (to D. Danieli-Betto
and E. Damiani), National Heart, Lung, and Blood Institute Grant HL-63903
(subcontract to D. Danieli-Betto), Consiglio Nazionale delle Ricerche institutional funds (to R. Betto and L. Dalla Libera), and a grant from the American
Heart Association (to R. A. Sabbadini).
REFERENCES
1. Allen DG, Lannergren J, and Westerblad H. Muscle cell function
during prolonged activity: cellular mechanisms of fatigue. Exp Physiol 80:
497–527, 1995.
2. Andrieu-Abadie N and Levade T. Sphingomyelin hydrolysis during
apoptosis. Biochim Biophys Acta 1585: 126 –134, 2002.
3. Baker AJ, Longuemare MC, Brandes R, and Weiner MW. Intracellular tetanic calcium signals are reduced in fatigue of whole skeletal muscle.
Am J Physiol Cell Physiol 264: C577–C582, 1993.
4. Bencini C, Squecco R, Piperio C, Formigli L, Meacci E, Nosi D,
Tiribilli B, Vassalli M, Quercioli F, Bruni P, Zecchi Orlandini S, and
Francini F. Effects of sphingosine 1-phosphate on excitation-contraction
coupling in mammalian skeletal muscle. J Muscle Res Cell Motil 24:
539 –554, 2003.
5. Betto R, Teresi A, Fachechi-Cassano G, Salviati G, and Sabbadini RA.
Sphingosine inhibits calcium release from sarcoplasmic reticulum membranes. Adv Exp Med Biol 311: 403– 404, 1992.
6. Betto R, Teresi A, Duca M, Turcato F, Danieli-Betto D, and Sabbadini
RA. Inability of sphingosine and calmodulin to control ryanodine receptor
in malignant hyperthermia. Basic Appl Myol 12: 159 –167, 2002.
7. Cavalli AL, Ligutti JA, Gellings NM, Castro E, Page MT, Klepper
RE, Palade PT, McNutt WT, and Sabbadini RA. The role of TNF␣ and
sphingolipid signaling in cardiac hypoxia: Evidence that cardiomyocytes
release TNF␣ and sphingosine. Basic Appl Myol 12: 167–176, 2002.
8. Chatterjee S. Neutral sphingomyelinase: past, present and future. Chem
Phys Lipids 102: 79 –96, 1999.
9. Dalla Libera L, Sabbadini R, Renken C, Ravara B, Sandri M, Betto R,
Angelini A, and Vescovo G. Apoptosis in the skeletal muscle of rats with
heart failure is associated with increased serum levels of TNF-␣ and
sphingosine. J Mol Cell Cardiol 33: 1871–1878, 2001.
10. Dalla Libera L, Ravara B, Volterrani M, Gobbo V, Della Barbera M,
Angelini A, Danieli-Betto D, Germinario E, and Vescovo G. Beneficial
effects of GH/IGF-1 on skeletal muscle atrophy and function in experimental heart failure. Am J Physiol Cell Physiol 286: C138 –C144, 2004.
11. Danieli-Betto D, Germinario E, Esposito A, Biral D, and Betto R.
Effects of fatigue on sarcoplasmic reticulum and myofibrillar properties of
rat single muscle fibers. J Appl Physiol 89: 891– 898, 2000.
12. Dettbarn CA, Betto R, Salviati G, Palade P, Jenkins GM, and Sabbadini RA. Modulation of cardiac sarcoplasmic reticulum ryanodine
receptor by sphingosine. J Mol Cell Cardiol 26: 229 –242, 1994.
13. Deutschman DH, Carstens JS, Klepper RL, Smith WS, Page MT,
Young TR, Gleason LA, Nakajima N, and Sabbadini RA. Predicting
obstructive coronary artery disease with serum sphingosine-1-phosphate.
Am Heart J 146: 62– 68, 2003.
14. Dobrzyń A and Górski J. Effect of acute exercise on the content of free
sphinganine and sphingosine in different skeletal muscle types of the rat.
Horm Metab Res 34: 523–529, 2002.
15. Dobrzyń A, Żendzian-Piotrowska M, and Górski J. Effect of endurance
training on the sphingomyelin-signalling pathway activity in the skeletal
muscles of the rat. J Physiol Pharmacol 55: 305–313, 2004.
16. Edman KA and Lou F. Myofibrillar fatigue versus failure of activation
during repetitive stimulation of frog muscle fibres. J Physiol 457: 655–
673, 1992.
17. Fitts RH. Cellular mechanisms of muscle fatigue. Physiol Rev 74: 49 –94,
1994.
18. Formigli L, Francini F, Meacci E, Vassalli M, Nosi D, Quercioli F,
Tiribilli B, Bencini C, Piperio C, Bruni P, and Zecchi Orlandini S.
Sphingosine 1-phosphate induces Ca2⫹ transients and cytoskeletal rearrangement in C2C12 myoblastic cells. Am J Physiol Cell Physiol 282:
C1361–C1373, 2002.
19. Ghosh N, Sabbadini R, and Chatterjee S. Identification, partial purification, and localization of a neutral sphingomyelinase in rabbit skeletal
muscle: neutral sphingomyelinase in skeletal muscle. Mol Cell Biochem
189: 161–168, 1998.
20. Hannun YA and Bell RM. Functions of sphingolipids and sphingolipid
breakdown products in cellular regulation. Science 243: 500 –507, 1989.
288 • JUNE 2005 •
www.ajpcell.org
Downloaded from ajpcell.physiology.org on July 1, 2009
useful voltage-dependent SR Ca2⫹ release, resulting in a beneficial effect of the lipids in the attempt to counteract the
reduction of tension output. Moreover, because it has been
shown that AP propagation into the T tubules is impaired
during fatigue, causing poor activation of central myofibers
(16), the action of S1P and SPH in causing prolongation of the
active state could have a beneficial result by favoring the
recruitment of all contractile elements. It has consistently been
demonstrated that SPH affects the electrophysiological properties of voltage-gated Na⫹ channels in isolated cardiac myocytes (44) by decreasing the maximum current carried (peak
current) and increasing the opening state. This should result in
a reduction of the AP maximum rate of rise and an increase in
AP amplitude. Moreover, S1P has been demonstrated to favor
the receptor-mediated activation of Cl⫺ currents, resulting in
the depolarization of keratinocytes (49), and also to cause
depolarization of oligodendrocytes (22), while it did not have an
effect on the L-type Ca2⫹ currents in an insulinoma cell line (23).
The present work also shows that SPH is an endogenous
component of mouse skeletal muscle, with concentrations
similar to those reported in rabbit, rat, and pig skeletal muscle
(5, 6, 12, 34, 38, 42). Moreover, during sustained contractile
activity, total endogenous levels of SPH increased in whole
muscle homogenates. Similar results were obtained previously
in rat skeletal muscle (14). The major source of SPH in cells is
sphingomyelin, which is sequentially hydrolyzed by sphingomyelinase and ceramidase to form ceramide and SPH, respectively. The rise in SPH levels (and subsequent S1P production
by the kinase) during fatigue may be due to the activation of
the lysosomal acid sphingomyelinase as a result of the decline
in intracellular pH that accompanies fatigue (17, 51). An
additional increase in the level of SPH could be the consequence of the stimulation of a neutral sphingomyelinase and
cell surface-specific receptors by circulating TNF-␣, IFN-␥,
and IL-1␤, known activators of the sphingomyelinase pathway
that have been shown to increase during sustained exercise (8,
26; for review, see Ref. 36). We also report that the endogenous SPH elevation may affect fatigue development by direct
actions on the SR Ca2⫹ release channel. It is known that SPH
is an endogenous modulator of the ryanodine receptor channel
(5, 6, 12, 34, 38, 42), altering ryanodine binding to the receptor
(5, 38, 42) and its gating properties at the single-channel level
(34). We hypothesized in the present study that the intracellular
actions of SPH on coupling also may be mitigated by the
rheostat mechanism hypothesized herein, whereby the SPH is
converted to S1P, which is known to have no influence on the
ryanodine receptor (42).
In conclusion, present results indicate that SPH exerts a
protective effect on the development of fatigue through mechanisms mostly accounted for by SPH’s conversion into S1P.
The protective effects of these important signaling lipids is
most likely attributable to their actions on the skeletal muscle
AP; however, the abundant presence of S1P receptors on the
skeletal muscle sarcolemma (4) suggests that G protein-coupled receptors may mediate S1P actions. Our laboratory is
currently investigating this possibility.
SPHINGOSINE 1-PHOSPHATE AND SKELETAL MUSCLE FATIGUE
AJP-Cell Physiol • VOL
37. Pyne S and Pyne NJ. Sphingosine 1-phosphate signalling in mammalian
cells. Biochem J 349: 385– 402, 2000.
38. Sabbadini RA, Betto R, Teresi A, Fachechi-Cassano G, and Salviati G.
The effects of sphingosine on sarcoplasmic reticulum membrane calcium
release. J Biol Chem 267: 15475–15484, 1992.
39. Sabbadini R, McNutt W, Jenkins G, Betto R, and Salviati G. Sphingosine is endogenous to cardiac and skeletal muscle. Biochem Biophys Res
Commun 193: 752–758, 1993.
40. Sabbadini RA, Danieli-Betto D, and Betto R. The role of sphingolipids
in the control of skeletal muscle function: a review. Ital J Neurol Sci 20:
423– 430, 1999.
41. Salvatori S, Furlan S, Millikin B, Sabbadini R, Betto R, Margreth A,
and Salviati G. Localization of protein kinase C in skeletal muscle
T-tubule membranes. Biochem Biophys Res Commun 196: 1073–1080,
1993.
42. Sharma C, Smith T, Li S, Schroepfer GJ Jr, and Needleman DH.
Inhibition of Ca2⫹ release channel (ryanodine receptor) activity by sphingolipid bases: mechanism of action. Chem Phys Lipids 104: 1–11, 2000.
43. Spiegel S and Milstien S. Sphingosine 1-phosphate: an enigmatic signalling lipid. Nat Rev Mol Cell Biol 4: 397– 407, 2003.
44. Yasui K and Palade P. Sphingolipid actions on sodium and calcium
currents of rat ventricular myocytes. Am J Physiol Cell Physiol 270:
C645–C649, 1996.
45. Yatomi Y, Igarashi Y, Yang L, Hisano N, Qi R, Asazuma N, Satoh K,
Ozaki Y, and Kume S. Sphingosine 1-phosphate, a bioactive sphingolipid
abundantly stored in platelets, is a normal constituent of human plasma
and serum. J Biochem (Tokyo) 121: 969 –973, 1997.
46. Yatomi Y, Ruan F, Megidish T, Toyokuni T, Hakomori S, and
Igarashi Y. N,N-Dimethylsphingosine inhibition of sphingosine kinase
and sphingosine 1-phosphate activity in human platelets. Biochemistry 35:
626 – 633, 1996.
47. Yatomi Y, Welch RJ, and Igarashi Y. Distribution of sphingosine
1-phosphate, a bioactive sphingolipid, in rat tissues. FEBS Lett 404:
173–174, 1997.
48. Young KW, Willets JM, Parkinson MJ, Bartlett P, Spiegel S, Nahorski SR, and Challiss RAJ. Ca2⫹/calmodulin-dependent translocation of
sphingosine kinase: role in plasma membrane relocation but not activation.
Cell Calcium 33: 119 –128, 2003.
49. Wang J, Carbone LD, and Watsky MA. Receptor-mediated activation
of a Cl⫺ current by LPA and S1P in cultured corneal keratocytes. Invest
Ophthalmol Vis Sci 43: 3202–3208, 2002.
50. Westerblad H and Allen DG. Mechanisms underlying changes of tetanic
[Ca2⫹]i and force in skeletal muscle. Acta Physiol Scand 156: 407– 416,
1996.
51. Westerblad H, Lee JA, Lannergren J, and Allen DG. Cellular mechanisms of fatigue in skeletal muscle. Am J Physiol Cell Physiol 261:
C195–C209, 1991.
288 • JUNE 2005 •
www.ajpcell.org
Downloaded from ajpcell.physiology.org on July 1, 2009
21. Hannun YA and Obeid LM. The ceramide-centric universe of lipidmediated cell regulation: stress encounters of the lipid kind. J Biol Chem
277: 25847–25850, 2002.
22. Hida H, Takeda M, and Soliven B. Ceramide inhibits inwardly rectifying
K⫹ currents via a Ras- and Raf-1-dependent pathway in cultured oligodendrocytes. J Neurosci 18: 8712– 8719, 1998.
23. Himmel HM, Meyer zu Heringdorf D, Windorfer B, van Koppen CJ,
Ravens U, and Jakobs KH. Guanine nucleotide-sensitive inhibition of
L-type Ca2⫹ current by lysosphingolipids in RINm5F insulinoma cells.
Mol Pharmacol 53: 862– 869, 1998.
24. Itagaki K and Hauser CJ. Sphingosine 1-phosphate, a diffusible calcium
influx factor mediating store-operated calcium entry. J Biol Chem 278:
27540 –27547, 2003.
25. Kluk MJ and Hla T. Signaling of sphingosine-1-phosphate via the
S1P/EDG-family of G-protein-coupled receptors. Biochim Biophys Acta
1582: 72– 80, 2002.
26. Kolesnick R and Golde DW. The sphingomyelin pathway in tumor
necrosis factor and interleukin-1 signaling. Cell 77: 325–328, 1994.
27. Maceyka M, Payne SG, Milstien S, and Spiegel S. Sphingosine kinase,
sphingosine-1-phosphate, and apoptosis. Biochim Biophys Acta 1585:
193–201, 2002.
28. Margreth A, Damiani E, and Tobaldin G. Ratio of dihydropyridine to
ryanodine receptors in mammalian and frog twitch muscles in relation to
the mechanical hypothesis of excitation-contraction coupling. Biochem
Biophys Res Commun 197: 1303–1311, 1993.
29. McDonough PM, Yasui K, Betto R, Salviati G, Glembotski CC,
Palade PT, and Sabbadini RA. Control of cardiac Ca2⫹ levels: inhibitory
actions of sphingosine on Ca2⫹ transients and L-type Ca2⫹ channel
conductance. Circ Res 75: 981–989, 1994.
30. Meacci E, Cencetti F, Donati C, Nuti F, Becciolini L, and Bruni P.
Sphingosine kinase activity is required for sphingosine-mediated phospholipase D activation in C2C12 myoblasts. Biochem J 381: 655– 663,
2004.
31. Meissner G. Ryanodine receptor/Ca2⫹ release channels and their regulation by endogenous effectors. Annu Rev Physiol 56: 485–508, 1994.
32. Midrio M, Danieli-Betto D, Megighian A, and Betto R. Early effects of
denervation on sarcoplasmic reticulum properties of slow-twitch rat muscle fibres. Pflügers Arch 434: 398 – 405, 1997.
33. Nava VE, Lacana E, Poulton S, Liu H, Sugiura M, Kono K, Milstien
S, Kohama T, and Spiegel S. Functional characterization of human
sphingosine kinase-1. FEBS Lett 473: 81– 84, 2000.
34. Needleman DH, Aghdasi B, Seryshev AB, Schroepfer GJ Jr, and
Hamilton SL. Modulation of skeletal muscle Ca2⫹-release channel activity by sphingosine. Am J Physiol Cell Physiol 272: C1465–C1474, 1997.
35. Olivera A and Spiegel S. Sphingosine kinase: a mediator of vital cellular
functions. Prostaglandins 64: 123–134, 2001.
36. Pedersen BK, Ostrowski K, Rohde T, and Bruunsgaard H. The
cytokine response to strenuous exercise. Can J Physiol Pharmacol 76:
505–511, 1998.
C1373