J Appl Physiol 103: 1412–1418, 2007.
First published August 2, 2007; doi:10.1152/japplphysiol.00288.2007.
Expression and activity of cyclooxygenase isoforms in skeletal muscles
and myocardium of humans and rodents
Marco Testa,1 Bianca Rocca,2 Lucia Spath,3 Franco O. Ranelletti,4 Giovanna Petrucci,4
Giovanni Ciabattoni,5 Fabio Naro,3 Stefano Schiaffino,6 Massimo Volpe,1,7,8 and Carlo Reggiani9
1
Department of Cardiology, Sant’Andrea Hospital, Departments of 2Pharmacology and 7Cardiology, II School of Medicine,
University of Rome “La Sapienza,” 3Department of Histology, I School of Medicine, University of Rome “La Sapienza,”
4
Department of Pathology, Catholic University, Rome; 5Department of Pharmacy, University of Chieti, Chieti; 8IRCCS
Neuromed, Pozzilli; and Departments of 6Biomedical Sciences and 9Human Anatomy and Physiology, University of Padua,
Padua, Italy
Submitted 13 March 2007; accepted in final form 11 July 2007
muscle; myocardium; prostaglandins; cyclooxygenases
hemostasis, reproduction, kidney function, gastric protection,
and inflammation are well known. TxA2 is a vasoconstrictor
and pro-aggregatory agent, PGI2 is a vasodilating and endothelium-protective compound, and PGE2 mediates inflammation (21). On the contrary, the possible presence and function
of COX and prostanoids in striated muscles are less defined.
More than 20 years ago, Young and Sparks (35) observed that
skeletal muscles could release PGE2, and similar data were
reported by Rodemann and Goldberg (22). Physical exercise
has been shown to increase PGE2 synthesis in normal muscles
(10, 31, 32). Elevated PGE2 synthesis has been documented in
muscles from patients with Duchenne muscular dystrophy or
from mdx mice (9, 13), as well as in regenerating skeletal
muscles of rodents (12). Nevertheless, it remains presently
unknown whether muscle fibers or other cell types (fibroblasts,
macrophages, cells of the vessel wall) account for PG release
and which COX isoform is present in striated muscles, both
myocardial and skeletal. Furthermore, PGs function in the
pathophysiology of muscle contraction, differentiation, and
regeneration remains unclear.
We sought to characterize the pattern of expression of each
COX isoform in striated muscles, including different skeletal
muscles and myocardium, from three different species: mouse,
rat, and human. Together with the expression, we also investigated the selective activity of each COX isoform as well as
the effects of the major product of COX activity on a rodent’s
muscle contraction.
MATERIALS AND METHODS
(AA) through the
sequential enzymatic activity of cyclooxygenases (COX)-1 or
-2 and of terminal synthases. Both COX-1 and -2 catalyze the
conversion of AA into an unstable intermediate, the prostaglandin (PG) G/H2, which is then transformed by distinct
terminal synthases into PGE2, PGF2, thromboxane (Tx) A2,
PGD2, or PGI2.
COX-1 and COX-2, are encoded by different genes, are
highly homologous, and catalyze the same reaction, although
they play different roles even within the same cell or tissue
(28). The role of COX-1 or -2 and of different prostanoids in
Tissue sampling. Skeletal muscles and hearts were obtained from
adult Wistar rats (weighing 130 –160 g) and adult CD1 mice (weighing 45– 60 g). Briefly, under deep anesthesia, animals were exsanguinated by carotid cutting and then killed by cervical dislocation. Heart,
intact soleus, and extensor digitorum longus (EDL) with short pieces
of tendons were excised. Small biopsies of normal human skeletal
muscles were obtained during orthopedic surgery after informed
consent of patients was obtained. For mRNA studies, tissue samples
were immediately frozen in liquid nitrogen and kept at ⫺80°C until
use. For immunohistochemistry, tissue samples were immediately
mounted in OCT, frozen in liquid nitrogen, and kept at ⫺80°C until
use or were immediately placed into buffered formalin and then
embedded in paraffin using standard procedures. Human heart samples were obtained from paraffinated samples of the Archives of the
Address for reprint requests and other correspondence: M. Testa, Dept. of
Cardiology, Azienda Ospedaliera Sant’Andrea, Via di Grottarossa 1035, 00189
Roma, 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.
PROSTANOIDS DERIVE FROM ARACHIDONIC ACID
1412
8750-7587/07 $8.00 Copyright © 2007 the American Physiological Society
http://www. jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.5 on June 17, 2017
Testa M, Rocca B, Spath L, Ranelletti FO, Petrucci G, Ciabattoni
G, Naro F, Schiaffino S, Volpe M, Reggiani C. Expression and activity
of cyclooxygenase isoforms in skeletal muscles and myocardium of
humans and rodents. J Appl Physiol 103: 1412–1418, 2007. First published August 2, 2007; doi:10.1152/japplphysiol.00288.2007.—Conflicting data have been reported on cyclooxygenase (COX)-1 and COX-2
expression and activity in striated muscles, including skeletal muscles
and myocardium, in particular it is still unclear whether muscle cells
are able to produce prostaglandins (PGs). We characterized the
expression and enzymatic activity of COX-1 and COX-2 in the
skeletal muscles and in the myocardium of mice, rats and humans. By
RT-PCR, COX-1 and COX-2 mRNAs were observed in homogenates
of mouse and rat hearts, and in different types of skeletal muscles
from all different species. By Western blotting, COX-1 and -2 proteins were detected in skeletal muscles and hearts from rodents, as
well as in skeletal muscles from humans. Immunoperoxidase stains
showed that COX-1 and -2 were diffusely expressed in the myocytes
of different muscles and in the myocardiocytes from all different
species. In the presence of arachidonic acid, which is the COX
enzymatic substrate, isolated skeletal muscle and heart samples from
rodents released predominantly PGE2. The biosynthesis of PGE2 was
reduced between 50 and 80% (P ⬍ 0.05 vs. vehicle) in the presence
of either COX-1- or COX-2-selective blockers, demonstrating that
both isoforms are enzymatically active. Exogenous PGE2 added to
isolated skeletal muscle preparations from rodents did not affect
contraction, whereas it significantly fastened relaxation of a slow type
muscle, such as soleus. In conclusion, COX-1 and COX-2 are expressed and enzymatically active in myocytes of skeletal muscles and
hearts of rodents and humans. PGE2 appears to be the main product of
COX activity in striated muscles.
STRIATED MUSCLE AND CYCLOOXYGENASES
J Appl Physiol • VOL
of their resting length. Every incubation was carried out in 1 ml of
medium, equilibrated with 95% O2-5% CO2 before use, and kept at
37°C under 95% O2-5% CO2 throughout the experiment. All muscles
were incubated following the same protocols.
To assess COX activity, skeletal muscles were subjected to two
subsequent incubations of 30 min each. After the last incubation, the
medium was removed and frozen for prostanoids measurements. In
different experiments, skeletal muscles and myocardium were preincubated for 30 min with Krebs buffer, preincubation medium was then
removed, and the same buffer with 20 ␮M AA was added for 30 min,
and the supernatants were frozen.
In experiments with COX inhibitors, samples were preincubated
for 30 min with Krebs buffer containing 20 ␮M indomethacin (a
nonselective COX-1 and -2 inhibitor), 100 nM SC-560 (a selective
COX-1 inhibitor), 100 nM NS-398 (a selective COX-2 inhibitor) (all
from Cayman Chemicals), or vehicle; medium was then removed and
the same buffer containing 20 ␮M AA and the correspondent inhibitor
or vehicle were incubated for 30 min. At the end of any incubation,
medium was collected and frozen for prostanoid measurements.
Determination of PGE2 and TxB2. PGE2 and TxB2 were determined in the medium of muscle incubation by using previously
validated radioimmunoassays (2).
Muscle contraction studies. Murine soleus and EDL muscles were
dissected and transferred to the myograph, where they were mounted
between the hook of a force transducer (AME 801; Aksjeselkapet
Mikkroelektronik, Norway) and a hook connected to a movable shaft
used to adjust muscle length. Muscles were tied to the hooks by means
of silk threads that had been previously fixed to the tendons. The
perfusion bath (2 ml) was filled with bicarbonate Krebs solution,
bubbled with a O2-CO2 mixture kept at constant temperature (25°C).
The perfusing solution could be quickly renewed (5–10 s) by flushing
through the chamber ⬃20 ml of the new solution. On both sides of the
perfusion bath at a distance of ⬃2 mm from the preparations, plate
platinum electrodes connected with a Grass stimulator allowed field
electrical stimulation. The preparations were stretched just above
slack length, and their response to electrical stimulation was tested.
Each preparation was stimulated for ⬃30 min with supramaximal,
low-frequency (0.1 Hz) stimuli. The influence of PGE2 (between
100 nM and 10 ␮M) on the parameters of the isometric twitch
(peak tension, time to peak and time to half relaxation) was
investigated. The concentrations used were in the range previously
used in vitro by other authors on airway smooth muscles (30) and
myometrium (18). The output of the tension transducer was stored in
a personal computer after analog-to-digital conversion and was recalled for analysis. Cambridge Electronic Design 1401 analog-todigital converter and CEA Spike2 software (CED, Cambridge, UK)
were used.
Statistical analysis. Values are reported as means ⫾ SD or SE, as
indicated. Statistical significance of the differences between means
was assessed by analysis of variance followed by the Student-Newman-Keuls test or by Student’s t-test for paired or nonpaired data, as
indicated. Statistical significance was set at P ⬍ 0.05.
RESULTS
COX expression and activity in skeletal muscles. Messenger
RNAs for COX-1 and COX-2 were detected in samples from
different muscle types of mice, rats, and humans (Fig. 1).
Consistently, Western blots showed immunoreactive COX-1
and COX-2 of the expected size (i.e., ⬃70 kDa) in muscle homogenates from rat, mouse, and human samples
(Fig. 2, A and B). Densitometric analysis of the Western-blot
bands, expressed as ratios of COX-1 or -2 vs. tubulin bands,
showed no major differences in levels of protein expression
among different types of muscles within the same species,
including predominantly slow (soleus), fast (EDL, tibialis
103 • OCTOBER 2007 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.5 on June 17, 2017
Pathology Department, originated from postmortem hearts in subjects
who died from noncardiac causes.
The animals were treated according to the Convention of Helsinki
on the utilization of animals in biomedical research and according to
the guidelines of the American Physiological Society. All experiments
were approved by the Ethical Commission of the Department of
Anatomy and Physiology, University of Padova.
RNA preparation. Total RNA was isolated from frozen tissues by
homogenization in Trizol Reagent (Life Technologies, Gaithersburg,
MD) following the manufacturer’s protocol. One microgram of total
RNA was reverse transcribed using Moloney murine leukemia virus
RT (GIBCO-Invitrogen, Carlsbad CA). PCR reactions were carried
out by standard technique using Taq PCRx DNA Polymerase
(GIBCO-Invitrogen, Carlsbad, CA) under the following conditions:
denaturation at 95°C for 30 s, annealing at 50 –55°C (for different
amplifications) for 30 s, and extension at 68°C for 40 s after the initial
denaturation at 95°C for 2 min (35 cycles). The following oligonucleotides were used to amplify COX-1, COX-2 and ␤-actin cDNA:
human COX-1 (forward 5⬘TTCTTGCTGTTCCTGCTCCTG3⬘;
reverse 5⬘GCATTGACAAACTCCCAGAAC3⬘); human COX-2 (forward 5⬘TCCTGGCGCTCAGCCATACA3⬘; reverse 5⬘GTAGCCATAGTCAGCATTGTA3⬘); mouse COX-1 (forward 5⬘GGTCCTGCTCGCAGATCCTG3⬘; reverse 5⬘AGGACCCATCTTTCCAGAGG3⬘);
mouse COX-2 (forward 5⬘CTGTACAAGCAGTGGCAA3⬘; reverse
5⬘TTACAGCTCAGTTGAACGCCT3⬘); rat COX-1 (forward 5⬘CCTTCCGTGTGCCAGATTAC3⬘; reverse 5⬘GGCTGGCCTAGAACTCACTG3⬘); rat COX-2 (forward 5⬘ACACTCTATCACTGGCATCC3⬘;
reverse 5⬘GAAGGGACACCCTTTCACAT3⬘); ␤-actin (forward
5⬘ACCAACTGGGACGACATGGAG3⬘; reverse 5⬘GACTACCTCATGAAGATCCTGACC3⬘). Integrity and equal loading of cDNA
in the PCR reactions were checked by quantification of ␤-actin. The
expected sizes of the amplicons were: 580 base pairs (bp) for mouse
COX-1; 530 bp for mouse COX-2; 475 bp for rat COX-1; 585 bp for
rat COX-2; 286 bp for human COX-1; 334 bp for human COX-2, 380
bp for mouse, rat, and human actin.
Western blotting. Frozen muscles were homogenized in ice-cold
RIPA buffer (10 mM Tris 䡠 HCl, pH 7.5, 10 mM EDTA, 0.5 M NaCl,
1% NP40) containing a protease inhibitors cocktail (Roche Molecular
Biochemicals, Manneim, Germany). Protein (30 ␮g) were separated
by SDS-PAGE on a 8% polyacrylamide gel, blotted onto a nitrocellulose membrane, and probed with anti-COX-1 and anti-COX-2
monoclonal antibodies (Cayman Chemicals, Ann Arbor, MI). Proteins
were assayed by the Bradford method. To verify equal loading, the
blotted membrane were probed also with anti-tubulin or actin monoclonal antibodies (Sigma-Aldrich, St. Louis, MO), as specified. After
incubation with peroxidase-conjugated anti-rabbit or anti-mouse IgG
(Sigma-Aldrich), reactions were revealed with the ECL reagent (Amersham, Arlington Heights, IL). Densitometric analysis of the bands
were performed by a Diana 95.1 camera (Raytest Isotopenmebgerate,
Straubenhardt, Germany) and analyzed by the Aida 2.1 software
(Raytest Isotopenmebgerate).
Immunohistochemistry. Tissue slides were rehydrated, treated with
0.3% H2O2 in methanol for 10 min to block endogenous peroxidase
and incubated for 1 h with one of the following primary Abs:
monoclonal BA-D5 (25) specific for type I skeletal muscle fibers,
monoclonal SC-71, specific for type IIa skeletal muscle fibers (25),
anti-COX-1 polyclonal Ab (Cayman Chemicals, Ann Arbor, MI),
anti-COX-2 polyclonal Ab (Cayman Chemicals). Rabbit and/or
mouse preimmune IgGs were used as negative controls. Indirect
immunoperoxidase stainings were performed using the ABC-peroxidase technique (Vector Laboratories, Burlingame, CA) developed
with the DAB substrate kit (Vector Laboratories).
COX activity and inhibitor studies. Immediately after excision,
soleus, EDL, and small fragments of myocardium (15–20 mg) were
washed twice in Krebs-Henseleit buffer (Sigma-Aldrich) to eliminate
blood and transferred to incubation chamber. Rat and mouse skeletal
muscles were mounted and gently stretched to maintain them at 100%
1413
1414
STRIATED MUSCLE AND CYCLOOXYGENASES
Fig. 1. Expression of mRNA for cyclooxygenase
(COX)-1 and -2 in striated muscles and in the heart.
RT-PCR samples amplified with primers for COX-1,
COX-2, or actin in mouse, rat, or human samples, as
indicated. EDL, extensor digitorum longus; H, heart; K,
kidney (positive control); mw, molecular weight markers; RA, rectus abdominis; Sol, soleus. The amplified
fragments were of the expected molecular weight (see
MATERIALS AND METHODS for details).
We next investigated the enzymatic activity of COX isoenzymes in striated muscles. In a first set of experiments using
soleus and EDL from mouse and rat, we observed that muscles
released both PGE2 and TxB2, with PGE2 being approximately
six times more abundant than TxB2 in both species (for
example: rat EDL released 9.4 ⫾ 3.3 pg/mg of tissue of PGE2
and 0.9 ⫾ 0.08 pg/mg of tissue of TXB2; rat soleus released
6.5 ⫾ 1.9 pg/mg of tissue of PGE2 and 1.9 ⫾ 0.9 pg/mg of
tissue of TxB2; n ⫽ 3 each determination). Thereafter, we
measured PGE2 released in 30-min incubation (Fig. 6). The
amounts of PGE2 released by rat and mouse muscles were four
to five times higher in the presence of AA loaded to the
medium (rat EDL: 10.3 ⫾ 3.6 vs. 1.4 ⫾ 0.6 pg/mg muscle,
with and without AA, respectively, n ⫽ 4; mouse soleus: 43 ⫾
14 vs. 13 ⫾ 9 pg/mg muscle, with and without AA, respectively, n ⫽ 4). The predominance of PGE2 indirectly demonstrated that muscle specimens were properly washed and depleted of blood and platelets, with TxA2 being the main
platelet-derived product instead (21). The amount of PGE2
released from mouse soleus and EDL, normalized for muscle’s
weight, was five to eight times higher than that released from
Fig. 2. Expression of COX-1 and -2 proteins
in rodent and human striated muscles. A and
B: Western blots of mouse, rat, or human
samples detected with anti COX-1, anti
COX-2, or anti-tubulin antibodies, as indicated
(see MATERIALS AND METHODS for details). C
and D: densitometric analyses of Western blot
bands for COX-1, COX-2, and tubulin of different muscles. Data are expressed as ratios vs.
tubulin band intensities, and are means ⫾ SD
of at least 3 different determinations. No significant variations were observed for either
COX-1 or COX-2 expression among different
muscles of the same species. Dia, diaphragm.
TA, tibialis anterior; tub, tubulin.
J Appl Physiol • VOL
103 • OCTOBER 2007 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.5 on June 17, 2017
anterior), or mixed (diaphragm, rectus abdominis) muscles
(Fig. 2, C and D).
To identify the cells expressing COX isoenzymes, samples
of skeletal muscles were studied by immunohistochemistry
using specific antibodies. COX-1 and COX-2 immunoreactivity was consistently observed in muscle fibers of rodents
(Figs. 3 and 4, A and B) and humans (Figs. 5, A and B). In the
same specimens, vascular smooth muscle cells, interstitial
fibroblasts, adipocytes (if present), and endothelial cells of
veins and smallest arteries were negative for either COX-1 or
-2, whereas endothelial cells of larger arteries, identified by a
thicker media, were positive for COX-1 and COX-2. This
pattern was consistent among all different species. We studied
different types of muscles, and the expression of both COX-1
and -2 protein was similar. To further assess whether COX
isoenzymes had any preferential expression in different types
of fibers, serial sections of rodent and human muscles were
stained with antibodies specific for type I fibers, type IIa fibers,
COX-1, or COX-2. No preferential expression of COX isoenzymes in either type I or type IIa fibers was detected (Figs. 3–5,
C and D).
STRIATED MUSCLE AND CYCLOOXYGENASES
rat muscles (Fig. 6). This difference might be attributed to the
different thickness of the muscles in the two species. Taking
into account the size of the rat muscles, it is likely that not the
whole muscles but only the outer layers of the muscles could
release the prostanoid into the incubation medium.
To discriminate the relative contribution of each COX isoform to PGE2 formation, muscle samples were preincubated
with the nonselective inhibitor indomethacin, the COX-1selective inhibitor SC-560, or the COX-2-selective inhibitor
NS-398. Among the three different inhibitors, indomethacin
(20 ␮M) showed the higher degree of inhibition of PGE2 in
both species and in all the muscles examined, as shown in Fig.
6. At variance with indomethacin, SC-560 and NS-398 at
concentrations known to be isoform selective such as 100 nM,
caused a lower degree of inhibition of PGE2 synthesis. Dose
response experiments showed similar degrees of inhibition
between 100 nM and 1 ␮M of both SC-560 and NS-398,
indicating that each isoform was maximally inhibited at those
Fig. 4. Immunoperoxidase staining of serial samples of mouse soleus.
A: COX-1. B: COX-2. C: type I myosin. D: type IIa myosin. Negative fibers
are visibile in C and D. Original magnification ⫽ ⫻20.
J Appl Physiol • VOL
Fig. 5. Immunoperoxidase staining of serial samples of human rectus abdominis. A: COX-1. B: COX-2. C: type I myosin. D: type IIa myosin. Negative
fibers are visibile in C and D. Original magnification ⫽ ⫻20.
concentrations (data not shown). We did not check concentrations higher than 1 ␮M because drug selectivity is lost by
increasing the concentrations (20). These experiments indicate
that both COX isoforms are enzymatically active and contribute to PGE2 synthesis in rodents, although COX-1 inhibition
Fig. 6. PGE2 released from rat (A) and mouse (B) muscles. PGE2 released in
30 min from soleus and EDL of rat and mouse was measured in the medium
with 20 ␮M AA loading in the presence of vehicle, indomethacin, SC-560, or
NS-398 (see MATERIALS AND METHODS for details). The amount of PGE2 was
normalized to muscle weight. Black bars indicate soleus muscle; white bars
indicate EDL muscle. Data are means ⫾ SD of 4 – 6 experiments in duplicate.
*P ⬍ 0.01 vs. vehicle.
103 • OCTOBER 2007 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.5 on June 17, 2017
Fig. 3. Immunoperoxidase staining of serial samples of rat soleus. A: COX-1.
B: COX-2. C: type I myosin. D: type IIa myosin. Negative fibers are visibile
in C and D. Original magnification ⫽ ⫻20.
1415
1416
STRIATED MUSCLE AND CYCLOOXYGENASES
Fig. 7. Effects of different concentrations of
PGE2 on contraction and relaxations of
mouse soleus and EDL muscles. A–C: histograms of means ⫾ SE (n ⫽ 3) of active
tension (A), time to peak tension (B), and
time to half relaxation (C) in the presence of
100 nM (open bars), 1 ␮M (crosshatched
bars), and 10 ␮M (filled bars) of PGE2 (see
MATERIALS AND METHODS for technical details). Values express % of vehicle-treated
samples. *P ⬍ 0.01 vs. vehicle. D and E:
effect of 10 ␮M PGE2 on the twitch of
soleus (D) and EDL (E). The solid lines
indicate contraction in control condition, and
the dashed lines indicate contraction in the
presence of 10 ␮M PGE2.
DISCUSSION
The present study demonstrates a constitutive expression of
both COX-1 and COX-2 in different types of muscles fibers
and in cardiomyocytes of three different species. At least in
rodents, COX isozymes are enzymatically active, being inhibJ Appl Physiol • VOL
ited by COX-1 or -2 selective blockers. Furthermore, in the
murine soleus, which is a slow type muscle, PGE2 in the low
micromolar range made relaxation significantly faster.
Whereas COX expression or PG generation has been previously described during myogenesis and skeletal muscle regeneration, more limited information is available on the role of
Fig. 8. Immunoperoxidase staining of myocardium samples from human
(A and B), mouse (C and D), and rat (E and F) samples. A, C, and E were
reacted with anti COX-1 antibodies; B, D, and F were reacted with anti-COX-2
antibodies and developed with peroxidase. Original magnification ⫽ ⫻20.
Inset in A: original magnification ⫽ ⫻40. *Vessels whose smooth muscle cells
within the tunica media do not react with anti-COX-1 or -2 antibodies.
103 • OCTOBER 2007 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.5 on June 17, 2017
consistently gave a higher degree of PGE2 reduction compared
with COX-2 inhibition (Fig. 6).
In vitro effects of exogenous PGE2 on muscle contraction.
The effects of PGE2 on the time and tension parameters of the
isometric twitch were investigated in vitro in both murine
soleus and EDL. No effects on active tension and on time to
peak tension were detected at concentrations up to 10 ␮M.
However, PGE2 induced a small but significant reduction in the
half-relaxation time that was evident in the slow muscle type,
i.e., the soleus, rather than in the EDL (Fig. 7).
COX expression and activity in the myocardium. Myocardial
samples from mouse and rat expressed mRNA and protein of
both COX-1 and -2, as shown by RT-PCR (Fig. 1) and Western
blot. By immunohistochemistry, COX-1 and COX-2 proteins
showed a diffuse pattern of expression in cardiomyocytes of
human, mouse, and rat hearts (Fig. 8).
Mouse and rat myocardial samples, loaded with 20 ␮M AA,
released 28 ⫾ 13 and 32.6 ⫾ 14.12 pg/mg of tissue of PGE2,
respectively (n ⫽ 6 for each species), in 30-min incubation.
Thus similar amounts of PGE2 were released from fragments
of mouse and rat myocardium.
In rat samples, SC-560 (100 nM) lowered the amount of
PGE2 released in 30 min to 6.4 ⫾ 3.4 pg/mg of tissue,
corresponding to an 80% inhibition (P ⬍ 0.05), whereas
NS-398 (100 nM) reduced the production to 12.2 ⫾ 4.4 pg/mg
of tissue, corresponding to a 63% inhibition (P ⬍ 0.05). In
mouse samples, SC-560 significantly inhibited PGE2 release
by 59% on average, whereas NS-398 caused an average 80%
inhibition. These data, showing that each selective COX inhibitor caused an incomplete inhibition of PGE2 production
from heart muscle, indicate that both isoforms are enzymatically active and contribute to PGE2 generation in the myocardium.
STRIATED MUSCLE AND CYCLOOXYGENASES
J Appl Physiol • VOL
cytes is quite novel. Although we recognize that our observations are on postmortem tissues, since the collection of fresh
and healthy myocardium is unethical, the consistency of a
constitutive expression of COX-2 among three different species indirectly supports the likelihood of our finding in living
human hearts as well. Wong et al. (36) described a strong
expression of COX-2 in failing human hearts but found no
expression in control hearts. This discrepancy might be due to
different sources of antibodies used in Wong et al.’s and our
experiments. On the contrary, our data are in agreement with
Liu and coworkers (11), who found a constitutive expression of
both COX-1 and -2 in rat hearts, and this expression was
enhanced by lipopolysaccharide infused in vivo. We report that
PGE2 was the main PG released by rodent hearts via COX-1
and COX-2. The prevalence of PGE2 is consistent with data
reported in vitro in cultured neonatal rat cardiomyocytes (16).
In conclusion, the present study demonstrates that COX
isozymes are both expressed and enzymatically active in striated muscles, both in myocardial and skeletal muscles. The
predominant PG produced is PGE2. Our data also suggest a
modulation of slow skeletal muscle relaxation by PGE2.
GRANTS
This study was supported in part by the European Commission FP6 grant
(LSHM-CT-2004-0050333). This publication reflects only the author’s views.
The Commission is not liable for any use that may be made of information
herein.
REFERENCES
1. Bondesen BA, Mills ST, Kegley KM, Pavlath GK. The COX-2 pathway
is essential during early stages of skeletal muscle regeneration. Am J
Physiol Cell Physiol 287: C465–C483, 2004.
2. Ciabattoni G, Pugliese F, Spaldi M, Cinotti GA, Patrono C. Radioimmunoassay measurement of prostaglandins E2 and F2alpha in human
urine. J Endocrinol Invest 2: 173–182, 1979.
3. Clifford PS, Hellsten Y. Vasodilatory mechanisms in contracting skeletal
muscle. J Appl Physiol 97: 393– 403, 2004.
4. David JD, Higginbotham CA. Fusion of chick embryo skeletal myoblasts: interactions of prostaglandin E1, adenosine 3⬘:5⬘ monophosphate,
and calcium influx. Dev Biol 82: 308 –316, 1981.
5. Dupouy VM, Ferre PJ, Uro-Coste E, Lefebvre HP. Time course of
COX-1 and COX-2 expression during ischemia-reperfusion in rat skeletal
muscle. J Appl Physiol 100: 233–239, 2006.
6. Entwistle A, Curtis DH, Zalin RJ. Myoblast fusion is regulated by a
prostanoid of the one series independently of a rise in cyclic AMP. J Cell
Biol 103: 857– 866, 1986.
7. Fortner CN, Breyer RM, Paul RJ. EP2 receptors mediate airway
relaxation to substance P, ATP, and PGE2. Am J Physiol Lung Cell Mol
Physiol 281: L469 –L674, 2001.
8. Froemming GR, Murray BE, Harmon S, Pette D, Ohlendieck K.
Comparative analysis of the isoform expression pattern of Ca2⫹-regulatory
membrane proteins in fasttwitch, slow-twitch, cardiac, neonatal, and
chronic low-frequency stimulated muscle fibers. Biochim Biophys Acta
1466: 151–168, 2000.
9. Jackson MJ, Brooke MH, Kaiser K, Edwards RH. Creatine kinase and
prostaglandin E2 release from isolated Duchenne muscle. Neurology 41:
101–104, 1991.
10. Karamouzis M, Langberg H, Skovgaard D, Bulow J, KjaerM, Saltin
B. In situ microdialysis of intramuscular prostaglandin and thromboxane
in contracting skeletal muscle in humans. Acta Physiol Scand 171: 71–76,
2001.
11. Liu SF, Newton R, Evans TW, Barnes PJ. Differential regulation of
cyclo-oxygenase-1 and cyclo-oxygenase-2 gene expression by lipopolysaccharide treatment in vivo in the rat. Clin Sci (Lond) 90: 301–306, 1996.
12. McArdle A, Edwards RH, Jackson MJ. Release of creatine kinase and
prostaglandin E2 from regenerating skeletal muscle fibers. J Appl Physiol
76: 1274 –1278, 1994.
103 • OCTOBER 2007 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.5 on June 17, 2017
PGs in adult skeletal muscle physiology and on the expression
in skeletal muscle fibers of the enzymes responsible for PG
production. More than 20 years ago, Rodemann and coworkers
(22, 23) suggested a role for exogenous PGE2 and PGF2␣ in the
acceleration of muscle protein turnover, well before the discovery of COX-2. A generic role for COX-derived compounds
on protein metabolism in skeletal muscle has also been hypothesized by some authors to explain the analgesic effect of
ibuprofen and acetaminophen after eccentric exercise in humans without characterizing the expression and the type of
COX isoenzymes involved (31, 32). A role for PGs in muscle
cell fusion and growth in different animal species has also been
reported (4, 6, 14, 17, 24, 29, 33, 38). More recently, it has
been reported that COX-2 is required during mouse skeletal
muscle regeneration after different models of injury, using
selective COX-2 inhibitors and/or COX-2-deleted mice (1, 5,
27), although none of those studies assessed the type of PG
involved. Data indicating a role for the AA cascade in regeneration is possibly consistent with our data, showing only
minor modulation on contractile properties, which might indirectly indicate a role for PGs in muscle housekeeping functions, such as muscle trophism or repair. Consistently, a possible involvement of PGs in degenerative muscle diseases is
suggested by the observation of an increased production in
muscles of patients with Duchenne muscular dystrophy or of
mdx mice (9, 13). PGE2 is known to cause relaxation via the
EP2 receptors in vascular (37) and airway smooth muscles (7,
26, 30). Arterial infusion of PGE2 leads to a marked increase in
blood flow in the human forearm as well as in canine skeletal
muscle arteries. Divergent findings on the role of prostanoids
in exercise hyperemia have been reported (3). Altogether, these
observations might imply a role of PGE2 in the regulation of
muscle microcirculation.
No data are available on a possible role of PGE2 on skeletal
muscle contraction. Our data show a modest reduction of
half-relaxation time induced by PGE2 in the low micromolar
range in the slow soleus muscle, whereas no effect was observed in the fast EDL muscle. This diversity could be explained by the difference among the slow and fast muscle fibers
in controlling cytosolic calcium. Slow muscle fibers share with
cardiac and smooth muscle, although to a limited extent, the
regulation of sarcoplasmic reticulum calcium pump via phospholamban (34) and the involvement of transarcolemmal calcium movements in the transient increase, which accompanies
contraction (8, 19). Further study will clarify the physiological
relevance.
Our data showed that cardiomyocytes also express enzymatically active COX-1 and -2 in different species, extending
previous reports from different groups. Using cultures of rat
neonatal ventricular myocytes, Mendez and Lapointe (15, 16)
demonstrated an induction of COX-2 in vitro, but they did not
find expression in cultured resting cells. At variance with these
observations, we found a constitutive presence of COX-2 in
cardiomyocytes of young adult rats and mice at both RNA and
protein levels. A possible explanation is that in vitro conditions
might downregulate COX-2 gene expression or that neonatal
myocytes have a different COX-2 expression compared with
fully differentiated adult cells. It would be interesting to
explore which is the trigger in the heart for a physiological
constitutive COX-2 expression in cardiomyocytes. The report
of a constitutive expression of COX-2 in human cardiomyo-
1417
1418
STRIATED MUSCLE AND CYCLOOXYGENASES
J Appl Physiol • VOL
26. Sheller JR, Mitchell D, Meyrick B, Oates J, Breyer R. EP2 receptor
mediates bronchodilation by PGE2 in mice. J Appl Physiol 88: 2214 –
2218, 2000.
27. Shen W, Li Y, Tang Y, Cummins J, Huard J. NS-398, a cyclooxygenase-2-specific inhibitor, delays skeletal muscle healing by decreasing
regeneration and promoting fibrosis. Am J Pathol 167: 1105–1117, 2005.
28. Simmons DL, Botting RM, Hla T. Cyclooxygenase isozymes: the
biology of prostaglandin synthesis and inhibition. Pharmacol Rev 56:
387– 437, 2004.
29. Templeton GH, Padalino M, Moss R. Influences of inactivity and
indomethacin on soleus phosphatidylethanolamine and size. Prostaglandins 31: 545–559, 1986.
30. Tilley SL, Hartney JM, Erikson CJ, Jania C, Nguyen M, Stock J,
McNeisch J, Valancius C, Panettieri RA Jr, Penn RB, Koller BH.
Receptors and pathways mediating the effects of prostaglandin E2 on
airway tone. Am J Physiol Lung Cell Mol Physiol 284: L599 –L606, 2003.
31. Trappe TA, Fluckey JD, White F, Lambert CP, Evans WJ. Skeletal
muscle PGF2␣ and PGE2 in response to eccentric resistance exercise:
influence of ibuprofen and acetaminophen. J Clin Endocrinol Metab 86:
5067–5070, 2001.
32. Trappe TA, White F, Lambert CP, Cesar D, Hellerstein M, Evans WJ.
Effect of ibuprofen and acetaminophen on postexercise muscle protein
synthesis. Am J Physiol Endocrinol Metab 282: E551–E556, 2002.
33. Vandenburgh HH, Hatfaludy S, Sohar I, Shansky J. Stretch-induced
prostaglandins and protein turnover in cultured skeletal muscle. Am J
Physiol Cell Physiol 259: C232–C240, 1990.
34. Vangheluwe P, Schuermans M, Zador E, Waelkens E, Raeymaekers
L, Wuytack F. Sarcolipin and phospholamban mRNA and protein expression in cardiac and skeletal muscle of different species. Biochem J
389: 151–159, 2005.
35. Young EW, Sparks HV. Prostaglandin E release from dog skeletal
muscles during restricted flow exercise. Am J Physiol Heart Circ Physiol
236: H596 –H599, 1979.
36. Wong SC, Fukuchi M, Melnyk P, Rodger I, Giaid A. Induction of
cyclooxygenase-2 and activation of nuclear factor-kappaB in myocardium
of patients with congestive heart failure. Circulation 98: 100 –103, 1998.
37. Zhang Y, Guan Y, Schneider A, Brandon S, Breyer RM, Breyer MD.
Characterization of murine vasopressor and vasodepressor prostaglandin
E(2)receptors. Hypertension 35: 1129 –1134, 2000.
38. Zalin RJ. Prostaglandins and myoblast fusion. Dev Biol 59: 241–248,
1977.
103 • OCTOBER 2007 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.5 on June 17, 2017
13. McArdle A, Edwards RHT, Jackson MJ. Effects of contractile activity
on muscle damage in the dystrophin-deficient mdx mouse. Clin Sci (Lond)
80: 367–371, 1991.
14. McLennan IS. Hormonal regulation of myoblast proliferation and myotube production in vivo: influence of prostaglandins. J Exp Zool 241:
237–245, 1987.
15. Mendez M, LaPointe MC. PGE2-induced hypertrophy of cardiac myocytes involves EP4 receptor-dependent activation of p42/44 MAPK and
EGFR transactivation. Am J Physiol Heart Circ Physiol 288: H2111–
H2117, 2005.
16. Mendez M, LaPointe MC. Trophic effects of the cyclooxygenase-2
product prostaglandin E2 in cardiac myocytes. Hypertension 39: 382–388,
2002.
17. Otis JS, Burkholder TJ, Pavlath GK. Stretch-induced myoblast proliferation is dependent on the COX2 pathway. Exp Cell Res 310: 417– 425,
2005.
18. Parkington HC, Tonta MA, Davies NK, Brennecke SP, Coleman HA.
Hyperpolarization and slowing of the rate of contraction in human uterus
in pregnancy by prostaglandins E2 and f2alpha: involvement of the Na⫹
pump. J Physiol 514: 229 –243, 1999.
19. Pereon Y, Dettbarn C, Lu Y, Westlund KN, Zhang JT, Palade P.
Dihydropyridine receptor isoform expression in adult rat skeletal muscle.
Pflügers Arch 436: 309 –314, 1998.
20. Rocca B, Spain LM, Ciabattoni G, Patrono C, FitzGerald GA. Differential expression and regulation of cyclooxygenase isozymes in thymic
stromal cells. J Immunol 162: 4589 – 4597, 1999.
21. Rocca B, Patrono C. Determinants of the interindividual variability in
response to antiplatelet drugs. J Thromb Haemost 3: 1597–1602, 2005.
22. Rodeman HP, Goldberg AL. Arachidonic acid, PGE2 and F2␣ influence
rates of protein turnover in skeletal and cardiac muscle. J Biol Chem 257:
1632–1638, 1982.
23. Rodeman HP, Waxamn L, Goldberg AL. The stimulation of protein
degradation in muscle by Ca2⫹ is mediated by prostaglandin E2 and does
not require the calcium-activated protease. J Biol Chem 257: 8716 – 8723,
1982.
24. Rossi MJ, Clark MA, Steiner SM. Possible role of prostaglandins in the
regulation of mouse myoblasts. J Cell Physiol 141: 142–147, 1989.
25. Schiaffino S, Gorza L, Sartore S, Saggin L, Ausoni S, Vianello M,
Gundersen K, Lomo T. Three myosin heavy chain isoforms in type 2
skeletal muscle fibres. J Muscle Res Cell Motil 10: 197–205, 1989.