Journal of Cellular Biochemistry 96:1231–1243 (2005)

Differential Activation of Stress-Responsive SignallingProteins Associated With Altered Loading in a RatSkeletal Muscle

Inho Choi,1* Kisoo Lee,1 Myungjoo Kim,1 Moonyong Lee,1 and Kyoungsook Park2**1Department of Life Science, College of Liberal Arts and Science, Yonsei University, Wonju,Republic of Korea2Molecular Therapy Research Center, Sungkyunkwan University, Seoul, Korea

Abstract Skeletal muscle undergoes a significant reduction in tension upon unloading. To explore intracellularsignalling mechanisms underlying this phenomenon, we investigated twitch tension, the ratio of actin/myosin filaments,and activities of key signallingmolecules in rat soleusmuscle during a3-weekhindlimb suspension and2-week reloading.Twitch tension andmyofilament ratio (actin/myosin) gradually decreased during unloading but progressively recovered toinitial levels during reloading. To study the involvement of stress-responsive signalling proteins during these changes, theactivities of protein kinase C alpha (PKCa) and three mitogen-activated protein kinases (MAPKs)—c-Jun NH2-terminalkinase (JNK), extracellular signal-regulated protein kinase (ERK), and p38MAPK—were examined using immunoblottingand immune complex kinase assays. PKCa phosphorylation correlated positively with the tension (Pearson’s r¼ 0.97,P< 0.001) and themyofilament ratio (r¼ 0.83, P< 0.01) over the entire unloading and reloading period. Treatment of thesoleus muscle with a PKC activator resulted in a similar paralleled increment in both PKCa phosphorylation and thea-sarcomeric actin expression. The three MAPKs differed in the pattern of activation in that JNK activity peaked only forthe first hours of reloading, whereas ERK and p38 MAPK activities remained elevated during reloading. These resultssuggest that PKCamay play a pivotal role in converting loading stress to intracellular changes in contractile proteins thatdetermine muscle tension. Differential activation of MAPKs may also help alleviate muscle damage, modulate energytransport and/or regulate the expression of contractile proteins upon altered loading. J. Cell. Biochem. 96: 1231–1243,2005. � 2005 Wiley-Liss, Inc.

Key words: alpha actin; MAPK; myofilament; PKC; reloading; twitch tension; unloading

A postural or antigravity muscle undergoessignificant atrophy during spaceflight or hin-dlimb suspension. The atrophy is caused byconsiderable loss of cellular components (pro-

teinsandorganelles) andextracellularfluid fromthe muscle [Fitts et al., 2001]. The most seriousconsequence of these changes is decreasedmuscle tension [Caiozzo et al., 1996]. Thismuscular problem has been a central issue inspace biology because astronauts can face ser-ious locomotor problems during an emergency inspace or after return to Earth [Baldwin, 1996].Various cellular and molecular factors areknown to altermuscle contraction upon changesin loading [Vandenburgh et al., 1999; Baeweret al., 2004], yet the intracellular signallingmechanisms underlying this phenomenonremain unclear.

Weight loading imposes continuous contrac-tile stress and physical stretch to posturalmuscles [Baewer et al., 2004]. This loading iscrucial for maintaining muscle proteins andmyofibrillar integrity, while its abnormal dis-placement causes various pathological pro-blems including musculo-skeletal diseases

� 2005 Wiley-Liss, Inc.

Grant sponsor: Basic Research Program of the KoreaScience and Engineering Foundation (to IC); Grantnumber: R05-2003-000-10589-0.

*Correspondence to: Dr. Inho Choi, Department of LifeScience, College of Liberal Arts and Science, YonseiUniversity, 234 Maeji-Ri, Heungup-Myon, Wonju, Gang-won-do 222-710, Republic of Korea.E-mail: ichoi@dragon.yonsei.ac.kr

**Correspondence to: Kyoungsook Park, Molecular Ther-apy Research Center, Sungkyunkwan University, Sam-sung Medical Center Annex 8F, 50 Ilwon-Dong, Gangnam-Gu, Seoul 135-710, Republic of Korea.E-mail: hannah05@dreamwiz.com

Received 19 July 2005; Accepted 25 July 2005

DOI 10.1002/jcb.20616

[Ingber, 2003; Hornberger and Esser, 2004]. Ifthe muscle is unloaded, then it progressivelyloses myofibrillar proteins, with a concomitantdecrease in protein synthesis and an increase inprotein degradation [Vandenburgh et al., 1999;Isfort et al., 2002]. Consequently, the musclewastes both actin and myosin, causing signifi-cant misalignment of myofilaments as well as areduction in the number of cross-bridges [Rileyet al., 2000; Adams et al., 2003]. If the muscle isreloaded, however, the tissue regainsmyofibril-lar proteins and tension production, although itis uncertainwhether the recoverymechanism isthe reverse of the atrophic process. Thus, tobetter understand the muscle tension problem,we need to investigate how the expression ofthese proteins is regulated by intracellularmechanisms upon unloading and reloading(UR), which may virtually explain the extent ofthealterationofmyofibrillar integrity.Anumberof studies have investigated the signallingmechanisms in overloaded skeletal muscle(e.g., exercise) [Donsmark et al., 2003; Longet al., 2004]. However, few studies have address-ed the signalling mechanisms in unloadedmuscle.

According to previous reports, loading stressor stretch can be directly transduced into thecytoplasm through G protein-coupled receptors(GPCRs), Ca2þ channels and integrins in thecell membrane [Li and Xu, 2000; Ingber, 2003].Once transduced, these stimuli selectivelyactivate stress-activated signalling moleculesand subsequent nuclear transcription of therelevant protein-coding genes. The primarycandidate signalling pathways are those invol-ving protein kinase C (PKC) and mitogen-activated protein kinases (MAPKs). PKCs areubiquitous serine/threonine protein kinasesthat are involved in cell proliferation, regenera-tion and stress responses [Liu, 1996; Korzicket al., 2004]. Todate, 12PKC isoformshave beenidentified and classified into 3 subfamilies(conventional, novel and atypical), dependingon their requirements for Ca2þ and diacylgly-cerol (DAG) derived from phosphatidylinositol-4,5-bisphosphate (PIP2) [Gutkind, 1998].MAPKs are a family of serine/threonine kinasesthat include c-Jun NH2-terminal kinase (JNK),p38 MAPK and extracellular signal-regulatedprotein kinase (ERK). As a part of downstreamsubstrates of PKCs, MAPKs regulate musclecell death, survival and regeneration, particu-larly in response to various stresses like

oxidative stress, inflammation, stretch andexercise overloading [Boppart et al., 2000;Widegren et al., 2001; Donovan et al., 2002].

Previous studies demonstrate that the activa-tion of PKC isoforms and MAPKs vary with thenature of the stress. For instance, denervationof skeletal muscle increases or decreases phos-phorylation of a conventional PKCa or novelPKCy, respectively [Sneddon et al., 2000],whereas exercise elevates the activation ofatypical rather than conventional and novelPKCs [Rose et al., 2004]. Activation of theMAPKsappears to bemore complicated becausethe levels of activation vary in a stimulusintensity- and duration-dependent manner[Long et al., 2004]. The JNK pathway isactivated rather rapidly and transiently ininjury-evoking exercise, whereas the activationof the ERK and p38 MAPK pathways may beassociated with energy transport and/or cytos-keletal actin reorganisation during musclecontraction [Boppart et al., 2000; Donsmarket al., 2003]. The link between these kinases andPKCs has remained obscure, although somemetabolism and myogenesis studies havedemonstrated such a link [Long et al., 2004].

Based on the work described above, weexplored the muscle tension problem by inves-tigating the link between the external loadingstress and signalling pathways that may reg-ulate contractile protein expression. We at-tempted twitch tension measurements becausethis contraction occurs only within 1 s, henceallowing us to focus on the muscle contractilemachinery and removing the potential effects ofother factors (e.g., energy generation capacity)on muscle tension [Cain et al., 1962]. In addi-tion, because muscle contraction increases thelevels of both DAG and intracellular Ca2þ, andelevated DAG or treatment with phorbol esterenhances tension production [Ikebe and Brozo-vich, 1996; Donsmark et al., 2003], we consid-ered that the phosphorylation of conventionalPKC isoforms (cPKCs; a, b, g) as a part of the keysignalling processes underlying myofibrillarintegrity.

On the basis of this rationale and the afore-mentioned studies, we hypothesised that phos-phorylation levels of cPKCs positively correlatewith twitch tension and myofibrillar integrity,with a decreasing and increasing trend duringunloading and reloading, respectively. We alsohypothesised that the activities of MAPKsdecrease with unloading but rapidly increase

1232 Choi et al.

upon reloading (as in exercise), with the ERKand p38MAPK pathways, rather than the JNKpathway, better linked to cPKC activation. Wetested these hypotheses by examining changesin twitch tension, the ratio of actin/myosinfilaments, and phosphorylation levels of cPKCsrelative to expression of the a-sarcomeric actinlevel, and activities of MAPKs in the rat soleusmuscle during 3weeks of unloading and2weeksof reloading.

MATERIALS AND METHODS

Subjects

All studies involving animals were approvedby the Yonsei University Animal Care and UseCommittee. Sprague-Dawley rats (5 weeks old,100–120 g) were maintained at 22–248C in a14:10 h light:dark cycle. Standard Purina ratchow and water were provided ad libitum. Onthe last day of week 5, the rats were randomlydivided into seven groups: two unloadinggroups of 1 week (U1w) and 3 week (U3w); fourreloading groups, which experienced the 3weekunloading followed by 1 h, 5 h, 1 day or 2 weekreloading (R1h, R5h, R1d and R2w, respec-tively); and one representative age-matchedcontrol group kept for 3 weeks under standardconditions (C3w). We did not include C5w (anage-matched control for R2w) because musclecontractile properties and myofilament struc-ture do not differ between C3w and C5w [Leeet al., 2004]. We used a modified model ofhindlimb suspension adopted fromNyhan et al.[2002]. Each animal was caged separately andanaesthetised intraperitoneally with pentobar-bital sodium (30 mg/kg). Two pieces of flexibleTygon tubing (5.6 mm OD) were placed alongthe dorsumof the animal andwere suturedwithsilk threads to provide a firm support. On thefollowing day (the first day of week 6), after theanimal recovered completely from the anaes-thetics, the tubing at the tip of the tail wasconnected to a tether so that the animal wassuspended at an incline of 358 in a head-downorientation. The animal was free to movearound the cage on its front feet. The meshjacket was replaced in 7–10 days as the animalgrew.Similar tubingand jacketswereapplied tothe control groups with the exception that thetail was not taped to the tubing. At the time ofreloading, the animals were freed from alltubing and jackets and were housed in indivi-dual cages.

Twitch Contraction

On the day of experiment, each subject wasanaesthetised with pentobarbital sodium (i.p.,50 mg/kg), and the soleus muscles of bothhindlimbs were quickly removed. The left limbmuscle was frozen immediately in liquid nitro-gen for immunoblotting and immune complexkinase assays or in 2.5% glutaraldehyde forelectron microscopy (EM). The right limbmuscle was soaked in 70 ml of oxygenatedRinger’s solution (115 mM NaCl, 5 mM KCl,4 mM CaCl2, 1 mM MgCl2, 1 mM NaH2PO4,24 mMNaHCO3, 11 mM glucose, 0.02 g/L tubo-curarine chloride, pH 7.39) at 258C. One tendonwas tied to a lever arm of a 300B-LR servomotorsystem (Aurora Scientific, Aurora, Canada),and the other tendon was tied to a micrometerto adjust to the optimal muscle length (Lo). Acustom DMC program (Aurora Scientific) wasused to control both the servo system and aGrass S48 stimulator. The stimulator wasconnected to apair of bright platinumelectrodesand supplied a 1.0-ms square wave pulse thatgenerated maximum twitch force. At the com-pletion of the twitch experiment, the tissue wasfrozen in place at Lo with liquid nitrogen,sectioned (8 mm) in aMicromHM505E cryostat,and stained by a routine haematoxylin andeosin procedure. The cross-sectional area (CSA)of each muscle was determined as describedpreviously [Lee et al., 2004]. The force datawereanalysed by custom software (DMA, AuroraScientific) and normalised to the CSA.

Electron Microscopy

The soleusmusclesused in theEMstudywereobtained from three rats per group. Each tissuewas pre-fixed for 4h in 2.5%glutaraldehyde, cutinto 1� 1� 2 mm blocks, rinsed for 15 min in0.1 M phosphate buffer (pH 7.2) and post-fixedfor 2 h in 1% OsO4. The fixation and rinsingprocesses were conducted at 48C to minimiseautolysis and extraction. The fixed sampleswere dehydrated two times in a series of ethanolsolutions (50%, 60%, 70%, 90% and 100%). Thesamples then were infiltrated with propyleneoxide, affixed sequentially on dodecenyl succi-nic anhydride, epon 812, nadic methyl anhy-dride and tri(dimethylaminomethyl) phenol(DMP-30) for 1 h each at room temperatureand then polymerised at 608C overnight. Aftertrimming and semi-thin sectioning, musclesamples were dried on a hot plate (808C) and

Muscle Tension and Stress Signalling 1233

were prestained with 0.5% toluidine blue. Thesemi-thin sections were trimmed, ultra-thinsectioned with an ultramicrotome and double-poststained with 1% uranyl acetate for 15 minand lead citrate for 5 min. Ultra-thin sections(0.05 mm) were supported on a grid. Cross-sections were adjusted to 120,000� magnifica-tion under an electron microscope (TEM, JEM-1200EX, Jeol, Tokyo, Japan), photographedbyadigital TEM camera (Megaview III), and ana-lysed using an Image Analysis System (SIS,Munster, Germany).

Wechose threephotographed images out of 12images per tissue, in which both actin andmyosin filaments were clearly evident. Wedivided each image into �10 quadrates(63 nm� 63 nm each) using Adobe Photoshop5.0 software (Adobe Systems, Inc., San Jose,CA), and randomly selected four quadrates.Actin and myosin filaments were then countedwithin each quadrate. Any actin or myosinfilament appearing on the edge of the quadratewas not counted.

Immunoblotting and Immune ComplexIn Vitro Protein Kinase Assays

The frozen muscle tissues were homogenisedin five volumes of tissue lysis buffer (25 mMHEPES, pH 7.5, 0.3 M NaCl, 1.5 mM MgCl2,0.2mMEDTA, 0.05% (w/v) TritonX-100, 20mMb-glycerophosphate, 0.1 mM orthovanadate,0.5 mM DTT, 0.4 mM phenylmethylsulfonylfluoride (PMSF), 1 g/ml leupeptin, 1 g/mlpepstatin), as described previously [Lee et al.,2002]. After a 15-min incubation on ice, tissuelysates were centrifuged at 12,000 rpm using amicrocentrifuge for 15 min at 48C. The super-natant was saved as the whole-tissue solublelysate. The Bradford assay was used to deter-mine the protein concentration in lysatesaccording to the manufacturer’s protocol (Bio-Rad,Hercules,CA).Whole-tissue lysates (50mg)were subjected to SDS–PAGE on 10% gelsfollowed by electrophoretic transfer of proteinbands to a nitrocellulose membrane and sub-sequent incubation with three antibodiesagainst cPKCs (PKCa, PKCb, PKCg). Immunecomplexes were detected with the ECL system(Amersham Pharmacia Biotech, Uppsala, Swe-den) and quantitated using a densitometer (Geldocument,Bio-Rad). The loading and transfer ofprotein was normalised to b-tubulin. The activ-ities of MAPKs (JNK, ERK, p38) were deter-mined by immune complex kinase assays, as

described previously [Lee et al., 2002]. Proteinfromsoluble lysates (500mg)was incubatedwithspecific antibodies against JNK, ERK or p38MAPK for 2 h at 48C. Immunoprecipitation wasachieved using a 1-h incubation with protein G-Sepharose. Immune complexes were washedtwice with Triton X-100 lysis buffer and threetimes with 20 mM HEPES (pH 7.4), andresuspended in 20 ml of kinase assay buffercontaining 10 mM MgCl2, 2 mM DTT, 0.2 mMsodium orthovanadate and 2 mg of substrate.GST-c-Jun, GST-ATF2 or MBP was used as asubstrate for JNK, p38 or ERK, respectively.The MAPK activity assays were carried out inthe presence of 20 mM unlabeled ATP and 2 mCiof [g-32P]ATP for 30min at 308Cand terminatedby the addition of 5 ml of 5ml SDS sample bufferand boiling for 5 min. The reactions wereseparated by SDS–PAGE on a 15% gel. Thephosphorylation of the substrate proteins wasexamined by autoradiography and quantitatedusing a densitometer (Gel document, Bio-Rad).The same blot was used to examine an equalload of immunoprecipitated JNK, ERK and p38MAPK by incubation with the correspondingantibody. Antisera and substrates used forimmune complex assays were: PKC antibodysampler kit (anti-PKCa, anti-PKCb and anti-PKCg) from Transduction Labs (Lexington,KY); anti-JNK, anti-ERK (1/2), and anti-p38from Santa Cruz Biotechnology (Santa Cruz,CA);GST-c-Jun,GST-ATF2, fromNewEnglandBiolabs (Beverly, MA); myelin basic protein(MBP) from Sigma (St. Louis, MO); anti-phosphorylated PKCa at Ser657 from UpstateBiotechnology (Charlottesville, VA); and pro-teinG-Sepharose and [g-32P]ATP (specific activ-ity, 5,000Ci/mmol) fromAmershamPharmacia.The secondary antibodies were horseradishperoxidase-conjugated anti-rabbit IgG or anti-mouse IgG from Zymed (San Francisco, CA).

Activation of PKCa Ex Vivo andExpression of a-Actin

To explore correlation between PKCa activa-tion and expression ofa-sarcomeric actin, soleusmuscles were dissected from 6-week-old ratsand then incubated ex vivo with the PKCactivator, phorbol 12-myristate 13-acetate(PMA, 100 nM, Sigma), dissolved in 100%dimethyl sulfoxide (DMSO) for various periods(0, 0.5, 1, 2, 5 h) at pH 7.39, 258C. The muscletissues were then homogenised in five volumesof tissue lysis buffer (described above) to

1234 Choi et al.

prepare whole-tissue lysates. Phosphorylationof PKCa and expression of a-actin were deter-mined by immunoblottingwith anti-phosphory-lated PKCa antibody (Ser657) and anti-a-actinantibody (Sigma), respectively.

Data Presentation

All data are presented as the mean� 1SEM,or as otherwise noted. Significance of inter-group differences was examined by one-wayanalysis of variance (ANOVA) and Bonferroni’spost hoc test. Correlations among twitchtension, actin/myosin ratio, and activities ofsignalling proteins, and also between PKCaphosphorylation and a-actin levels were deter-mined by Pearson’s coefficients (r), and thesignificance was examined by the two-tailedt-test on r. The statistics were performed withSPSS/PCþ at the significance level P¼ 0.05.

RESULTS

Body Mass and Muscle Mass

Body mass and muscle properties for thecontrol and the UR groups are summarised inTable I. As expected, muscle mass relative tobody mass decreased by 39.8% in U3w butrecovered to the control level in R2w (P< 0.001;Fig. 1A).MuscleCSAalso decreased by 43.9% inU3w but increased to 122.1% of the control inR2w (P< 0.001; Table I).

Twitch Tension

Figure 1B and C presents the maximumtwitch force and twitch tension, respectively,of the soleus muscle for each loading condition.Twitch force of C3w was 10.2� 0.8 g; this forcedecreased by 61.5% in U3w and 62.4% in R5h,and returned to 128% of the control in R2w(P< 0.001). Twitch tension of C3w was 39.2�

2.8 mN/mm2; this tension decreased signifi-cantly by 34.1% in U3w and 37.3% in R1h, andreturned to 96.4% of the control in R2w(P< 0.001). Twitch rise time between onsetand the point of peak force ranged from 23 to46 ms for all controls and UR groups examined.

Electron Microscopic Analysis

Figure 2A–D illustrates representative elec-tron micrographs of cells in cross-sectionedsoleus muscles for groups C3w, U3w, R1h andR2w. In C3w, actin filaments were fairly wellorganised in a hexagonal array around eachmyosin filament, whereas in U3w and R1h theactin arrangement was somewhat irregular. InR2w, the number of actin filaments around eachmyosin filament was greater than that in C3w.

Since the magnification of the micrographswas the same, the numbers of actin and myosinfilaments in each quadrate were counted.Figure 3A and B presents the average numberof actin filaments per quadrate and the ratio ofactin/myosin filaments for each loading condi-tion. Each quadrate contained 24–33 myosinfilaments, depending on the loading status ofthe muscle. Actin and myosin filaments weredifferentially affected by the loading conditionsin that the decrease in the number of actinfilaments was greater than that of myosinfilaments, with the lowest value at R1h. As aresult, the actin/myosin filament ratio reducedsignificantly upon the loading status from3.65 in C3w, 2.00 in R1h and 4.01 in R2w(F6,143¼ 84.08, P< 0.001).

Phosphorylation of cPKC Isoforms

Since muscle contractility depends on Ca2þ

concentration, we examined the phosphoryla-tion status of three conventional PKC isoforms(a, b, g) in soleus whole-tissue lysates using

TABLE I. Summary of Morphological Analysis of the Rat Soleus Muscle During Unloadingand Reloading Conditions

Groups C3w U1w U3w R1h R5h R1d R2w

VariablesBody mass (g)a 269.88� 3.12 202.57� 6.89 230.20� 7.06 227.84� 6.31 223.72� 10.91 230.49� 11.92 358.91� 23.54Muscle mass (g)b 0.107�0.009 0.076�0.004 0.054� 0.005 0.057� 0.004 0.058� 0.005 0.061� 0.002 0.141� 0.008Cross-sectionalarea (CSA) (mm2)c

4.142�0.281 3.249�0.220 2.355� 0.189 2.340� 0.149 2.318� 0.228 2.621� 0.096 5.102� 0.262

Values are mean�1SEM.aThe value of C3w differed significantly from those of U1w and R2w; the value of R2w differed significantly from all others (one-wayanalysis of variance (ANOVA) and Bonferroni’s post hoc test; F6,53¼19.8, P< 0.001).bValues differed significantly between all pairs of groups (F6,53¼ 38.1, P< 0.001).cThe value of C3w differed significantly from those of U3w, R1h, R5h, R1d and R2w; the value of R2w differed significantly from allothers (F6,53¼23.3, P< 0.001).

Muscle Tension and Stress Signalling 1235

correspondingphospho-PKC-specific antibodies.Compared with the control, PKCa phosphoryla-tion levels decreased steadily to 67.3% at U3w,62.4% at R1h and then returned to 89.4% at R2w(n¼ 6, P< 0.01) (Fig. 4A and B). In contrast, our

preliminaryresultshadrevealed thatexpressionof both PKCb and PKCg as well as phosphoryla-tion levels of PKCb andPKCgwere at the limit ofdetection in our rat muscle tissues, making itdifficult to accurately calculate any variationamong the groups (data not shown). Since PKCawas theprominent isoformdetected inour study,wenext examined the correlation betweenPKCaand a-actin expression in soleus strips aftertreatment with 100 nM PMA, a PKC activator,over four different time periods (n¼ 6 for each).PMA treatment induced significant PKC phos-phorylation at 30 min post-treatment, whichthen decreased gradually over the next 4.5 h(P< 0.05). A similar pattern of induction wasobserved for a-actin expression, with peakinduction occurring at 1 h post-treatment(P< 0.05) (Fig. 4C and D). Taken together, thedata indicate a significant correlation betweenPKCa phosphorylation and a-actin expression(Pearson’s r¼ 0.51; P< 0.05) (Fig. 4E).

Changes in MAPK Activities

MAPKs are important for mediating intra-cellular signalling brought on by many types ofstress. To explore the involvement of MAPKs intransmitting external loading signals in thesoleusmuscle, we used immune complex kinaseassays (two samples per group) to examinechanges in the temporal profiles of MAPKsignalling activities in response to loadingchanges (Fig. 5A, C and E). The U1w sampleswere not included because preliminary studiesshowed that the activity of each MAPK in U1wwas similar to that in C3w or U3w. Relative tothe level of C3w, JNK activity increaseddramatically 11.3� 4.9 fold in R1h and rapidlyreturned to the control level after R5h (Fig. 5B).ERK activity, however, remained similar to thecontrol level until R1h and then increased afterR5h, reaching a level 7.4� 2.7 fold that of thecontrol in R2w (Fig. 5D). The activity of p38MAPK, on the other hand, increased �2.0 foldbetween R1h and R1d, and then decreased to1.5 fold of the basal level in R2w (Fig. 5F).

Relationships Among Twitch Tension, Ratioof Actin/Myosin Filaments and PKCa

Phosphorylation

Twitch tension, myofilament ratio and PKCaresponses showed similar U-shaped trendsacross the UR period, and thus we conductedregression and correlation analyses betweenpairs of datasets (Table II). The regression slope

Tw

itch

tens

ion

(mN

mm

-2)

0

10

20

30

40

50

U1w U3w R1h R5h R1d R2w

* *

C

C3w

Tw

itch

forc

e (g

)

0

2

4

6

8

10

12

14

16

* * *

*

B

*

*

Mus

cle

mas

s/bo

dy m

ass

(x10

-3)

0.0

0.1

0.2

0.3

0.4

0.5

0.6 A

* **

*

Fig. 1. The soleus muscle undergoes significant atrophy and areduction in tension during unloading, but recovers to the initiallevels as reloading proceeds. A: Soleus muscle mass relative tobodymass, (B)maximum twitch force and (C) normalised tensionare illustrated for each loading condition. Twitch force wasgenerated at 258Cbya1.0-mspulse at supramaximal voltage andwas normalised by the cross-sectional area (CSA) of the tissue.Each data point represents the mean�1SEM; n¼15 for C3w,and n¼ 9 for unloading or reloading groups. *Significantlydifferent from both C3w and R2w (one-way analysis of variance(ANOVA) and Bonferroni’s post hoc test, P< 0.05).

1236 Choi et al.

between twitch tension with actin/myosin ratiowas 0.97, and the correlation coefficient (Pear-son’s r) between these two variables was 0.88(P< 0.01). The correlation coefficients betweentwitch tension and the PKCa phosphorylationand between the actin/myosin ratio and PKCaphosphorylation were 0.97 (P< 0.001) and 0.81(P¼ 0.028), respectively.

DISCUSSION

In our experimental system, the twitch ten-sion of the soleus muscle peaked in less than0.1 s, under both normal and unloaded condi-tions. During this brief time, the muscle wouldutilise already-available phosphagens (ATP,creatine phosphate), which may be present insufficient quantities in the sarcoplasm for sim-ple cross-bridge turnover [Cain et al., 1962].

Therefore, the key factor directly limitingmuscle tension must be the contractile machin-ery (e.g., myofibrillar components). From thisbaseline, the current study aimed to elucidatethe stress-mediated signalling mechanismsthat link external stimuli to tension capacityvia myofibrillar alterations.

Our results show that there is a good correla-tion between twitch tension, the ratio of actin/myosin filaments and PKCa phosphorylationover the entireURperiod.Wealso observed thatthree kinds of MAPKs (JNK, p38, ERK) aredifferentially regulated after reloading: Activa-tion of JNKwas rapid but transient only duringearly reloading; activation of ERK remained atthe basal level for the first several hours ofreloading and rapidly increased thereafter; andp38MAPKwas activated for at least 1 day afterreloading. Thus, in general our results re-

Fig. 2. The number of actin and myosin filaments in cross-sectioned soleusmuscle cells changesmarkedlywith the loadingstatus, as shown in the representative electron micrographs forC3w (A), U3w (B), R1h (C) andR2w (D). Three soleus tissues from

three subjects were used in each group, with four quadratesselected per section (totalling 24–36 quadrates per group). Eachmicrograph shown here is a single quadrate (63 nm� 63 nm), asused for quantitation in Figure 3.

Muscle Tension and Stress Signalling 1237

garding the relationships between PKC andmyofibrillar integrity and muscle tension sup-port our first hypothesis. The hypothesis re-garding the temporal trends of MAPKs over theloading conditions as well as their molecularlinks with PKCa needs further investigation.

The soleus muscle lost a higher percentage ofmass thandid thewhole body during the 3-weekunloading period, indicating greater atrophy ofthe tissue compared with the rest of the body.The serious consequences of this atrophy werealterations in muscle tension and myofibrillarintegrity (indicated by the ratio of actin/myosinfilaments). The 34% decrease in the tensionproduction capacity in our study is essentiallythe same as that (33%) reported previously[Caiozzo et al., 1996]. This tension reductioncould be explained in part by a disproportional

loss of actin filaments relative to myosin, suchthat the actin/myosin ratio decreased by 31%during the 3-week period (Fig. 3; also see Rileyet al. [2000]). Importantly, the overall temporaltrend of themyofilament ratio closely paralleledthat of twitch tension during the UR period,with the regression slope near unity and acorrelation coefficient of 0.88 (Table II). Becauseboth the tension and actin/myosin ratio weremeasured on the basis of the muscle cross-section, the reduced tension must be the resultof decreased numbers of cross-bridges per CSA,which must be directly affected by the loadingstatus. The unexplained variation in the corre-lation (ca. 23%)might arise fromseveral factors,such as misalignments between actin andmyosin filaments or sarcomere lesions in themyofibres experiencing altered loading [Fittset al., 2001].

Our next approach to the issue of muscletension reduction was to understand how thismyofibrillar alteration might be regulated byintracellular mechanisms that are linked tomechanical stimuli. As shown in Figure 4A andB, PKCa phosphorylation was significantlyaltered by the loading status. The levels of theother two cPKCs (b and g) were too low to assessthe loading effect. Low levels of these isoformsalso have been reported in other studies onrodent skeletal muscle [Hilgenberg et al., 1996;Boczan et al., 2000; Goel and Dey, 2002]. Suchresponsiveness of PKCa to variable loadingconditions could be anticipated considering therole of the soleusmuscle on antigravity balance.Thus, under the normal loading, the PKCa iscontinually translocated from the sarcoplasm tothe cell membrane and activated through bind-ing with DAG and calcium (released from thesarcoplasmic reticulum) [Liu, 1996; Gutkind,1998]. However, the signalling pathway couldbe down regulated during unloading and reac-tivated upon reloading.

This pattern of PKCa phosphorylation underdifferent loading conditions may affect tensionproduction in two ways: first, by regulating theexpression of contractile proteins (the focus ofour study), and second, by changing the bio-chemical properties of existing myofibrillarcomponents (e.g., affecting Ca2þ sensitivity ofmyofilaments and myosin light chain phospha-tase activity in smooth muscle) [Ikebe andBrozovich, 1996; Tanaka et al., 2002]. Regard-ing the first proposition, our data show thatPKCa phosphorylation correlates well with both

Fig. 3. Actin andmyosin filaments are differentially affected bythe loading stress. The number of actin filaments decreasedmoresubstantially than that of myosin filaments during unloading.A: The number of actin filaments per quadrate, and (B) the ratio ofactin/myosin filaments are shown for each loading level.*Significantly different from both C3w and R2w; **significantlydifferent from C3w (one-way ANOVA and Bonferroni’s post hoctest, P<0.05).

1238 Choi et al.

the actin/myosin filament ratio (Table II) and a-actin levels (Fig. 4D and E). Upon treatmentwith thePKCactivator,PMA,a1.4 foldelevationof PKC phosphorylation was followed by asimilar increase in a-actin expression. About a30-min delay between peaks of two variableswould be the time lag from the post-transmem-brane to post-translational processes.A cellular mechanism underlying functional

association of PKCa and a-actin has yet to beclarified. Emerging evidence suggest that actinexpression would depend on the cytoplasmiclevel of monomeric actins (G-actin) and beha-

vior of PIP2 (the sarcolemmal secondary mes-senger) that responds to external stimulus[Lyubimova et al., 1997; Laux et al., 2000].Synthesis of actin molecules was increasedwhen the G-actin level was decreased duringpolymerisation of actin in various cells [Lyubi-mova et al., 1997; Laux et al., 2000]. Further-more, PKC substrates like myristoylatedalanine-rich C kinase substrate (MARCKS)were found to stimulate actin assembly bypromoting clustering of PIP2 in the sarcolemma[Laux et al., 2000; Bittner and Holz, 2005]. Itmay thus be speculated that the external stress

Fig. 4. The phosphorylation level of protein kinase C alpha(PKCa) decreases significantly during unloading and earlyreloading, but gradually recovers to the basal level as reloadingproceeds. A: Immunoblot analysis of PKCa phosphorylation atSer657. Whole-tissue lysates were subjected to SDS–PAGEfollowed by incubation with a phospho-PKC(Ser657)-specificantibody. Total PKCa level was also measured by immunoblot-ting with an antibody against PKCa. B: Quantitation of PKCaphosphorylation. Phosphorylation levels are relative to thatmeasured for C3w. Each data point represents the mean� 1SEM(n¼6). *Significantly different from C3w (one-way ANOVA andBonferroni post hoc test, P< 0.01). C–D: Association of PKCa

phosphorylation with a-actin expression. Soleus muscles weretreated with 100 nM phorbol 12-myristate 13-acetate (PMA) forthe indicated period, and then the phosphorylation of PKCa andexpression of a-sarcomeric actin were determined by immuno-blotting with the indicated antibody. The same immunoblot wasprobed with a PKCa antibody to show equal loading. Each datapoint represents themean� 1SEM(n¼ 6). *Significantly differentfrom 0 h (control) (P< 0.05). E: Regression and correlationanalyses between PKCa phosphorylation and a-actin levelobtained from the 100-nM PMA treatment experiment(Y¼ 0.77Xþ0.23; Pearson’s r¼0.51, P< 0.05).

Muscle Tension and Stress Signalling 1239

Fig. 5. The three mitogen-activated protein kinases (MAPKs)show differential activation upon altered loading stress. Whole-tissue soluble lysates were used to determine the activities of c-Jun NH2-terminal kinase (JNK) (A, B), extracellular signal-regulated protein kinase (ERK) (C, D) and p38 MAPK (E, F) usingimmune complex kinase assays. Purified recombinant GST-c-

Jun,MBP or GST-ATF-2were used as substrates for JNK, ERK andp38MAPK respectively. The levels of total JNK, ERK, p38 MAPKwere measured by immunoblotting with respective antibodies.The levels changed little during the experiment. Two animalswere used per group, and the data are representative of the twosample sets tested.

TABLE II. Regression and Correlation Analyses Among TwitchTension, Myofilament Ratio and PKCa Phosphorylation

Y variables

X variables

Actin/myosin ratio PKCa phosphorylation

Twitch tensionRegression equation Y¼0.97 Xþ 0.053 Y¼ 0.79 Xþ 0.267Correlation coefficienta r¼0.877 r¼ 0.966Significance for ra P<0.01 P< 0.001

Actin/myosin ratioRegression equation — Y¼ 0.85 X� 0.024Correlation coefficienta r¼ 0.807Significance for ra P¼ 0.028

aSignificance of correlation was determined by the two-tailed t-test on Pearson’s coefficient (r).

1240 Choi et al.

or PMA treatment causes PKC activation andlocal PIP2 clustering by the PKC substrates,which upholds actin polymerisation, reductionin the G-actin level and elevation in actinexpression. In the unloading condition, theseprocesses would be reversed, with down regula-tion in the PKC signalling pathway and actinexpression. An integrative study includingmechanotransduction, associated signallingand capping should be approached to compre-hend the detailed mechanism of the PKCactivation to actin expression connection. Incardiac and smooth cells, mechanical stimuluswas reported to induce expression of theskeletal a-actin gene, or actin polymerisationthat increased the ratio of fibrous actin:G-actin,possibly via PKC activation [Komuro et al.,1991; Yazaki et al., 1993; Poussard et al., 2001;Cipolla et al., 2002]. We did not study the role ofPKCa phosphorylation on myosin proteinexpression, although this issue also needsclarification. From the literature and otherinvestigations, the phosphorylated form ofPKC is considered to be the active form thatbinds to the cell membranes (see, Gutkind[1998] for review). In our present study, wemeasured PKC phosphorylation in soleuswhole-tissue extract because we were unableto obtain sufficient amounts of sarcolemmalextract from the tissue at these ages of rats. Itwill be informative to investigate PKCa phos-phorylation using subcellular fractionationfollowed by immunoblotting or immunofluores-cence microscopy using a phospho-PKC-specificantibody. Taken together, these results implythat PKCa may play a critical role in linkingexternal loading stress to the regulated expres-sion of a-sarcomeric actin (and thus to muscletension). To clarify the role of PKCa activationon muscle tension capacity, we will furtheraddress the effect of PKC activation on muscletension using PMA.It is worth noting a study of Sneddon et al.

[2000] that addressed the preventive effect ofclenbuterol (a b-adrenergic agonist) on dener-vation-induced muscle atrophy, a system thatrepresents a good comparison to our currentstudy.These investigators showed that the levelof PKCa phosphorylation increases upon dener-vation of the rat soleus muscle (where the‘passive’ stretch is intact), and a further increasein phosphorylation upon clenbuterol treatmentsignificantly attenuates atrophy of the dener-vated tissue. In our present study, the level of

PKCa phosphorylation decreased in theunloaded muscle (where the neural input isintact; Fig. 4AandB).Thus, thephysical stretch,rather than motor influence per se, seemsimportant to activate this signalling pathway.The increased PKCa activation in the dener-vated muscle might indicate that this tissue, inthe absence of a feedback control, experienced agreater stretch than did the normal muscle.Since the further increase inPKCaphosphoryla-tion (by clenbuterol treatment) amelioratedmuscle atrophy, PKCa activation may be impor-tant to maintain myofibrillar constituents orincrease the efficiency of mRNA translation[Sneddon et al., 2000].

In addition to PKC activation, MAPK cas-cades also convert mechanical stimuli intorelevant intracellular responses [Long et al.,2004]. We observed differential activation ofthree kinds of MAPKs, particularly during thereloading period (Fig. 5). From the viewpoint ofmechanical stress, reloading and exerciserepresent comparable loading stimuli becauseboth cases render the tissue ‘overloaded’ follow-ing either unloading or rest. Previous reportsshow that JNK activity in human or rat skeletalmuscle increases 6–15 fold above the basal levelafter exercise or sciatic-stimulated muscle con-traction [Aronson et al., 1997; Boppart et al.,1999, 2000]. These studies also demonstratethat activation of JNKreaches amaximumafter15 min of contraction and remains elevated forabout 1 h. This pattern of JNK activation iscomparable to that of the soleus muscle duringearly reloading (Fig. 5A). Thus, activation ofJNK signalling is transient for the first severalhours of muscle activity, and may reflect aresistance response of the soleusmuscle againstcellular damage or lesion.

Several studies also show that mechanicalstimuli activate the other twoMAPKs (ERKandp38) in skeletalmuscles. For instance, ERKwasactivated at a maximum level (15–31 fold overbasal level) in 10–30 min during sciatic stimu-lation or cycling exercise, and remained ele-vated throughout the 1-h exercise period[Widegren et al., 2001]. Similarly, p38 MAPKactivity increased 2.2 fold above the control andremained elevated throughout the 1-h exercise.In addition, Boppart et al. [2000] demonstratedthat among p38 isoforms (a, b and g) in skeletalmuscle, p38g had the highest response (4 foldincrease in activity) immediately after exercisebut returned to the basal level at 1 day post-

Muscle Tension and Stress Signalling 1241

exercise. We also observed relatively moderateactivation (2 fold) of p38 for at least 1 day afterreloading.

Molecular links between PKC and MAPKpathways in response to unloading have beenpoorly understood in skeletalmuscle, but recentdata from various systems are worth noting. Inthe exercising muscle, a PKC–ERK signallingpathway was found to be involved in thestimulation of hormone-sensitive lipase fortriacylglycerol catabolism [Donsmark et al.,2003; Langfort et al., 2003; Long et al., 2004].Furthermore, accelerated glucose transportupon exercise is likely to be regulated by Ca2þ-sensitive cPKC-mediated signalling, in whichp38 MAPK may be involved [Widegren et al.,2001]. In differentiating skeletal muscle, downregulation of PKCs (mostly PKCa) increasedJNK activation but decreased the activation ofthe ERK and p38 pathways [Goel and Dey,2002]. Also, both the PKC–ERK and PKC–p38MAPK pathways are important in mediatinghypertrophy and the migration of smoothmuscle cells upon mechanical stress, althoughtheir responses depend on the duration andmagnitude of the stress [Li and Xu, 2000]. In thecase of the cardiac hypertrophic response,elevated contractile output of myocytesincreased the activation of the PKC–ERK path-way but not the activation of JNK or p38[Sugden, 2001]. These results and our findingssuggest that the ERK pathway is more likelyrelated to the activation of PKCa for energymobilisation or cell differentiation, whereas theJNKpathwaymay benegatively associatedwithPKCa activation, possibly during the injury-evoking process. The activation pattern of p38MAPK, and its association with PKCa, seems tobe intermediary between JNK and ERK.

In conclusion, our data demonstrate thatPKCa may play a principle role in linkingexternal stimuli with the expression of contrac-tile proteins, which then determines the level oftwitch exertion. Differential activation ofMAPKs may also be important for protectingmyofibres against damage during early reload-ing, mobilising energy sources or regulating theexpression of contractile proteins. Because theacute effects of unloading on activation of thesemolecules cannot be ruled out, this issue needsto be further addressed, particularly withregard to elucidating the potential reversibilityof signalling mechanisms during unloadingversus reloading. Moreover, as recent studies

on exercise have revealed that atypical PKCsmediate MAPK activation [Chen et al., 2002;Rose et al., 2004], future research shouldaddress the involvement of diverse PKC iso-forms in signal transduction upon unloading.The findings of our study may provide valuableinformation to aid the search for measures tocounter muscle contractile depression in spacemicrogravity or extended periods of bed rest.

ACKNOWLEDGMENTS

We thank Dr. Marni Boppart at University ofIllinois—Urbana Champaign and professor UnJin Zimmerman at University of Pennsylva-nia for critically reviewing this manuscript,Mr. Suk Kim for processing EM micrographsand Dr. Timothy Taylor for editing themanuscript.

REFERENCES

AdamsGR, Caiozzo VJ, Baldwin KM. 2003. Skeletal muscleunweighting: Spaceflight and ground-based models.J Appl Physiol 95:2185–2201.

Aronson D, Dufresne SD, Goodyear LJ. 1997. Contractileactivity stimulates the c-Jun NH2-terminal kinase path-way in rat skeletal muscle. J Biol Chem 272:25636–25640.

Baewer DV, Hoffman M, Romatowski JG, Bain JL, FittsRH, Riley DA. 2004. Passive stretch inhibits centralcorelike lesion formation in the soleus muscles ofhindlimb-suspended unloaded rats. J Appl Physiol97:930–934.

Baldwin KM. 1996. Effect of spaceflight on the functional,biochemical, and metabolic properties of skeletal muscle.Med Sci Sports Exerc 28:983–987.

Bittner MA, Holz RW. 2005. Phosphatidylinositol-4,5-bisphosphate: Actin dynamics and the regulation ofATP-dependent and -independent secretion. Mol Phar-macol 67:1089–1098.

Boczan J, Boros S, Mechler F, Kovacs L, Biro T. 2000.Differential expressions of protein kinase C isozymesduring proliferation and differentiation of human skele-tal muscle cells in vitro. Acta Neuropathol (Berl) 99:96–104.

Boppart MD, Aronson D, Gibson L, Roubenoff R, Abad LW,Bean J, Goodyear LJ, Fielding RA. 1999. Eccentricexercise markedly increases c-Jun NH(2)-terminalkinase activity in human skeletal muscle. J Appl Physiol87:1668–1673.

BoppartMD, Asp S,Wojtaszewski JF, Fielding RA,Mohr T,Goodyear LJ. 2000. Marathon running transientlyincreases c-Jun NH2-terminal kinase and p38 activitiesin human skeletal muscle. J Physiol 526(Pt 3):663–669.

Cain DF, Infante AA, Davies RE. 1962. Chemistry ofmuscle contraction: Adenosine triphosphate and phos-phoylcreatine as energy suppliers for single contractionsof working muscle. Nature 196:214–217.

Caiozzo VJ, Haddad F, Baker MJ, Herrick RE, Prietto N,Baldwin KM. 1996. Microgravity-induced transforma-tions of myosin isoforms and contractile properties ofskeletal muscle. J Appl Physiol 81:123–132.

1242 Choi et al.

Chen HC, Bandyopadhyay G, Sajan MP, Kanoh Y, Stan-daertM, Farese RV, Jr., Farese RV. 2002. Activation of theERK pathway and atypical protein kinase C isoforms inexercise- and aminoimidazole-4-carboxamide-1-beta-D-riboside (AICAR)-stimulated glucose transport. J BiolChem 277:23554–23562.

CipollaMJ, GokinaNI, Osol G. 2002. Pressure-induced actinpolymerization in vascular smoothmuscle as amechanismunderlying myogenic behavior. FASEB J 16:72–76.

Donovan N, Becker EB, Konishi Y, Bonni A. 2002. JNKphosphorylation and activation of BAD couples thestress-activated signaling pathway to the cell deathmachinery. J Biol Chem 277:40944–40949.

Donsmark M, Langfort J, Holm C, Ploug T, Galbo H. 2003.Contractions activate hormone-sensitive lipase in ratmuscle by protein kinase C and mitogen-activatedprotein kinase. J Physiol 550:845–854.

Fitts RH, Riley DR, Widrick JJ. 2001. Functional andstructural adaptations of skeletal muscle to micrograv-ity. J Exp Biol 204:3201–3208.

Goel HL, Dey CS. 2002. PKC-regulated myogenesis isassociated with increased tyrosine phosphorylation ofFAK, Cas, and paxillin, formation of Cas–CRK complex,and JNK activation. Differentiation 70:257–271.

Gutkind JS. 1998. The pathways connecting G protein-coupled receptors to the nucleus through divergent mito-gen-activated protein kinase cascades. J Biol Chem 273:1839–1842.

Hilgenberg L, Yearwood S, Milstein S, Miles K. 1996.Neural influence on protein kinase C isoform expressionin skeletal muscle. J Neurosci 16:4994–5003.

Hornberger TA, Esser KA. 2004. Mechanotransduction andthe regulation of protein synthesis in skeletal muscle.Proc Nutr Soc 63:331–335.

Ikebe M, Brozovich FV. 1996. Protein kinase C increasesforce and slows relaxation in smoothmuscle: Evidence forregulation of the myosin light chain phosphatase.Biochem Biophys Res Commun 225:370–376.

Ingber DE. 2003. Mechanobiology and diseases of mechan-otransduction. Ann Med 35:564–577.

Isfort RJ, Wang F, Greis KD, Sun Y, Keough TW, FarrarRP, Bodine SC, AndersonNL. 2002. Proteomic analysis ofrat soleus muscle undergoing hindlimb suspension-induced atrophy and reweighting hypertrophy. Proteo-mics 2:543–550.

Komuro I, Katoh Y, Kaida T, Shibazaki Y, Kurabayashi M,Hoh E, Takaku F, Yazaki Y. 1991. Mechanical loadingstimulates cell hypertrophy and specific gene expressionin cultured rat cardiac myocytes. Possible role of proteinkinase C activation. J Biol Chem 266:1265–1268.

KorzickDH, LaughlinMH, BowlesDK. 2004. Alterations inPKC signaling underlie enhanced myogenic tone inexercise-trained porcine coronary resistance arteries.J Appl Physiol 96:1425–1432.

Langfort J, Donsmark M, Ploug T, Holm C, Galbo H. 2003.Hormone-sensitive lipase in skeletal muscle: Regulatorymechanisms. Acta Physiol Scand 178:397–403.

Laux T, Fukami K, Thelen M, Golub T, Frey D, Caroni P.2000. GAP43, MARCKS, and CAP23 modulate

PI(4,5)P(2) at plasmalemmal rafts, and regulate cellcortex actin dynamics through a common mechanism.J Cell Biol 149:1455–1472.

Lee M, Choi I, Park K. 2002. Activation of stress signalingmolecules in bat brain during arousal from hibernation.J Neurochem 82:867–873.

Lee K, Lee YS, Lee M, Yamashita M, Choi I. 2004.Mechanics and fatigability of the rat soleus muscleduring early reloading. Yonsei Med J 45:690–702.

Li C, Xu Q. 2000. Mechanical stress-initiated signaltransductions in vascular smooth muscle cells. CellSignal 12:435–445.

Liu JP. 1996. Protein kinase C and its substrates. Mol CellEndocrinol 116:1–29.

Long YC, Widegren U, Zierath JR. 2004. Exercise-inducedmitogen-activated protein kinase signalling in skeletalmuscle. Proc Nutr Soc 63:227–232.

Lyubimova A, Bershadsky AD, Ben-Ze’ev A. 1997. Auto-regulation of actin synthesis responds to monomericactin levels. J Cell Biochem 65:469–478.

Nyhan D, Kim S, Dunbar S, Li D, Shoukas A, BerkowitzDE. 2002. Impaired pulmonary artery contractileresponses in a rat model of microgravity: Role of nitricoxide. J Appl Physiol 92:33–40.

Poussard S, Dulong S, Aragon B, Jacques Brustis J,Veschambre P, Ducastaing A, Cottin P. 2001. Evidencefor a MARCKS–PKCalpha complex in skeletal muscle.Int J Biochem Cell Biol 33:711–721.

Riley DA, Bain JL, Thompson JL, Fitts RH, Widrick JJ,Trappe SW, Trappe TA, Costill DL. 2000. Decreased thinfilament density and length in human atrophic soleusmuscle fibers after spaceflight. J Appl Physiol 88:567–572.

Rose AJ, Michell BJ, Kemp BE, Hargreaves M. 2004.Effect of exercise on protein kinase C activity andlocalization in human skeletal muscle. J Physiol 561:861–870.

Sneddon AA, Delday MI, Maltin CA. 2000. Amelioration ofdenervation-induced atrophy by clenbuterol is associatedwith increased PKC-alpha activity. Am J Physiol Endo-crinol Metab 279:E188–E195.

Sugden PH. 2001. Signalling pathways in cardiac myocytehypertrophy. Ann Med 33:611–622.

Tanaka S, Kanaya N, Homma Y, Damron DS, Murray PA.2002. Propofol increases pulmonary artery smoothmuscle myofilament calcium sensitivity: Role of proteinkinase C. Anesthesiology 97:1557–1566.

Vandenburgh H, Chromiak J, Shansky J, Del Tatto M,Lemaire J. 1999. Space travel directly induces skeletalmuscle atrophy. FASEB J 13:1031–1038.

Widegren U, Ryder JW, Zierath JR. 2001. Mitogen-activated protein kinase signal transduction in skeletalmuscle: Effects of exercise and muscle contraction. ActaPhysiol Scand 172:227–238.

Yazaki Y, Komuro I, Yamazaki T, Tobe K, Maemura K,Kadowaki T, Nagai R. 1993. Role of protein kinasesystem in the signal transduction of stretch-mediatedprotooncogene expression and hypertrophy of cardiacmyocytes. Mol Cell Biochem 119:11–16.

Muscle Tension and Stress Signalling 1243