doi:10.1152/physiolgenomics.00232.2007 35:86-95, 2008. First published 1 July 2008;Physiol. Genomics

Marianna Evans, Kevin Morine, Cyelee Kulkarni and Elisabeth R. Bartonmasticationsupports fiber size economy in the murine muscles of Expression profiling reveals heightened apoptosis and

You might find this additional info useful...

51 articles, 21 of which can be accessed free at:This article cites http://physiolgenomics.physiology.org/content/35/1/86.full.html#ref-list-1

2 other HighWire hosted articlesThis article has been cited by

  [PDF] [Full Text] [Abstract]

, July 15, 2009; 587 (14): 3703-3717.J PhysiolStephan Klossner, Roberto Bottinelli and Martin FlückAnne-Cécile Durieux, Giuseppe D'Antona, Dominique Desplanches, Damien Freyssenet,muscle phenotypeFocal adhesion kinase is a load-dependent governor of the slow contractile and oxidative 

[PDF] [Full Text] [Abstract], May , 2010; 108 (5): 1069-1076.J Appl Physiol

Elisabeth R. Barton, J DeMeo and Hanqin Leiexpression and muscle hypertrophy after local IGF-I productionThe insulin-like growth factor (IGF)-I E-peptides are required for isoform-specific gene

including high resolution figures, can be found at:Updated information and services http://physiolgenomics.physiology.org/content/35/1/86.full.html

can be found at:Physiological Genomicsabout Additional material and information http://www.the-aps.org/publications/pg

This infomation is current as of July 26, 2011. 

the American Physiological Society. ISSN: 1094-8341, ESSN: 1531-2267. Visit our website at http://www.the-aps.org/.July, and October by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2008 bytechniques linking genes and pathways to physiology, from prokaryotes to eukaryotes. It is published quarterly in January, April,

publishes results of a wide variety of studies from human and from informative model systems withPhysiological Genomics

on July 26, 2011physiolgenom

ics.physiology.orgD

ownloaded from

Expression profiling reveals heightened apoptosis and supports fiber sizeeconomy in the murine muscles of mastication

Marianna Evans,1 Kevin Morine,2 Cyelee Kulkarni,1 and Elisabeth R. Barton1,3

1Department of Anatomy and Cell Biology, School of Dental Medicine, 2Department of Physiology, School of Medicine,and 3Pennsylvania Muscle Institute, University of Pennsylvania, Philadelphia, Pennsylvania

Submitted 5 October 2007; accepted in final form 27 June 2008

Evans M, Morine K, Kulkarni C, Barton ER. Expression profilingreveals heightened apoptosis and supports fiber size economy in the murinemuscles of mastication. Physiol Genomics 35: 86–95, 2008. First publishedJuly 1, 2008; doi:10.1152/physiolgenomics.00232.2007.—Distinctions be-tween craniofacial and axial muscles exist from the onset of devel-opment and throughout adulthood. The masticatory muscles are aspecialized group of craniofacial muscles that retain embryonic fiberproperties in the adult, suggesting that the developmental origin ofthese muscles may govern a pattern of expression that differs fromlimb muscles. To determine the extent of these differences, expressionprofiling of total RNA isolated from the masseter and tibialis anterior(TA) muscles of adult female mice was performed, which identifiedtranscriptional changes in unanticipated functional classes of genes inaddition to those attributable to fiber type. In particular, the massetersdisplayed a reduction of transcripts associated with contractile andcytoskeletal load-sensing and anabolic processes, and heightenedexpression of genes associated with stress. Associated with theseobservations was a significantly smaller fiber cross-sectional area inmasseters, significantly elevated load-sensing signaling (phosphory-lated focal adhesion kinase), and increased apoptotic index in masse-ters compared with TA muscles. Based on these results, we hypoth-esize that masticatory muscles may have a fundamentally differentstrategy for muscle design, compared with axial muscles. Specificallythere are small diameter fibers that have an attenuated ability tohypertrophy, but an increased propensity to undergo apoptosis. Theseresults may provide insight into the molecular basis for specificmuscle-related pathologies associated with masticatory muscles.

caspase-12; focal adhesion complex; endoplasmic reticulum-stressapoptosis; myosin heavy chain

DISTINCTIONS BETWEEN THE CRANIOFACIAL muscles and the axialmuscles of the limb and trunk are evident at the inception ofdevelopment. During embryogenesis, myogenic precursor cellsarise from epithelial somites, which have been segmented fromthe paraxial mesoderm flanking the neural tube (30, 36).Myogenic precursor cells from these somites produce all of themuscles in the limb and trunk and a subset of cranial muscles(the tongue). The remaining craniofacial muscles, however, arederived from more anterior nonsomitic cells of the prechordalhead mesoderm (25). The factors required for early phases ofdevelopment differ in these muscle groups. For example, Pax3and Myf-5 are needed for the existence of limb, trunk, andtongue muscles, but the absence of these factors does not affectthe development of craniofacial muscles (43). In contrast, thebasic helix-loop-helix transcription factor, MyoR, is importantfor the progression of craniofacial muscle development, where

its absence in combination with its homologous protein cap-sulin by gene targeting renders mice lacking craniofacial mus-cles (24). Therefore, both cell lineage and developmentalregulation of craniofacial and limb muscles are distinct.

The developmental patterns may contribute to distinguishingfeatures in adult head muscles. In the postnatal period, fibertype differences persist, where adult craniofacial muscle fibersretain fetal myosin heavy chain (MHC) and in some species�-cardiac MHC, neither of which are present in limb muscles(22, 48, 51). Within the craniofacial muscles, functional cate-gorization exists, which includes the masticatory, laryngeal,and extraocular muscles, and these groups also bear uniqueproperties. For example, with age, most skeletal muscles in thelimbs and trunk exhibit a fast to slow shift in the fiber typedistribution. However, jaw muscles shift in the opposite direc-tion, where there is an increase in the fast fiber population inaging masseter muscles (29). Furthermore, it has been demon-strated that the repair process after acute damage is impaired inmasticatory muscles compared with limb muscles (37). In thesame study, a decreased rate of proliferation was observed inmyoblasts isolated from regenerating masseters, suggestingthat the satellite cells were, at least in part, responsible for theinability to repair. These results imply that unlike limb mus-cles, masticatory muscles are slow to remodel or regenerate, aproperty that could exacerbate the damaged state or hinder thehealing process from any oral/facial intervention.

The properties of the craniofacial muscle groups challenge thedogma that there is generalized unified tissue called skeletalmuscle in the entire body. To date, this has been illustrated mostclearly in the extraocular muscles, where expression profiling hashelped to define the components that support rapid, fatigue-resistant activity for eye movements (15), and careful monitoringof proliferation has revealed constant remodeling of the extraoc-ular muscles by active satellite cells (26). However, this analysismay not be relevant for the craniofacial muscles involved inmastication or speech. Because limb and masticatory muscleshave been shown to differ at the developmental, physiological,and pathological levels, we sought to define a molecular signaturefor the muscles of mastication, which could help to identify thefactors contributing to these distinctions.

METHODS

Animals. All studies were approved by the university’s animal carecommittee. Adult C57BL/6 female mice, aged 6 mo, were utilized forall experiments to eliminate potential effects of growth, aging, or sexon the measurements. For each experiment, mice were killed by CO2

Address for reprint requests and other correspondence: E. R. Barton, Dept.of Anatomy and Cell Biology, 441A Levy Bldg., 240 S. 40th St., Univ. ofPennsylvania, Philadelphia, PA 19104 (e-mail: erbarton@biochem.dental.upenn.edu).

The costs of publication of this article were defrayed in part by the paymentof page charges. The article must therefore be hereby marked “advertisement”in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Physiol Genomics 35: 86–95, 2008.First published July 1, 2008; doi:10.1152/physiolgenomics.00232.2007.

1094-8341/08 $8.00 Copyright © 2008 the American Physiological Society86

on July 26, 2011physiolgenom

ics.physiology.orgD

ownloaded from

asphyxiation, after which skeletal muscles were rapidly dissected andfrozen in liquid nitrogen (for RNA and protein analysis). For histo-logical analysis samples were rinsed in PBS, blotted, weighed, andcovered in mounting medium (OCT) prior to freezing in meltingisopentane. Muscle samples for RNA and protein were stored in liquidnitrogen, and samples for histology were stored in �80°C. Anadditional set of samples were fixed in 4% paraformaldehyde andembedded in paraffin for terminal deoxynucleotidyltransferase-medi-ated nick end labeling (TUNEL) analysis.

Expression analysis. To identify differences between craniofacialand axial muscles at the molecular level, expression profiling wasperformed on total RNA isolated from the masseter and tibialisanterior (TA) muscles of 3 mice. Total RNA was isolated from thefrozen muscle samples (TRIzol; Invitrogen, Carlsbad, CA) and furtherpurified using RNeasy columns (Qiagen, Valencia, CA). RNA integ-rity was confirmed by gel electrophoresis. Profiling measurementswere completed at the Penn Microarray Core Facility utilizing Af-fymetrix Mouse 430 2A chips. All protocols were conducted asdescribed in the Affymetrix GeneChip Expression Analysis TechnicalManual. Briefly, total RNA was converted to first-strand cDNA usingSuperscript II reverse transcriptase primed by a poly(T) oligomer thatincorporated the T7 promoter. Second-strand cDNA synthesis wasfollowed by in vitro transcription for linear amplification of eachtranscript and incorporation of biotinylated UTP. The cRNA productswere fragmented to 200 nucleotides or less, heated at 99°C for 5 min,and hybridized for 16 h at 45°C to six microarrays. The microarrayswere then washed at low (6� SSPE) and high (100 mM MES, 0.1 MNaCl) stringency and stained with streptavidin-phycoerythrin. Flu-orescence was amplified by adding biotinylated anti-streptavidinand an additional aliquot of streptavidin-phycoerythrin stain. A con-focal scanner was used to collect fluorescence signal after excitationat 570 nm.

Initial data analysis. The Affymetrix Microarray Suite 5.0 algo-rithm in GeneChip Operating Software was used to quantitate expres-sion levels for targeted genes; default values provided by Affymetrixwere applied to all analysis parameters. Border pixels were removed,and the average intensity of pixels within the 75th percentile wascomputed for each probe. The average of the lowest 2% of probeintensities occurring in each of 16 microarray sectors was set asbackground and subtracted from all features in that sector. Probe pairswere scored positive or negative for detection of the targeted sequenceby comparing signals from the perfect match and mismatch probefeatures. The number of probe pairs meeting the default discrimina-tion threshold (� � 0.015) was used to assign a call of absent, present,or marginal for each assayed gene, and a P value was calculated toreflect confidence in the detection call. A weighted mean of probefluorescence (corrected for nonspecific signal by subtracting themismatch probe value) was calculated using the one-step Tukey’sbiweight estimate. This Signal value, a relative measure of theexpression level, was computed for each assayed gene. Global scalingwas applied to allow comparison of gene Signals across multiplemicroarrays: after exclusion of the highest and lowest 2%, the averagetotal chip Signal was calculated and used to determine what scalingfactor was required to adjust the chip average to an arbitrary target of150. All Signal values from one microarray were then multiplied bythe appropriate scaling factor.

Transcripts that received an absent call in all TA and massetersamples were eliminated from subsequent analysis. Pairwise compar-isons of the probe set signals were performed between the TA andmasseter for each animal using significance analysis of microarrays(SAM) (46) to detect �1.5-fold changes between samples with afalse discovery rate �7%. This included transcripts present ormarginal in TA, masseter, or all samples. Gene ontology of thedifferentially expressed transcripts was performed using Affymetrixanalysis tools (www.affymetrix.com/analysis/netaffx). In addition,literature searches were performed to determine functions of tran-scripts without gene ontology assignments or to find specific functions

within muscle. Expression profiling results were submitted to theNCBI-GEO (GSE11114) and the EMBL-EBI ArrayExpress (E-MEXP-1288) repositories.

Quantitative RT-PCR. Total RNA was isolated as described abovefrom the TA and masseter muscles of an additional seven adult femalemice. Quantification of RNA was performed with Ribogreen quanti-tation kit (Molecular Probes, Eugene, OR), and RNA integrity wasconfirmed by gel electrophoresis. Equal amounts of total RNA fromeach sample was subjected to single strand reverse transcription(Applied Biosystems, Foster City, CA). The resultant cDNA wasutilized for quantitative RT-PCR (qRT-PCR) with oligonucleotidesspecific for genes listed in Table 2 using the Roche Lightcyclersystem, and reagents (LightCycler FastStart DNA MasterPLUS SYBRGreen I; Roche Applied Science, Indianapolis, IN) as previouslydescribed (4). Expression levels were assessed by calculating thecrossing point (Cp), where measured fluorescence rises above back-ground, as the second derivative maximum of the reaction curve. Eachsample was analyzed in duplicate and the resulting data averaged. Themean Cp values for the TA and masseters were calculated from therespective muscle samples. The relative change in expression betweenmuscle groups was based on the assumption that a difference of 1 Cpresulted from a twofold change in expression. Melting point analysisof experimental samples confirmed that all primers were specific fortheir respective transcripts, where there was only one melting pointobserved for each primer pair. Controls included RNA not subjectedto reverse transcription, and water only.

Immunoblotting. Muscles were removed from liquid nitrogen andhomogenized in 10 vol/muscle wet wt of modified lysis buffer [50mM Tris �HCl pH 7.4, 1% wt/vol Triton X-100, 0.25% sodiumdeoxycholate, 150 mM NaCl, 1 mM EGTA, protease inhibitor andphosphatase inhibitors (Sigma, St. Louis, MO)]. Homogenates werecentrifuged to pellet debris, and the total protein was measured in thesupernatant (Bio-Rad, Hercules, CA). Equal amounts of protein fromeach muscle lysate were separated by SDS-polyacrylamide gel elec-trophoresis (SDS-PAGE) and transferred to Polyvinylidene fluoridemembranes (Immobilon-P; Millipore, Bedford, MA). Membraneswere incubated in a blocking buffer [5% nonfat dry milk in Tris-buffered saline plus 0.1% Tween 20 (5% milk-TTBS)], and thenincubated with primary antibody diluted in 5% milk-TTBS overnightat 4°C. Primary antibodies included those for caspase-12 (1:1,000,#2202; Cell Signaling, Beverly, MA); phosphorylated and total focaladhesion kinase (FAK, 1:1,000 for each; #MAB1144, #06543; Up-state, Waltham, MA); integrin-�1D (1:1,000, ab8991; Abcam, Cam-bridge, MA); spectrin (1:500, ab11182; Abcam) and talin (1:2,000,T3287; Sigma). Antibodies for myomesin 1 (chMy2, 1:10,000) andmyomesin 2 (AA269, 1:100) were a kind gift from Irina Agarkova(Zurich, Switzerland). Antibodies for a muscle-ankyrin repeat protein(MARP) family members (diluted to 2 g/ml) were a kind gift fromSiegfried Labeit (Mannheim, Germany). Membranes were thenwashed in 5% milk-TTBS, and incubated with horseradish peroxi-dase-conjugated secondary antibody. After a series of washes in 5%milk-TTBS, TTBS, and TBS, detection was performed using en-hanced chemiluminescence scanned immediately by Kodak mm4000detection system (Eastman Kodak, Rochester, NY). Membranes werestained with Coomassie brilliant blue R-250 after immunoblotting toconfirm equal protein loading.

Muscle morphology and fiber typing. We subjected 10 m frozencross sections taken from the midbelly of each muscle to immuno-histochemistry for laminin (rabbit anti-laminin Ab-1; Neomarkers,Fremont, CA) to outline the muscle fibers. Fiber typing was per-formed with antibodies recognizing MHC 2a (SC-71), MHC 2b(BF-F3), MHC 1 (BAF-8), and fibers lacking MHC 2X (BF-35) aspreviously described (3, 40). Nuclei were counterstained with 4,6-diamidino-2-phenylindole. Stained sections were visualized on aLeica DMR microscope, captured with a digital camera using imageanalysis software (OpenLab, Improvision, UK). Cross-sectional areas

87EXPRESSION PROFILING OF MASSETERS REVEALS APOPTOSIS

Physiol Genomics • VOL 35 • www.physiolgenomics.org

on July 26, 2011physiolgenom

ics.physiology.orgD

ownloaded from

were assessed in at least 500 fibers from four nonoverlapping fields at�100 magnification from n � 3 muscle pairs.

MHC composition. Frozen muscles taken from n � 2 adult femalemice were homogenized in Laemmli buffer (23). Total protein in eachhomogenate was determined by a Bradford assay (Bio-Rad) andnormalized to total tissue wet weight, and equal protein amounts werethen separated using SDS-PAGE (44). Resolved gels were stained withCoomassie blue R-250 for 45 min followed by destaining for 2 h in 50%methanol-7% acetic acid. Images of the gels were captured with ahigh-performance digital camera (Eastman Kodak, Rochester, NY).

Apoptosis measurements. TUNEL staining was performed on 7 mparaffin sections of masseter and TA muscles (n � 5 muscle pairs)using TdT-FragEL DNA Fragmentation Detection kit (Calbiochem,EMD, Gibbstown, NJ) to determine the apoptotic index of the mus-cles. Subsequent immunohistochemistry was performed with anti-laminin (as above) and anti-dystrophin (DYS2, 1:100; Novacastra) todetermine the proportion of TUNEL positive nuclei within the musclefibers. TUNEL-positive myonuclei were calculated for four nonover-lapping fields at �200 magnification from each muscle, averaged, andpresented as percent TUNEL-positive myonuclei per sample. Nega-tive control samples were incubated without TdT enzyme.

Statistics. Data are presented as means � SE. Paired t-tests wereutilized for comparisons between masseter and TA from the sameanimal for qRT-PCR validation. Unpaired t-tests were utilized forcomparisons between groups for immunoblotting and TUNEL anal-ysis. Fiber size distributions were not normal and required the Mann-Whitney test for comparisons. Statistical significance was accepted forP � 0.05. Statistical analysis for the expression profiling data utilizedSAM (46) and is described above.

RESULTS

To identify differences between masticatory and axial mus-cles at the molecular level, expression profiling was performedon total RNA isolated from the masseter and TA muscles ofthree adult female mice. The quality of the microarray data issummarized in Table 1. All cRNA samples had a 260:280 ratiobetween 1.8 and 2.0. The percentage of transcripts expressed inall samples utilized for analysis exceeded 40% of the totalnumber of probe sets. Ratios of 3:5 for �-actin and GAPDHwere �3.0 for all samples. Out of �22,000 probe sets repre-sented on the chip, 8,220 transcripts were present in all mas-seter and TA samples. A pairwise comparison of signalstrength was performed between TA and masseter samples.Out of these, 158 transcripts were expressed at �1.50-folddifferences. Gene ontology of the differentially expressed tran-scripts was performed, and Fig. 1 represents the differentiallyexpressed transcripts for each masseter sample compared withthe respective TA sample for each functional classification.Transcripts with asterisks to the right were validated by addi-tional methods on independent samples including qRT-PCRand immunoblotting. Further analysis was limited to two func-tional classes of genes: the contraction/cytoskeleton transcriptsand those associated with stress and apoptosis.

Oligonucleotides were designed for seven of the identifiedtranscripts and GAPDH (Table 2), to validate the profilingresults by qRT-PCR on additional RNA samples (n � 7 musclepairs). These transcripts represented a range of differences inexpression level. The mean Cp were significantly lower in themasseters for Myh8, Ddit4l, Casp12, Egln3, and Mst4, whereasCp values for Myl3 and Il15 were significantly higher (Table 3).Using the mean difference between Cp values for each tran-script, we calculated the fold change in expression level be-tween masseter and TA muscles. In all cases, the trendsobserved by qRT-PCR were similar to those found by expres-sion profiling (Table 3), thereby validating the results for thisgroup of transcripts.

Genes indicating fiber type differences were observed(Figs. 1 and 2A), consistent with several previous studies (13,31, 47, 48). These included a 53-fold increase in neonatalMHC (Myh8) as well as a 66-fold decrease in slow skeletalalkali myosin light chain (Myl3) in masseters compared withTA muscles. The highest expression levels of MHC isoformsin masseters were observed in MHC 2x and 2a, whereas therewas an absence of MHC 1/�, embryonic, and �-cardiac in thistissue. MHC content was determined by separation of themyosin isoforms by SDS-PAGE (Fig. 2B). The predominantband in masseter muscles was MHC 2a/x, whereas TA muscleshad more MHC 2b plus MHC 2a/x, consistent with the expres-sion profiling results. A faint band in masseters migrated at thesame level as MHC 2b or �-cardiac MHC. The proportion ofthe MHC isoforms within this band was not determined due tothe lack of a specific antibody for the cardiac myosin isoform,but because there was no expression of �-cardiac MHC, thisband was most likely MHC 2b. Immunohistochemistry withantibodies recognizing the MHC isoforms also revealed fiberscoexpressing multiple isoforms (Fig. 2C).

Changes were also observed in genes that could regulatemuscle fiber size. Interleukin-15 (IL-15) is an anabolic cyto-kine (39) that was downregulated fivefold in masseters. DNA-damage-inducible transcript 4-like (Ddit4L, also known asSmhs1, RTP801L) had sixfold higher expression in masseters.This is a negative regulator of the mammalian target of rapa-mycin (mTOR) pathway (8), and expression increases in ratskeletal muscle subjected to hindlimb suspension (9). Fibercross-sectional area was determined in masseters and TAs byimmunohistochemistry with anti-laminin (Fig. 3A) and wassignificantly smaller in masseters (Fig. 3B) consistent withincreased expression of Ddit4l and decreased expression ofIL-15.

Surprisingly, the masseter possessed heightened expressionof genes responsive to cell stress and involved in apoptosis(Fig. 1). Validation of several of these genes was performedby qRT-PCR (Table 3). Specifically, expression of Ddit4l,caspase-12, Mst4, and Egln3 were elevated in masseter muscle.

Table 1. Quality of RNA samples for expression profiling

Muscle cRNA 260:280

Present Probe Sets (out of 22,690) �-Actin GAPDH

n % Signal 3:5 Signal 3:5

Masseter 1.92�0.02 10,338�862 45.6�2.2 3,114�148 2.76�0.03 8,494�570 2.67�0.13TA 1.91�0.02 10,018�293 44.1�1.3 3,029�100 2.69�0.06 11,953�100 1.94�0.04

Data are represented as means � SE for n � 3 samples per muscle.

88 EXPRESSION PROFILING OF MASSETERS REVEALS APOPTOSIS

Physiol Genomics • VOL 35 • www.physiolgenomics.org

on July 26, 2011physiolgenom

ics.physiology.orgD

ownloaded from

Todeterminewhetherincreasedexpressionofstressandproapop-totic genes is associated with apoptosis in this tissue, weperformed TUNEL staining on paraffin sections of masseterand TA muscles (n � 5 muscle pairs). Masseter muscles hadmore TUNEL-positive nuclei than TA muscles, which werefound at the periphery of the muscle fibers (Fig. 4A). To

determine whether the TUNEL-positive nuclei are myonu-clei, we performed subsequent immunohistochemistry withantibodies for laminin and dystrophin (Fig. 4B). The mas-seters exhibited significantly more TUNEL-positive myonu-clei than the TA muscles (Fig. 4C). Nuclei between thedystrophin and laminin staining (presumably satellite cells)

Fig. 1. Expression profiling of masseter andtibialis anterior (TA) muscles from adult fe-male mice. Values for expression differencesare presented as masseter:TA for each animal(labeled F1, F2, and F3). Blue tones indicatelower expression in masseters compared withTA; orange tones indicate higher expression inmasseters compared with TA. Classificationwas assigned through gene ontology tools onthe Affymetrix website (NetAffx), and litera-ture searches. *Transcripts that were validatedby quantitative RT-PCR or immunoblottingand displayed in subsequent figures.

89EXPRESSION PROFILING OF MASSETERS REVEALS APOPTOSIS

Physiol Genomics • VOL 35 • www.physiolgenomics.org

on July 26, 2011physiolgenom

ics.physiology.orgD

ownloaded from

had no detectable TUNEL staining. The identity of TUNEL-positive nuclei outside of the basal lamina was not deter-mined or quantified.

Immunoblotting for caspase-12 was performed on massetersand TA muscles (Fig. 5A, n � 3 muscle pairs). The caspase-12antibody epitope was immediately COOH-terminal to K158(CS in Fig. 5B), and so the protease involved in caspase-12cleavage could be identified (32). In masseters, cleavage andactivation of caspase-12 occurred at K158, which is an m-calpain site, and at D318, to result in an 18 kDa band. Cleavageand activation in the TA muscles occurred at the recognitionsite for caspase-7 (DEDD 94) resulting in the 40 kDa band.

Cytoskeletal genes were also differentially expressed. Ex-pression of talin, a component of the focal adhesion complex,was reduced by 3.6-fold in masseters. Immunoblotting forfocal adhesion complex proteins was performed to determinewhether the lower expression of talin found in masseters wasindicative of a decrease of the entire membrane complex.Because the masseter muscle fibers are smaller than those inTA muscles (Fig. 3) and the fiber surface area-to-volume ratioin masseter muscles is higher than in TA muscles, proteinlevels were normalized to spectrin, which has similar sar-colemmal localization. Spectrin expression did not differ be-tween masseter and TA muscles. There was no significantdifference in talin or �1D-integrin protein levels betweenmasseter and TA muscles (Fig. 6A, n � 3 muscle pairs),counter to the expression profiling results. To test if masseterswere sensing load through the focal adhesion complex, immu-noblotting for phosphorylated and total FAK was performed.Proportional phosphorylation of FAK was significantly higherin masseters (Fig. 6B).

Genes expressing components of the M-line (myomesins 1and 2) and I-band (Ankrd2) were at even lower levels than theother contractile genes in the microarray experiment. Myome-

sin expression was 1.7- and 5.3-fold lower in masseters formyomesins 1 and 2, respectively. Ankrd2 expression wasbetween 4- and 30-fold lower in masseters. Immunoblotting forthese components (normalized to tubulin) confirmed that myo-mesin 1 and 2 and Ankrd2 were lower in masseters than TA(Fig. 7). Furthermore, blotting for other family members ofAnkrd2 revealed that other proteins did not compensate forlower expression of Ankrd2 in masseters.

DISCUSSION

Past studies have shown that masticatory muscle is morpho-logically and physiologically distinct from axial muscles. Acontributing factor may be the embryonic origin of thesemuscles, where axial muscles are derived from somites, andmasticatory muscles arise from the neural crest. As establishedby several previous studies, fiber size is smaller, and there isincreased expression of neonatal MHC. Our study has revealedadditional and potentially novel features of masticatory mus-cle. In particular, there was an upregulation of stress-inducedgenes. Consistent with these changes, masseter muscle exhib-ited increased TUNEL-positive myonuclear staining, as well asa change in the cleavage/activation pattern of caspase-12,which is the canonical caspase for endoplasmic reticulum (ER)stress. The focal adhesion complex showed elevated activationas indicated by FAK phosphorylation, suggesting that in thenormal state, murine masseters are loaded to a greater extentthan TA muscles. Components associated with the cytoskele-ton that respond to and report mechanical stress (myomesin 1,2, and Ankrd2) were expressed at significantly lower levels,suggesting that mechanical loading in masseters may not beefficiently transmitted to the myonuclei. In contrast to cardiacand limb skeletal muscle, where sustained increased loaddrives muscle hypertrophy (16, 38), the apparent load onmasseters is associated with small fibers and high apoptoticindex, supporting a model in which load exacerbates massetermuscle damage and apoptosis. The masseter shares some butnot all features of expression profiling with extraocular, aging,regenerating, and unloaded muscles. Therefore, to our knowl-edge, the masseter has a unique molecular signature within thespectrum of muscle properties, which may provide insight intonovel load-sensing and stress mechanisms in skeletal muscle.

The stress-associated genes found in the masseters representmany different stress pathways. Ddit4l expression rises inresponse to DNA damage, hypoxia, and stress (8) and caninhibit Akt-mediated cell survival. The anabolic cytokine,IL-15, which was fivefold lower in masseters, also protectsagainst TNF-�-mediated apoptosis (6). In addition to thesetranscripts, there was a 3.5-fold increase of caspase-12, which

Table 2. Oligonucleotides and band sizes for qRT-PCR

Gene Accession No. Sense Primer (5-3) Antisense Primer (5-3) Band, bp

Myh8 NM_177369 acacatcttgcagaggaagg taacccagagaggcaagtg 104Myl3 NM_010859 atcacatacgggcagtgtggg gttgatgcagccgttggagtc 340IL15 NM_008357 gctgtgtcagtgtaggtctcc tcctccagctcctcacattcc 313Ddit4L NM_031043 agaacccggccagcatttcag tcacgataacacaacccc 300Casp12 NM_009808 ctggaaggaatctgtggggtg gaagagggaaccagtcttgcc 336Egln3 BC044926 aggccatcaacttcctcctgtc tgtggattcctgcggtctgacc 322Mst4 NM_133729 gcagggaaagtaacctcacc ggggattcatccgcagaacac 296

qRT-PCR, quantitative RT-PCR.

Table 3. Validation of profiling results by qRT-PCR

Gene �Cp (TA-Masseter)

Expression Difference (Ma:TA)

qRT-PCR Expression Profile

Myh8 6.53�0.86* 92.15�29.58 53.05�17.78Myl3 �5.34�0.52* �40.50�21.85 �66.70�18.99IL15 �2.24�0.71* �4.74�1.78 �5.11�0.38Ddit4L 2.53�0.64* 5.78�1.59 6.31�2.39Casp12 2.51�0.82* 5.69�1.86 3.55�1.94Egln3 3.20�0.72* 9.21�4.24 5.12�1.12Mst4 4.55�0.52* 23.51�14.06 9.72�1.95

Data are represented as means � SE for n � 7 muscle pairs. Cp, crossingpoint. *P � 0.05 for comparisons between masseters (Ma) and tibialis anterior(TA) muscles.

90 EXPRESSION PROFILING OF MASSETERS REVEALS APOPTOSIS

Physiol Genomics • VOL 35 • www.physiolgenomics.org

on July 26, 2011physiolgenom

ics.physiology.orgD

ownloaded from

is the member of the cysteine protease family associated withER stress-mediated apoptosis (33). The serine/threonine pro-tein kinase, Mst4 (MASK), had 10-fold higher expression inmasseters. Overexpression of Mst4 can induce apoptosis inmammalian cells, and the protein is also a caspase-3 substrate(10). Egln3 (SM-20) is a prolyl hydroxylase with fivefoldhigher expression in masseters. In response to anoxia, Egln3

modifies hypoxia-inducible factor alpha (HIF-�), to target thistranscription factor subunit for proteasomal degradation, thusinhibiting the ability of HIF to transactivate gene targets (17).Egln3 has also been shown to mediate cytochrome c releaseand caspase-dependent cell death in a neuronal cell line (42).At this point, it is not clear which of these multiple pathwayscontribute to the heightened apoptotic index in masseter mus-cles.

If myonuclear apoptosis were not counterbalanced by satel-lite cell proliferation and fusion to the fibers, muscle sizewould dwindle to nothing. Because this is not the case formurine masseters, it suggests that satellite cells are activelyreplacing these nuclei. Alternatively, it may be the satellite cellpool that is dividing and undergoing apoptosis without disturb-ing the nuclei within the fibers. The TUNEL-positive nuclei inthe masseters appeared to be within the fibers, indicated bytheir location within the dystrophin border (Fig. 4B), and therewas no evidence of apoptotic satellite cells between the lamininand dystrophin staining. The remaining TUNEL-positive nu-clei appear to be interstitial, but their identity was not deter-mined. Thus, it appears that the apoptotic myonuclei are beingreplaced by activated satellite cells, but this has not beenconfirmed. A similar phenomenon has been observed in ex-traocular muscles, where continued remodeling and increasedapoptosis exist simultaneously (26). The lack of central nucleiin the masseters and in the extraocular muscles, which is ahallmark of muscle remodeling, suggests that the mechanismof satellite cell fusion and muscle turnover may be verydifferent in the craniofacial muscles. Furthermore, the expres-sion profiling results support divergent regulation of prolifer-ation and differentiation in masseters and TA muscles (Fig. 1).It will require further investigation to determine whether a linkexists between proliferation and apoptosis in these muscles.

Caspase-12 is highly expressed in skeletal muscle, andlevels increase with age (7, 12). In the current study, there wasmore than threefold increase in caspase-12 expression in mas-seters. ER stress-mediated activation of caspase-12 by accu-mulation of unfolded/misfolded proteins or loss of calciumhomeostasis can result in apoptosis (reviewed in Ref. 28). This

Fig. 2. Myosin heavy chain (MHC) analysis shows that both murine masseter and TA muscles are composed of fast fiber types. A: MHC isoforms from profilingresults show that MHC 2x has the highest expression level in masseters. There was no apparent expression of embryonic (emb), �-cardiac (card), or 1/�-MHCin masseter or TA muscles. neo, Neonatal MHC (Myh8). B: SDS-PAGE MHC separation shows the predominant band for masseters in MHC 2a/x and for TAmuscles in MHC 2b. Dia, diaphragm; Mas, masseter; Ht, heart. C: fiber typing of adjacent sections of a masseter shows that there is coexpression of MHC 2xand 2a in a subset of fibers (*) and that most fibers contain MHC 2x by the lack of positive staining by the BF-35 antibody (�2x panel). Scale bar, 50 m.

Fig. 3. A: fibers appear smaller in the masseters compared with TA muscles,as shown by immunohistochemistry for laminin Scale bar, 100 m. B: fibersize distribution of masseter and TA muscles. In females, masseters havesignificantly smaller fibers relative to TA muscles by Mann-Whitney test.

91EXPRESSION PROFILING OF MASSETERS REVEALS APOPTOSIS

Physiol Genomics • VOL 35 • www.physiolgenomics.org

on July 26, 2011physiolgenom

ics.physiology.orgD

ownloaded from

activation occurs by m-calpain, triggered by high intracellularcalcium. Alternatively, mitochondrial disturbances can alsolead to caspase-12 activation via caspase-7. The cleavage sitesfor each protease are distinct (shown in Fig. 5), but both enablethe activation of caspase-12 (32). Our results demonstrate thatm-calpain is the primary activator of caspase-12 in masseters,suggesting the presence of high intracellular calcium, or aninability to properly handle calcium. Genes expressing calcium-handling proteins were not differentially expressed between thetwo muscle groups. Indeed, because the flux of ER/sarcoplas-mic reticular calcium is high during activation of any skeletalmuscle, it is not clear why the TA and masseter cleavagepatterns differed. A recent study clearly demonstrated that ERstress was high in limb skeletal muscle and increased with age(19), but the level of ER stress in masticatory muscles was notdetermined. Comparisons of ER stress in different musclegroups should be addressed in future studies.

Caspase activation is also involved in nonapoptotic pro-cesses, which are necessary for proper muscle development.Differentiation of satellite cells and myoblasts requires caspaseactivation to proceed. For example, primary myoblasts fromcaspase-3-deficient mice were defective in myotube formation(14), demonstrating that cleavage of caspase targets is neces-sary for muscle differentiation. More recently, ER stress hasbeen implicated as an enhancer of myotube formation (34),either by eliminating fragile cells at the onset of differentiationor by positively modulating the differentiating process. Itremains to be determined what role caspase-12 plays in mas-ticatory muscle: one leading to ER stress-induced apoptosis orone regulating muscle differentiation.

Because the focal adhesion complex is one of the primaryand well-characterized load sensors in muscle, our initial

impression was low talin expression in masseters indicated adeficit in load sensing via this complex. However, no differ-ences at the protein level were detected. The complex seemsintact and, in fact, had increased signaling contrary to ourinitial hypothesis (Fig. 6). Furthermore, genes encoding thedystrophin glycoprotein complex, which are also involved inload sensing, were present at levels similar to the TA (data notshown). Therefore, other differences in the expression profiledata may indicate increased load-induced cell stress or changesin the cellular response to load, perhaps within the contractileapparatus. These may include the reduction in myomesinexpression, a protein that is found at the M-line and has beenproposed to be part of the intracellular load-sensing apparatus(1). Studies of extraocular muscles have shown the M-line inthe outer orbital layer and inner global displays a lack oforganization and diminished myomesin expression (1, 49),suggesting this may be a common feature of the craniofacialmusculature. Second, deficits were also observed in Ankrd2,which is found at the I band. Ankrd2, a MARP family member,senses dynamic load and translocates to the nucleus from the Iband after stretch, contractile activity, or injury, where itinteracts with the tumor suppressor p53 to potentially regulategene expression (21, 27, 45). Whether these cytoskeletal pro-teins contribute to integration of load sensing bears furtherinvestigation.

Comparison of the masseter profile to expression profiling ofregenerating muscle shows similarities to TA and gastrocne-mius at 2 wk after cardiotoxin injection (50, 53). Neonatalmyosin expression remains high (5–10� increase) from 3 to 14days of regeneration. In addition, expression of myomesins 1and 2 is significantly reduced during this phase of musclerepair. These patterns may be indicative of muscle undergoing

Fig. 4. A: terminal deoxynucleotidyltransferase-mediated nickend labeling (TUNEL)-positive nuclei are at the periphery ofmuscle fibers in the masseters (arrowheads) and not evident in theTA muscles. Negative control displays the lack of backgroundstaining in the cryosections. B: identification of TUNEL-positivemyonuclei was achieved by immunohistochemistry with lamininand dystrophin antibodies subsequent to TUNEL staining. Thepanel provides one example (*) of a TUNEL-positive nucleuswithin the dystrophin and laminin borders. DAPI, 4,6-diamidino-2-phenylindole. C: masseter muscles have more TUNEL-positivemyonuclei than TA by unpaired t-test. The mean percentages ofTUNEL-positive nuclei were calculated from 4 nonoverlappinghigh-powered fields per sample.

92 EXPRESSION PROFILING OF MASSETERS REVEALS APOPTOSIS

Physiol Genomics • VOL 35 • www.physiolgenomics.org

on July 26, 2011physiolgenom

ics.physiology.orgD

ownloaded from

turnover. However, increased expression of the basic helix-loop-helix factors, which coordinate myogenesis and are tran-siently upregulated during muscle regeneration and reloadingregardless of muscle fiber type (11, 50, 53), was not evident in

the masseter profile. Furthermore, no changes in any of thestress-associated genes identified in the current study werefound in either muscle regeneration profile. These comparisonscounter a model in which masticatory muscle is continuouslyremodeling.

From a pathological perspective, masticatory muscles arethe site for myogenic pain in the largest proportion of sufferersof temporomandibular joint disorder (TMD), for which there isno established cause (18). Previous studies have proposed thatinefficient repair of masticatory muscle could lead to theprolonged pain associated with TMD (37). Our observationssuggest that repair and muscle turnover may be occurring, butthat nuclei also undergo cell death, thereby preventing resolu-tion of damage. If the native state of masticatory muscle is oneof heightened cell stress, this could provide a pathologicalenvironment for the onset of disease or myogenic pain.

Genetic diseases including both Duchenne muscular dystro-phy and myotonic dystrophy (DM) also affect the muscles ofmastication (20, 52). The genetic basis of each disease has beenidentified, but how the disease manifests in masticatory mus-cles compared with limb and trunk muscles differs in timingand severity. In DM patients, for example, the masseter andanterior temporalis are more severely affected than other mas-ticatory muscles, including significant muscle atrophy andweakness, and fatty infiltration. In contrast, jaw muscles suchas the anterior digastric, display milder pathohistologicalchanges (2, 35).

Differences in susceptibility and symptoms could be depen-dent on the development of these muscle groups or theirphysiological state. Alternatively, masticatory muscles maysimply adapt to the patterns of activity, as in other examples ofmuscles plasticity, resulting in a muscle with very differentfiber properties than limb muscles. For instance, a correlationbetween fiber type and the stress gene profile may exist but hasnot been directly addressed in this study. Part of the adaptationmay include muscle turnover or compensatory hypertrophy,which would give rise to activated satellite cells and expressionof embryonic and neonatal myosin. Indeed, the association ofmasseter hypertrophy with TMD pathology has been proposed(41) and could be part of a spectrum of responses within themasticatory muscles.

Ultimately, the emerging story is that there may be a fun-damentally different strategy for muscle design in the cranio-

Fig. 5. Caspase-12 processing and differential cleavage in masseter and TAmuscles. A: immunoblotting reveals distinct banding patterns exist between mas-seter and TA muscles. MW, molecular weight (in kDa). B: letters designate bandsin the map produced by cleavage sites in caspase-12 and detectable by thecaspase-12 antibody (CS). Caspase-7 utilizes DEDD94 (black arrow) and m-calpain uses T132 and K158 (white arrows) for initial cleavage (32). Fullactivation requires cleavage between the large and small subunits (*). Band sizesin masseter immunoblotting lanes can only arise by cleavage of caspase 12 atm-calpain site K158.

Fig. 6. Immunoblotting for members of the focal adhesion complex. A: talinand �1D-integrin (�1D Int) protein levels in masseter (Mas) and TA musclesdo not differ. Intensity was normalized to spectrin on the same membrane afterstripping, However, increased focal adhesion kinase phosphorylation (P-FAK)was elevated in masseters. T-FAK, total focal adhesion kinase. B: relative FAKphosphorylation is significantly higher in masseters compared with TA mus-cles (n � 3 for each muscle). *P � 0.05 by unpaired t-test.

Fig. 7. Immunoblotting for myomesin and a muscle-ankyrin repeat proteinfamily members (normalized to tubulin). Myomesin 1 (Myom1) and myomesin2 (Myom2) are lower in masseters (Ma) consistent with expression profilingresults. Ankrd2 is also lower in masseters, with no change in diabetes-associated ankyrin repeat protein (DARP) or cardiac ankyrin repeat protein(CARP).

93EXPRESSION PROFILING OF MASSETERS REVEALS APOPTOSIS

Physiol Genomics • VOL 35 • www.physiolgenomics.org

on July 26, 2011physiolgenom

ics.physiology.orgD

ownloaded from

facial muscles in general compared with axial muscles. Spe-cifically, there are small diameter fibers that have an attenuatedability to hypertrophy, but an increased propensity to undergoapoptosis. This strategy seems evident on several levels. Forexample, masticatory muscles have increased expression ofDdit4L, a known inhibitor of the primary cell survival andhypertrophic pathway in muscle: the Akt/mTOR pathway (5,8). Ddit4L may limit a normal drive for growth or survival inthe muscles of mastication. Second, lower expression of IL-15,an anabolic cytokine, may prevent progrowth signals frombeing executed. Third, masseters appear to have heightened ERstress, which could turn down the gain on anabolic signalsthrough a reduction in general translation. Many of thesecellular processes lie at the interface between growth andapoptosis, and so the question remains if apoptosis is aninevitable consequence of attenuating hypertrophy or an addi-tional strategy for myonuclear elimination and growth preven-tion. These features are apparent not only in the masseters butalso in the extraocular muscles, where small muscle fibers areundergoing continuous nuclear turnover (26), suggesting thismay a generalized strategy for the craniofacial muscles.Whether this adaptation positions masticatory muscles at a setpoint that is ideal for their functions, or if these strategies leadto heightened susceptibilities to specific genetic or use-depen-dent pathologies is unknown. Further study is needed to deter-mine if and how these characteristics are linked to the physi-ology and pathology associated with this muscle group.

ACKNOWLEDGMENTS

Expression profiling was performed at the University of PennsylvaniaMicroarray Facility under the direction of Dr. D. A. Baldwin.

GRANTS

Support for this study was from a Rabinowitz Award for Excellence inResearch. C. Kulkarni was supported by a summer research fellowship fromUniversity of Pennsylvania School of Dental Medicine.

REFERENCES

1. Agarkova I, Perriard JC. The M-band: an elastic web that crosslinksthick filaments in the center of the sarcomere. Trends Cell Biol 15:477–485, 2005.

2. Bakke M, Kirkeby S, Jensen BL, Hansen HJ, Kreiborg S, Hjorting-Hansen E, Michler L, Moller E. Structure and function of masticatorymuscles in a case of muscular dystrophy. J Oral Pathol Med 19: 335–340,1990.

3. Barton ER, Gimbel JA, Williams GR, Soslowsky LJ. Rat supraspinatusmuscle atrophy after tendon detachment. J Orthop Res 23: 259–265, 2005.

4. Barton ER. Viral expression of insulin-like growth factor-I isoformspromotes different responses in skeletal muscle. J Appl Physiol 100:1778–1784, 2006.

5. Bodine SC, Stitt TN, Gonzalez M, Kline WO, Stover GL, Bauerlein R,Zlotchenko E, Scrimgeour A, Lawrence JC, Glass DJ, YancopoulosGD. Akt/mTOR pathway is a crucial regulator of skeletal muscle hyper-trophy and can prevent muscle atrophy in vivo. Nat Cell Biol 3: 1014–1019, 2001.

6. Bulfone-Paus S, Ungureanu D, Pohl T, Lindner G, Paus R, Ruckert R,Krause H, Kunzendorf U. Interleukin-15 protects from lethal apoptosisin vivo. Nat Med 3: 1124–1128, 1997.

7. Chung L, Ng YC. Age-related alterations in expression of apoptosisregulatory proteins and heat shock proteins in rat skeletal muscle. BiochimBiophys Acta 1762: 103–109, 2006.

8. Corradetti MN, Inoki K, Guan KL. The stress-inducted proteinsRTP801 and RTP801L are negative regulators of the mammalian target ofrapamycin pathway. J Biol Chem 280: 9769–9772, 2005.

9. Cros N, Tkatchenko AV, Pisani DF, Leclerc L, Leger JJ, Marini JF,Dechesne CA. Analysis of altered gene expression in rat soleus muscleatrophied by disuse. J Cell Biochem 83: 508–519, 2001.

10. Dan I, Ong SE, Watanabe NM, Blagoev B, Nielsen MM, Kajikawa E,Kristiansen TZ, Mann M, Pandey A. Cloning of MASK, a novelmember of the mammalian germinal center kinase III subfamily, withapoptosis-inducing properties. J Biol Chem 277: 5929–5939, 2002.

11. Dapp C, Schmutz S, Hoppeler H, Fluck M. Transcriptional reprogram-ming and ultrastructure during atrophy and recovery of mouse soleusmuscle. Physiol Genomics 20: 97–107, 2004.

12. Dirks AJ, Leeuwenburgh C. Aging and lifelong calorie restriction resultin adaptations of skeletal muscle apoptosis repressor, apoptosis-inducingfactor, X-linked inhibitor of apoptosis, caspase-3, and caspase-12. FreeRadic Biol Med 36: 27–39, 2004.

13. Eason JM, Schwartz GA, Pavlath GK, English AW. Sexually dimor-phic expression of myosin heavy chains in the adult mouse masseter.J Appl Physiol 89: 251–258, 2000.

14. Fernando P, Kelly JF, Balazsi K, Slack RS, Megeney LA. Caspase 3activity is required for skeletal muscle differentiation. Proc Natl Acad SciUSA 99: 11025–11030, 2002.

15. Fischer MD, Gorospe JR, Felder E, Bogdanovich S, Pedrosa-DomellofF, Ahima RS, Rubinstein NA, Hoffman EP, Khurana TS. Expressionprofiling reveals metabolic and structural components of extraocularmuscles. Physiol Genomics 9: 71–84, 2002.

16. Fluck M, Carson JA, Gordon SE, Ziemiecki A, Booth FW. Focaladhesion proteins FAK and paxillin increase in hypertrophied skeletalmuscle. Am J Physiol Cell Physiol 277: C152–C162, 1999.

17. Freeman RS, Hasbani DM, Lipscomb EA, Straub JA, Xie L. SM-20,EGL-9, and the EGLN family of hypoxia-inducible factor prolyl hydroxy-lases. Mol Cell 16: 1–12, 2003.

18. Greene CS. Concepts of TMD etiology: effects on diagnosis and treat-ment. In: TMDs: An Evidence-Based Approach to Diagnosis and Treat-ment, edited by Laskin DM, Greene CS, Hylander WL. Hanover Park, IL:Quintessence, 2006, p. 219–228.

19. Iwawaki T, Akai R, Kohno K, Miura M. A transgenic mouse model formonitoring endoplasmic reticulum stress. Nat Med 10: 98–102, 2004.

20. Kiliaridis S, Katsaros C. The effects of myotonic dystrophy and Duch-enne muscular dystrophy on the orofacial muscles and dentofacial mor-phology. Acta Odontol Scand 56: 369–374, 1998.

21. Kojic S, Medeot E, Guccione E, Krmac H, Zara I, Martinelli V, ValleG, Faulkner G. The Ankrd2 protein, a link between the sarcomere and thenucleus in skeletal muscle. J Mol Biol 339: 313–325, 2004.

22. Korfage JAM, Van Eijden TMGJ. Myosin Isoform Composition of theHuman Medial and Lateral Pterygoid. J Dent Res 79: 1618–1625, 2000.

23. Laemmli UK. Cleavage of structural proteins during the assembly of thehead of bacteriophage T4. Nature 227: 680–685, 1970.

24. Lu JR, Bassel-Duby R, Hawkins A, Chang P, Valdez R, Wu H, GanL, Shelton JM, Richardson JA, Olson EN. Control of facial muscledevelopment by MyoR and capsulin. Science 298: 2378–2381, 2002.

25. Marcucio RS, Noden DM. Myotube heterogeneity in developing chickcraniofacial skeletal muscles. Dev Dyn 214: 178–194, 1999.

26. McLoon LK, Rowe J, Wirtschafter J, McCormick KM. Continuousmyofiber remodeling in uninjured extraocular myofibers: myonuclearturnover and evidence for apoptosis. Muscle Nerve 29: 707–715, 2004.

27. Miller MK, Bang ML, Witt CC, Labeit D, Trombitas C, Watanabe K,Granzier H, McElhinny AS, Gregorio CC, Labeit S. The muscleankyrin repeat proteins: CARP, ankrd2/Arpp and DARP as a family oftitin filament-based stress response molecules. J Mol Biol 333: 951–964,2003.

28. Momoi T. Caspases involved in ER stress-mediated cell death. J ChemNeuroanat 28: 101–105, 2004.

29. Monemi M, Kadi F, Liu JK, Thornell LE, Eriksson PO. Adversechanges in fibre type and myosin heavy chain compositions of human jawmuscle vs. limb muscle during ageing. Acta Physiol Scand 167: 339–345,1999.

30. Mootoosamy RC, Dietrich S. Distinct regulatory cascades for head andtrunk myogenesis. Development 129: 573–583, 2002.

31. Morris TJ, Brandon CA, Horton MJ, Carlson DS, Sciote JJ. Maxi-mum shortening velocity and myosin heavy-chain isoform expression inhuman masseter muscle fibers. J Dent Res 80: 1845–1848, 2001.

32. Nakagawa T, Yuan J. Cross-talk between two cysteine protease families.Activation of caspase-12 by calpain in apoptosis. J Cell Biol 150: 887–894, 2000.

33. Nakagawa T, Zhu H, Morishima N, Li E, Xu J, Yankner BA, Yuan J.Caspase-12 mediates endoplasmic-reticulum-specific apoptosis and cyto-toxicity by amyloid-beta. Nature 403: 98–103, 2000.

94 EXPRESSION PROFILING OF MASSETERS REVEALS APOPTOSIS

Physiol Genomics • VOL 35 • www.physiolgenomics.org

on July 26, 2011physiolgenom

ics.physiology.orgD

ownloaded from

34. Nakanishi K, Dohmae N, Morishima N. Endoplasmic reticulum stressincreases myofiber formation in vitro. FASEB J 21: 2994–3003, 2007.

35. Odman C, Kiliaridis S. Masticatory muscle activity in myotonic dystro-phy patients. J Oral Rehabil 23: 5–10, 1996.

36. Parkyn G, Mootoosamy RC, Cheng L, Thorpe C, Dietrich S. Hypaxialmuscle development. Results Probl Cell Differ 38: 127–141, 2002.

37. Pavlath GK, Thaloor D, Rando TA, Cheong M, English AW, Zheng B.Heterogeneity among muscle precursor cells in adult skeletal muscles withdiffering regenerative capacities. Dev Dyn 212: 495–508, 1998.

38. Pham CG, Harpf AE, Keller RS, Vu HT, Shai SY, Loftus JC, Ross RS.Striated muscle-specific beta(1D)-integrin and FAK are involved in car-diac myocyte hypertrophic response pathway. Am J Physiol Heart CircPhysiol 279: H2916–H2926, 2000.

39. Quinn LS, Anderson BG, Drivdahl RH, Alvarez B, Argiles JM. Overex-pression of interleukin-15 induces skeletal muscle hypertrophy in vitro:implications for treatment of muscle wasting disorders. Exp Cell Res 280:55–63, 2002.

40. Schiaffino S, Salviati G. Molecular diversity of myofibrillar proteins:isoform analysis at the protein and mRNA level. In: Methods in MuscleBiology, edited by Emerson CP and Sweeney HL. New York: AcademicPress, 1997, p. 349–369.

41. Sciote JJ, Horton MJ, Rowlerson AM, Link J. Specialized cranialmuscles: how different are they from limb and abdominal muscles? CellsTissues Organs 174: 73–86, 2003.

42. Straub JA, Lipscomb EA, Yoshida ES, Freeman RS. Induction ofSM-20 in PC12 cells leads to increased cytochrome c levels, accumulationof cytochrome c in the cytosol, and caspase-dependent cell death. J Neu-rochem 85: 318–328, 2003.

43. Tajbakhsh S, Rocancourt D, Cossu G, Buckingham M. Redefining thegenetic hierarchies controlling skeletal myogenesis: Pax-3 and Myf-5 actupstream of MyoD. Cell 89: 127–138, 1997.

44. Talmadge RJ, Roy RR. Electrophoretic separation of rat skeletal musclemyosin heavy-chain isoforms. J Appl Physiol 75: 2337–2340, 1993.

45. Tsukamoto Y, Hijiya N, Yano S, Yokoyama S, Nakada C, Uchida T,Matsuura K, Moriyama M. Arpp/Ankrd2, a member of the muscleankyrin repeat proteins (MARPs), translocates from the I-band to thenucleus after muscle injury. Histochem Cell Biol 129: 55–64, 2008.

46. Tusher VG, Tibshirani R, Chu G. Significance analysis of microarraysapplied to the ionizing radiation response. Proc Natl Acad Sci USA 98:5116–5121, 2001.

47. Tuxen A, Kirkeby S. An animal model for human masseter muscle:histochemical characterization of mouse, rat, rabbit, cat, dog, pig, and cowmasseter muscle. J Oral Maxillofac Surg 48: 1063–1067, 1990.

48. Widmer CG, Morris-Wiman JA, Nekula C. Spatial distribution ofmyosin heavy-chain isoforms in mouse masseter. J Dent Res 81: 33–38,2002.

49. Wiesen MH, Bogdanovich S, Agarkova I, Perriard JC, Khurana TS.Identification and characterization of layer-specific differences in extraoc-ular muscle m-bands. Invest Ophthalmol Vis Sci 48: 1119–1127, 2007.

50. Yan Z, Choi S, Liu X, Zhang M, Schageman JJ, Lee SY, Hart R, LinL, Thurmond FA, Williams RS. Highly coordinated gene regulation inmouse skeletal muscle regeneration. J Biol Chem 278: 8826–8836, 2003.

51. Yu F, Stal P, Thornell LE, Larsson L. Human single masseter musclefibers contain unique combinations of myosin and myosin binding proteinC isoforms. J Muscle Res Cell Motil 23: 317–326, 2002.

52. Zanoteli E, Yamashita HK, Suzuki H, Oliveira AS, Gabbai AA.Temporomandibular joint and masticatory muscle involvement in myo-tonic dystrophy: a study by magnetic resonance imaging. Oral Surg OralMed Oral Pathol Oral Radiol Endod 94: 262–271, 2002.

53. Zhao P, Iezzi S, Carver E, Dressman D, Gridley T, Sartorelli V,Hoffman EP. Slug is a novel downstream target of MyoD. Temporalprofiling in muscle regeneration. J Biol Chem 277: 30091–30101,2002.

95EXPRESSION PROFILING OF MASSETERS REVEALS APOPTOSIS

Physiol Genomics • VOL 35 • www.physiolgenomics.org

on July 26, 2011physiolgenom

ics.physiology.orgD

ownloaded from