effects of maternal hyperthermia on myogenesis-related factors in developing upper limb

Effects of Maternal Hyperthermia on Myogenesis-RelatedFactors in Developing Upper Limb

Jin Lee,1 Philip E. Mirkes,2 Doo Jin Paik,1 and Won Kyu Kim1*1Department of Anatomy and Cell Biology, College of Medicine, Hanyang University, Seoul, Korea

2Department of Veterinary Physiology and Pharmacology, Texas A&M University, College Station, Texas

Received 31 March 2008; Revised 5 September 2008; Accepted 14 September 2008

BACKGROUND: Maternal hyperthermia is one causative factor in various congenital anomalies in experi-mental animals and humans. In the present study, we assessed the effects of high temperature on limb myo-genesis in mice. METHODS: Pregnant mice, C57BL/6 strain, were exposed to hyperthermia (438C, 5 minutes)on embryonic day (ED) 8. Fetuses on ED 11, 13, 15, and 17 and neonates on postnatal day (PD) 1 were col-lected. To characterize the effects of hyperthermia on myogenesis-related factors Pax3, MyoD, myogenin,and myosin heavy chain (MyHC) during skeletal muscle development, we performed RT-PCR, western blot-ting, immunohistochemistry, and transmission electron microscopy. RESULTS: Pax3 gene expression was stilldetected on ED 13 in hyperthermia-exposed fetuses. The expression of MyoD protein was down-regulatedin fetuses exposed to hyperthermia. In contrast, myogenin and MyHC protein expression were up-regulatedon PD 1 and ED 17, respectively, in the group exposed to hyperthermia. Immunohistochemical analysis con-firmed the findings from western blot analysis. Compared with control neonates, a TEM study revealedimmature muscle fibers in PD 1 hyperthermia neonates. Thus, our studies showed that maternal hyperther-mia induced delayed expression of Pax3 and inhibited expression of MyoD proteins, which are known toplay important roles in migration of myogenic progenitor cells, and in myoblast proliferation. In addition,maternal hyperthermia also delayed the expression of myogenin protein for the formation of myotubes, andMyHC protein, which is one of the final muscle differentiation factors. CONCLUSION: Our data suggest thatmaternal hyperthermia delays limb myogenesis in part by disregulating the expression of key myogenesis-related factors. Birth Defects Research (Part A) 85:184–192, 2009. � 2009 Wiley-Liss, Inc.

Key words: maternal hyperthermia; myogenesis-related factors; skeletal muscle development; developingupper limb; delayed myogenesis

INTRODUCTION

Numerous perturbations of internal and external envi-ronments during pregnancy are known or postulated toaffect normal development. For example, maternal hyper-thermia, a well-known animal and human teratogen,increases prostaglandin secretion by both mother and fe-tus (Adrianakis et al., 1989) and activates uterine contrac-tions that can induce fetal distress (Morishima et al.,1975). Consequently, heat exposure during pregnancycan result in abortion, growth retardation, and birthdefects (Edwards et al., 2003). Generally, hyperthermiacan be defined as 1.58C or more above normal core tem-perature. Many reports document that maternal hyper-thermia is associated with a variety of congenital anoma-lies in human and experimental animals. In experimentalanimals, hyperthermia induces a variety of birth defects,such as anencephaly, exencephaly, microphthalmia,

arthrogryposis, abdominal wall defects, and limb anoma-lies. In addition, epidemiologic studies provide strongevidence that maternal hyperthermia is an importantcausative factor that is linked to several congenitalanomalies in humans (Martinez-Frias et al., 2001).

This work was presented at the 47th annual meeting of Teratology Soci-ety, June 23–28, 2007, Pittsburgh, Pennsylvania.

*Correspondence to: Won Kyu Kim M.D., Ph.D., Department of Anatomy andCell Biology, College of Medicine, Hanyang University, Seoul, South Korea.E-mail: [email protected]

This work was supported by the Korean Research Foundation Grant fundedby the Korean Government (MOEHRD, Basic Research Promotion Fund)(KRF-2007-000-0000-5739)

Published online 29 January 2009 in Wiley InterScience (www.interscience.wiley.com).DOI: 10.1002/bdra.20538

Birth Defects Research (Part A): Clinical and Molecular Teratology 85:184–192 (2009)

� 2009 Wiley-Liss, Inc. Birth Defects Research (Part A) 85:184–192 (2009)

Cell death, membrane disruption, vascular disruption,and placental infarction have been implicated in causingembryonic damage exposed to hyperthermia. Amongthem, heat-induced vascular disruption has been impli-cated in the pathogenesis of Moebius syndrome, oroman-dibular-limb hypogenesis syndrome, and the amyoplasiaform of arthrogryposis in humans (reviewed by Grahamet al., 1998). Webster et al. (1987, 1988) reported that anumber of agents, including hyperthermia, caused em-bryonic vascular damage with hemorrhage and resultedin hypoplasia of limbs and digits, as well as otherdefects. Martinez-Frias et al. (2001) also reported that ahuman female fetus with severe limb deficiencies wasexposed to hyperthermia for 2 days during the 2nd and4th months, and suggested vascular damage with throm-bosis (infarction) as the pathogenetic mechanism.

Although the mechanisms by which hyperthermiacauses congenital anomalies have been studied exten-sively, particularly neural tube defects, the effects ofhyperthermia on the skeletal muscle development, espe-cially in developing upper limbs, have not been wellstudied. Skeletal muscles of vertebrates are derived frommyogenic progenitor cells (MPCs) that develop fromsomites lateral to the neural tube and notochord (Christand Ordahl, 1995). MPCs ventrolateral to the dermomyo-tome continue to proliferate and migrate toward thedeveloping limb bud (Rawls and Olson, 1997). Duringmuscle development, signals and transcription of manyfactors are required, and spatiotemporal expression pat-terns of these factors occur sequentially (Chen and Gold-hamer, 2004). Pax3, one of these factors, plays a key rolein the establishment and survival of cells in the hypaxialdermomyotome, and in the delamination and migrationof MPCs (Tajbakhsh and Buckingham, 2000). Duringmigration, MPCs express Pax3, which induces the expres-sion of one of myogenic regulatory factors (MRFs), MyoD(Kassar-Duchossoy et al., 2004). Once MPCs start tomigrate to the developing limb bud, a number of MRFsare activated. Among them, myogenic determination fac-tors such as MyoD and Myf5 are expressed. Thereafter,myogenic differentiation factors such as myogenin, Mrf4,and MyHC, which are required for proliferation and mat-uration of muscle fibers, are activated (Buckingham et al.,2003). Once migrating cells arrive in their final destina-tion, they cease mitosis and fuse to form multinucleatedmyotubes, which are subsequently lengthened to becomemuscle fibers (Olson et al., 1991). In skeletal muscle de-velopment, MRFs play an activating role in muscle differ-entiation program. MyoD, which is activated by Pax3, isexpressed in early stage of muscle development and dif-ferentiation of MPCs into the myoblasts forming hypaxialmuscle. Myogenin plays a role in fusion of myoblasts,activation of muscle-related genes of adult type and mus-cle differentiation (Kablar et al., 1997). In addition, struc-tural muscle proteins, such as MyHC, are expressed inthe period of formation of myotubes. MyHC is synthe-sized in embryonic, neonatal, and adult muscles, andshows expression in myotube and muscle fibers (McKoyet al., 1998). Thus, skeletal myogenesis in developingupper limbs is closely related with several factors such asPax3, MyoD, myogenin and MyHC, which collaboratewith each other sequentially and complete muscle devel-opment through complex processes.

In the studies reported herein, we assessed the effectsof maternal hyperthermia on the myogenesis in develop-

ing upper limb. After heat exposure on embryonic day(ED) 8, upper limbs were excised and used to character-ize the expression of Pax3, MyoD, myogenin, and MyHCgenes and proteins by RT-PCR, western blotting,and immunohistochemistry, and transmission electronmicroscopy.

MATERIALS AND METHODSMaternal Hyperthermia and Collection of

Fetuses and Neonates

We used primigravida C57BL/6 strain mice as experi-mental animals. The morning following copulation, atwhich the vaginal plug was observed, was designatedday 0 of gestation. Time-mated pregnant mice wereexposed to hyperthermia in vivo on ED 8 using the fol-lowing method. Pregnant mice were placed in 50-ml per-forated Falcon tube, and dipped in a 438C water bath.Using a digital thermometer possessing two probes, oneprobe is inserted into the mouse rectum, and the other isdipped in the water bath for checking the water tempera-ture during heat treatment. Once body temperaturereached 438C, the pregnant mouse was maintained atthat temperature for 5 minutes To prevent hypothermiaafter heat treatment, the mice were wiped with papertowels and placed in a 388C incubator. In parallel, for thecontrol group, other pregnant C57BL/6J mice receivedthe rectal probe but not restraint or heat treatment, werewiped with clean water by wet paper towel, and wereplaced in a 388C incubator. Fetuses on ED 11, 13, 15, and17 and neonates on postnatal day (PD) 1 were collected.

Total RNA Isolation

Total RNA was isolated from whole upper limbs on ED11 and flexor muscles on ED 13, 15, and 17 and bicepsbrachii muscle on PD 1 by using TRIzol (Invitrogen, Carls-bad, CA). Then 1 lg of total RNA was treated with DNaseI (Life Technologies, Karlsruhe, Germany). Total RNA con-centration was calculated (Smartspec Plus spectrophotom-eter, BioRAD, Hercules, CA, USA).

RT-PCR

The RNA samples were reverse-transcribed withSuperScript II (Gibco BRL, Rockville, MD) and mix ofoligo dT and random primers in a 12 ll total reactionvolume at 458C for 30 minutes. PCR was performedusing the cDNA as a template. The PCR reaction using aTaq DNA polymerase (TAKARA) and pair of primersdesigned using BLAST algorithms (GenBank). The primersequences used and conditions of synthesis are providedin Table 1. PCR products were separated on 1.2% to 1.5%agarose gels using a T3000 Thermocycler (Biometra,Goettingen, Germany). The densities of each band wereanalyzed by gel documentation (Gel Doc 2000, QuantityOne, Bio-Rad, Hercules, CA).

Western Blotting

Whole upper limbs on ED 11 and flexor muscles onED 13, 15, and 17 and biceps brachii muscle on PD 1were homogenized in lysis buffer (Cell Signaling Tech-nology, Danvers, MA) with PMSF at 48C and then centri-fuged (13,0003 < g). The protein content of each samplewas determined by the Bradford method with bovine

185MATERNAL HYPERTHERMIA AND MYOGENESIS-RELATED FACTORS

Birth Defects Research (Part A) 85:184–192 (2009)

serum albumin as a standard. Protein samples (35 lg)were boiled with 53 < sample buffer, electrophoresis onpolyacrylamide gels, followed by transfer to a nitrocellu-lose membrane at 15 V overnight.

The membrane was washed, blocked, and incubatedfor Western blotting to detect the MyoD, Myogenin,MyHc (1:1000, 1:1500, 1:2500; Santa Cruz Biotechnology,Santa Cruz, CA) for 12 hours at 48C. Horseradish peroxi-dase-conjugated antirabbit secondary antibody (SantaCruz Biotechnology, Santa Cruz, CA) was used at 1:2000for 1 hour at room temperature. Membranes werewashed and visualized by autoradiography after devel-opment with ECL Plus (Amersham Life Sciences). Densi-tometry was performed with gel documentation (Gel Doc2000, Quantity One, Bio-Rad, Hercules, CA).

Immunohistochemistry for MyoD, Myogeninand MyHc

Tissue samples were fixed in immunofixative (4% para-formaldehyde, 0.1% glutaraldehyde, 0.1 M phosphatebuffer) for 2 hours. After fixation, all tissues were dehy-drated in ethanol and embedded in paraffin. After cuttingthrough the longitudinal plane at a thickness of 6 lm, alltissue sections were deparaffinized, cleared, and hydratedto phosphate buffered saline (PBS) using a descending se-ries of ethanol concentrations. The sections were blockedfor 40 minutes at 378C with 3% goat serum in PBS fol-lowed by quenching endogenous peroxidase activity byexposing slides to 0.3% H2O2 and 10% methanol for5 minutes. Primary antibodies for MyoD, Myogenin, andMyHc (1:45, 1:50, 1:100 dilution, rabbit polyclonal; SantaCruz) were added and incubated overnight at 48C. Thenext day, the slides were washed in PBS three times for5 minutes, then secondary antibody (goat antirabbit; 1:200Vactor Laboratory, Burlington, ON, Canada) was appliedfor 30 minutes at room temperature. The peroxidase activ-ity was then visualized with a DAB (3,3’-diaminobenzi-dine) substrate kit (Vector Laboratories). Finally, the slideswere counterstained with 0.5% methyl green. For negativecontrols the primary antibodies were omitted.

Transmission Electron Microscopy

The biceps brachii muscles from PD 1 neonates werefixed in cold fixative of 2.5% glutaraldehyde in 0.1 Mphosphate buffer, pH 7.4. After 2 hours of fixation, thespecimens were postfixed in 1% OsO4 and dehydrated in

graded alcohol. Decreasing concentrations of propyleneoxide and increasing EmBed-812 (Electron MicroscopyServices, Fort Washington, PA) were used. After purefresh resin embedding and polymerization, 1-lm–thicksections were initially cut and stained with methyl blue.After determining appropriate site, ultrathin sections (60to 80 nm thick) were made, stained with uranyl acetateand lead citrate, and observed with transmission electronmicroscope (Hitachi S7600, Tokyo, Japan) at 80 kV.

Statistical Analysis

Statistical analyses were performed using SPSS software(SPSS Inc, Chicago, IL). All data were expressed as mean6 SD. The whole bodyweight data were analyzed by one-way ANOVA, and concentration of myogenesis-related fac-tors genes and proteins in the upper limb were performedwith Student’s t test. After finding significant differences inunivariate tests, least significant difference as a post hocanalysis was used to determine significant differences. p <0.05 was considered as statistically significant.

RESULTSEffects of Hyperthermia on Expression of Pax-3,MyoD, Myogenin, and MyHC Genes and Proteins

during Limb Development

Pax-3 gene expression gradually decreased in upperlimbs from the control group and the hyperthermiagroup across ED 11, 15, and 17 embryos and fetuses;however, Pax-3 gene expression was significantly ele-vated in ED 13 upper limbs from hyperthermia-treatedfetuses compared to controls (Fig. 1A, B). Pax 3 proteinexpression mirrored Pax3 gene expression in ED 13 (Fig.1C, D). The level of gene expression of MyoD (Fig. 2A, B)was similar between hyperthermia-treated and controlupper limbs at all stages examined. In contrast, westernblot analysis revealed that the amount of MyoD proteinin upper limbs of fetuses exposed to hyperthermia wassignificantly lower than in the upper limbs of controlfetuses on ED 13, 15, and 17 (Fig. 2C, D). Likewise, thepattern of levels of gene expression of myogenin (Fig. 3A,B) was similar between treated and control groups. Theamount of myogenin protein in heat treated fetuses wassignificantly decreased on ED 11, 13, 15, and 17compared with the control group. However, myogenin

Table 1Primer Sequences for Amplification by RT-PCR

Gene UP(u) and Down(d) primer sequences Product size (bp) Annealing(8C)/Cycle

Pax3 u50-AACACTGTGCCCTCAGTGAGTTCT-30 589 55/30d50-TTTGGTGTACAGTGCTCGGAGGAA-30

MyoD u50-AGGACACGACTGCTTTCTTC-30 389 53/27d50-AGCACCGCAGTAGAGAAGTGT-30

Myogenin u50-CTGGGGACCCCTGAGCATTG-30 272 55/27d50-ATCGCGCTCCTCCTGGTTGA-30

MyHc u50-GCAAAGACCCGTGACTTCACCTCTAG-30 150 58/30d50-GCATGTGGAAAAGTGATACGTGG-30

GAPDH u50-GTCGTGGAGTCTACTGGTGT-30 250 55/20d50-CAAAGTTGTCATTGAGAGCA-30

186 LEE ET AL.


protein expression on PD 1 in the hyperthermia groupwas increased compared to control group, which sug-gests that fusion of myoblast (i.e., formation of myotubes)continues into the neonatal period. Finally, the level ofgene expression of MyHC was also similar betweentreated and control groups, whereas MyHC protein wassignificantly lower on ED 17 and PD 1 in upper limbsfrom treated fetuses compared to controls (Fig. 4C, D).

Effects of Hyperthermia on Expression of MyoD,Myogenin, and MyHC Proteins during Limb

Development: Immunohistochemical Localization

Immunohistochemical staining showed patterns similarto those found by western blotting. MyoD immunoposi-tive cells were dispersed within developing muscle. Ingeneral immunostaining for MyoD was stronger in the

Figure 1. RT-PCR (A) and western blotting (C) of Pax3 in the upper limbs of fetuses. The gene and protein of Pax3 are highly expressedin hyperthermia-exposed group on ED 13 compared to control group on ED 13. Densitometric analysis of RT-PCR (B) and western blot-ting (D) are shown. Data are represented as mean 6 SEM (n 5 3). *p < 0.05; **p < 0.001.

Figure 2. RT-PCR (A) and western blotting (C) of MyoD in the upper limbs of fetuses. The MyoD gene expression is similar in upperlimbs from treated and control fetuses on ED 11, 13, 15 and 17. The protein expression of MyoD is significantly decreased in hyperther-mia-exposed group on ED 13, 15, and 17, compared with control group. Densitometric analysis of RT-PCR (B) and western blotting (D)are shown. Data are represented as mean 6 SEM (n 5 3). *p < 0.05; **p < 0.001.



control group compared to the hyperthermia-exposedgroup (Fig. 5). Similarly, myogenin immunostaining wasstronger in the control group on ED 13, 15, and 17 com-pared with the hyperthermia-exposed group. In contrast,immunostaining was stronger in hyperthermia-exposedfetus samples on PD 1 compared with the control group(Fig. 6). MyHC immunostaining was stronger in the con-trol group on ED 17 and PD 1 compared with the hyper-thermia-exposed group (Fig. 7).

Transmission Electron Microscopic Findings

We compared ultrastructural differences between con-trol and hyperthermia-exposed developing muscle onPD 1. In the control group, the nuclei of muscle fiberswere peripherally located. Satellite cells were localizednear the nucleus of muscle cells and covered by a basallamina of muscle cell (Fig. 8A). Muscle fibers were com-posed of well-arranged myofibrils, which showed distinctZ-lines and several bands. Mitochondria, sarcoplasmic

Figure 3. RT-PCR (A) and western blotting (C) of myogenin in upper limbs of fetuses and neonates. The gene expression of myogenin isnot significantly different between hyperthermia-exposed group and control group. However, myogenin protein of treated group is sig-nificantly increased on ED 11, 13, 15, and 17, except for PD 1 (C). Densitometric analysis Western Blots (B, D) are shown. Data are repre-sented as mean 6 SEM (n 5 3). *p < 0.05; **p < 0.001).

Figure 4. RT-PCR (A) and western blotting (C) of myosin heavy chain (MyHC) in the upper limbs of fetuses and neonates. The geneexpression of MyHC is similar between hyperthermia-exposed group and control group. However, MyHc protein of treated group is sig-nificantly decreased on ED 17 and PD 1. Densitometric analysis of RT-PCR (B) and western blotting (D) are shown. Data are representedas mean 6 SEM (n 5 3). *p < 0.05; **p < 0.001.

188 LEE ET AL.


reticulum, and glycogen particles were located betweenmyofibrils (Fig. 8B). Nerve fibers penetrated muscle fibers(Fig. 8C, arrows). In contrast, satellite cells from thehyperthermia-exposed group were also closely located inmuscle fiber but not enveloped. In addition, myofibrils

showed indistinct bands and Z-lines (Fig. 8D), andmitotically-active myoblasts were observed (Fig. 8E).Nonpenetrating nerve fibers were seen on the peripheryof muscle fibers, and the space between myofibrils waswide (Fig. 8F, arrow).

Figure 5. MyoD immunohistochemistry in the upper limbs of control (A, C, E) and hyperthermia-exposed fetuses (B, D, F). The controlgroup shows stronger immunostaining than hyperthermia-exposed group on ED 13, 15, and 17.

Figure 6. Myogenin immunohistochemistry in the upper limbs of control (A, C, E, G) and hyperthermia-exposed fetuses (B, D, F, H).Myogenin immunostaining of control group is stronger than hyperthermia-exposed group in fetal period, however, weaker on PD 1.



DISCUSSION

In skeletal muscle development in developing limbs,individual mesenchymal muscle precursor cells delami-nate from dermomyotome, migrate into the limb, prolif-erate, differentiate, and eventually form individualmuscles. In recent years, many genes and signaling mole-cules have been identified during skeletal muscle devel-opment in vertebrates (Christ and Brand-Saberi, 2002).These genes and signaling molecules are expressed spa-tiotemporally during skeletal myogenesis. During in vitrodifferentiation of fetal myoblasts, MyoD-positive cellswere detected first, followed by the appearance of cellspositive for both MyoD and myogenin and finally by theappearance of differentiated myocytes and myotubesexpressing MyHC.Pax3, the early marker gene for myogenic lineage, is

essential for early myogenesis (i.e., migration of muscleprecursor cells). This gene is expressed initially through-out the dermomyotome and is later confined to thehypaxial muscle lineage (Buscher and Belmonte, 1999).Pax3 begins to be expressed in dermomyotome and thelimb bud on ED 9 and 9.5, respectively, and starts todecrease its expression on ED 11 and 12 when MyoDand myogenin are detected in premuscle mass in mice(Hayashi and Ozawa, 1991; Bober et al., 1994). In Pax3

Figure 7. MyHc immunohistochemistry in the upper limbs ofcontrol (A, C) and hyperthermia-exposed (B, D) fetuses and neo-nates. MyHc of hyperthermia-exposed group on ED 17 and PD 1shows weaker immunostaining than in the control group.

Figure 8. Electron micrographs of developing muscle cells in upper limbs on PD 1: control group (A, B, C) and hyperthermia-exposedgroup (D, E, F). In the control group, the nucleus of muscle fiber (N) is peripherally located. The satellite cell (SC) is enveloped by basallamina of muscle cell (A). Mitochondria and sarcoplasmic reticulum are located between myofibrils (B). Note that nerve fiber (arrows)penetrates muscle fibers (C). Alternatively, in the hyperthermia-exposed group, the satellite cell is not enveloped by basal lamina. Myofi-brils show indistinct bands and Z-line (D). A mitotic figure of myoblast is observed (E) and a nonpenetraing nerve fiber (arrow) is seenon the periphery of the muscle fiber (F).

190 LEE ET AL.


mutant mice, disintegration of the hypaxial dermomyo-tome was observed, and formation, delamination andmigration of MPCs did not occur, and no limb muscledevelopment proceeded (Buckingham, 2001). Using RT-PCR and western blotting, we confirmed that the Pax3gene and protein are highly expressed on ED 11, whichcorresponds to the migrating period for MPCs, and thatthey begin to be down-regulated on ED 13 in upperlimbs of control fetuses. In contrast, in hyperthermia-exposed groups, Pax3 gene and protein are expressed atsignificantly higher levels compared with the controlgroup on ED 13. This result suggests that in the hyper-thermia-exposed group, MPCs are still migrating on ED13 and early myogenesis is delayed.

The other important step in limb skeletal myogenesisis the transition from a Pax3-positive to a myogenic-posi-tive state. When Pax3 is down-regulated, MyoD and Myf5(members of myogenic regulatory factors [MRFs]) are up-regulated (Pourquie et al., 1995). MRFs are subdividedinto primary and secondary MRFs. The primary MRFsMyoD and Myf5 appear to be required for myogenicdetermination, whereas the secondary MRFs myogeninand Mrf4 are required downstream of MyoD and Myf5 asdifferentiation factors. MRFs are not expressed whilehypaxial MPCs are migrating toward limb bud (Ottet al., 1991). Overall, MRFs are responsible for fate deter-mination, differentiation of myoblast, formation of multi-nucleated muscle cell, and transformation of nonmusclecells into muscle cells after completion of migration ofMPCs (Muscat et al., 1995). MyoD and Myf5 are essentialgenes for formation and survival of myoblasts and playdistinct roles during the formation of hypaxial and epax-ial muscles, respectively. MyoD2/2 mice display normalepaxial (paraspinal and intercostals) muscle develop-ment, whereas hypaxial (limb and abdominal wall) mus-cle development is delayed by 2.5 days. In contrast,Myf52/2 embryos exhibit normal muscle developmentin the limbs and branchial arches, and markedly delayeddevelopment of epaxial muscles (Kablar et al., 1997). Inaddition, mice deficient for both MyoD and Myf5 die atbirth because of a complete absence of skeletal myoblastsand muscle. MyoD and Myf5 are upstream of myogeninand Mrf4 (Rudnicki et al., 1993). Myogenin and Mrf4 arerequired in myoblast differentiation and control terminaldifferentiation of myotubes; thus, these factors areexpressed after fusion of myoblasts (Patapoutian et al.,1995). Mice lacking myogenin display a normal number ofmyoblasts, but die at birth because of an absence of myo-fibers (Hasty et al., 1993; Nabeshima et al., 1993). In thepresent study, we tried to detect gene and proteinexpression levels of MyoD and myogenin as myogenicdetermination and differentiation factors by RT-PCR andwestern blotting. Detection of Myf5 was omitted becauseit is already well known that the role of Myf5 is concen-trated on the epaxial muscle development (Kablar et al.,1997). Genes of MyoD and myogenin were detected on ED11, 13, 15, and 17 and PD 1 in both control and hyper-thermia-exposed groups. However, protein expressionwas different between both groups. In the hyperthermia-exposed group, MyoD and myogenin proteins weredown-regulated compared with the control group duringthe latter stages of limb development (ED 13–17). Thisresult suggests that transformation of MPCs into myo-blasts and differentiation of myoblasts were delayed byprevious hyperthermic exposure. In contrast, on PD1,

expression of myogenin protein was higher in the hyper-thermia group than in the control group. This result sug-gests that fusion of myoblasts (i.e., formation of myo-tubes) continued into the neonatal period. Gene expres-sion patterns (MyoD, myogenin and MyHC) did not matchtheir protein expression patterns on every developmen-tal day examined. These results suggest that posttran-scriptional regulation of genes may occur at the RNAprocessing level. Cusalla-De Angelis et al. (1992)reported that, in 10.5-day–postconception embryos,somites expressing myogenin did not cross-react with amonoclonal antibody against myogenin protein. Sanchezand Robbinson (1994) also insisted on the possibility ofposttranscriptional regulation of the myogenin transcriptduring embryogenesis using PCR and in situ analyses;that is, a significant pool of unprocessed myogeninRNA was accumulated in the developing somites ofmouse embryos (10.5 days postconception), and disap-peared with the formation of skeletal muscle at day 14.In this study, we performed RT-PCR and western blot-ting at the same time on a given development day.Genes of MyoD, myogenin and MyHC showed similarexpression patterns between control and hyperthermia-exposed groups; however, their respective proteins wereusually lower in the treated group than in the controlgroup. These results suggest that hyperthermia may notaffect gene expression, but instead the translation toproteins.In normal skeletal muscle development in mice, pri-

mary muscle fibers appear at ED 11 to 14, and second-ary fibers form at the time when nerve innervationbegins to be established at ED 14 to 16 (Ontell andKozeka, 1984). Muscle fibers are more compact andthicker after ED 17 (van Swearingen and Lance-Jones,1995). During this time, MyHC plays an important rolein formation of primary fibers and several bands estab-lishment in myofibrils (Agbulut, 2003). MyHC proteinis the major component of the thick filament and themost abundant protein of the sarcomere, and thus isprimarily responsible for muscle contraction. Duringmyogenesis it is mainly the developmental isoforms(i.e., embryonic and neonatal MyHC); these are gradu-ally replaced by the adult-type isoforms during theearly postnatal period (Schiaffino and Reggiani, 1996).Thus, we tried to examine the expression of MyHC formaturation of skeletal muscle fibers using RT-PCR andwestern blotting. On ED 17 and PD 1, MyHC gene wasdetected on ED 17 and PD 1 in both control and hyper-thermia-exposed groups; however, the expression ofMyHC protein in limbs from the hyperthermia-exposedgroup was lower than in the control group. We werealso able to confirm this western blotting result usingimmunohistochemistry. In addition, our transmissionelectron microscope study demonstrated nonpenetrat-ing peripherally located nerve fibers, dispersed myofi-brils, indistinct bands and Z-lines, and satellite cellsoutside muscle fiber in hyperthermia-exposed muscleon PD 1. These results suggest that maternal hyperther-mia may affect myotube formation and fusion of myo-blasts during the fetal period and may delay final mus-cle differentiation into the neonatal period. Thus, ourresults suggest that maternal hyperthermia delays skel-etal myogenesis in the developing upper limb, in part,by disregulating the expression of key myogenesis-related factors.



REFERENCES

Adrianakis P, Walker DW, Ralph MM, Thorborn GD. 1989. Effect of in-hibiting prostaglandin synthesis in pregnant sheep with 4-aminoanti-pyrine under normothermic and hyperthermic conditions. Am JObstet Gynecol 161:241–247.

Agbulut O, Noirez P, Beaumont F, Butler-Browne G. 2003. Myosin heavychain isoforms in postnatal muscle development of mice. Biol Cell95:399–406.

Bober E, Franz T, Arnold HH, et al. 1994. Pax-3 is required for the devel-opment of limb muscles: a possible role for the migration of dermo-myotomal muscle progenitor cells. Development 120:603–612.

Buckingham M. 2001. Skeletal muscle formation in vertebrates. Curr OpinGenet Dev 11:440–448.

Buckingham M, Bajard L, Chang T, et al. 2003. The formation of skeletalmuscle: from somite to limb. J Anat 202:59–68.

Buscher D, Belmonte JCI. 1999. Muscle development during vertebratelimb outgrowth. Cell Tissue Res 296:131–139.

Chen JC, Goldhamer DJ. 2004. The core enhancer is essential for propertiming of MyoD activation in limb buds and branchial arches. DevBiol 265:502–512.

Christ B, Brand-Saberi B. 2002. Limb muscle development. Int J Dev Biol46:905–914.

Christ B, Ordahl CP. 1995. Early stages of chick somite development.Anat Embryol 191:381–396.

Cusella-De Angelis MG, Lyons G, et al. 1992. MyoD, myogenin independ-ent differentiation of primordial myoblasts in mouse somites. J CellBiol 116:1243–1255.

Edwards MJ, Saunders RD, Shiota K. 2003. Effects of heat on embryosand foetuses. Int J Hyperthermia 19:295–324.

Hasty P, Bradley A, Morris JH, et al. 1993. Muscle deficiency and neona-tal death in mice with a targeted mutation in the myogenin gene. Na-ture 364:501–506.

Hayashi K, Ozawa E. 1991. Vital labelling of somite-derived myogeniccells in the chicken limb bud. Roux’s Arch Dev Biol 200:188–192.

Kablar B, Krastel K, Ying C, et al. 1997. Myo D and Myf-5 differentiallyregulate the development of limb versus trunk skeletal muscle. De-velopment 124:4729–4738.

Kassar-Duchossoy L, Gayraud-Morel B, Gomes D, et al. 2004. Mrf4 deter-mines skeletal muscle identity in Myf5: Myod double-mutant mice.Nature 431:466–471.

Martinez-Frias ML, Garcia Marario MJ, Caldas CF, et al. 2001. Highmaternal fever during gestation and severe congenital limb disrup-tions. Am J Med Genet 98:201–203.

McKoy G, Leger ME, Bacou F, Goldspink G. 1998. Differential expressionof myosin heavy chain mRNA and protein isoforms in four function-ally diverse rabbit skeletal muscles during pre- and postnatal devel-opment. Dev Dyn 211:193–203.

Morishima HO, Glaser B, Niemann WH, James LS. 1975. Increased uter-ine activity and fetal deterioration during maternal hyperthermia.Am J Obstet Gynecol 121:531–538.

Muscat GE, Downes M, Dowhan DH. 1995. Regulation of vertebrate mus-cle differentiation by thyroid hormone: the role of the myoD genefamily. Bioessays 17:211–218.

Nabeshima Y, Hanaoka K, Hayasaka M, et al. 1993. Myogenin gene dis-ruption results in perinatal lethality because of severe muscle defect.Nature 364:532–535.

Olson EC, Brennan TJ, Chakraborty T, et al. 1991. Molecular control ofmyogenesis: antagonism between growth and differentiation. MolCell Biochem 104:7–13.

Ontell M, Kozeka K. 1984. The organogenesis of murine striated muscle:a cytoarchitectural study. Am J Anat 171:133–148.

Ott MO, Bober E, Lyons G, et al. 1991. Early expression of the myogenicregulatory gene, myf5, in precursor cells of skeletal muscle in themouse embryo. Development 111:1097–1107.

Patapoutian A, Yoon JK, Miner JH, et al. 1995. Disruption of the mouseMRF4 gene identifies multiple waves of myogenesis in the myotome.Development 121:3347–3358.

Rawls A, Olson EN. 1997. MyoD meets its marker. Cell 89:5–8.Rudnicki MA, Schnegelsberg PN, Stead RH, et al. 1993. MyoD or Myf5 is

required for the formation of skeletal muscle. Cell 75:1351–1359.Schiaffino S, Reggiani C. 1996. Molecular diversity of myofibrillar pro-

teins: gene regulation and functional significance. Physiol Rev76:371–423.

Tajbakhsh S, Buckingham M. 2000. The birth of muscle progenitor cells inthe mouse: spatiotemporal considerations. Curr Top Dev Biol 48:225–268.

Van Swearingen J, Lance-Jones C. 1995. Slow and fast muscle fibers arepreferentially derived from myoblasts migrating into the chick limbbud at different developmental times. Dev Biol 170:321–337.

Webster WS, Lipson AH, Brown-Woodman PD. 1987. Uterine trauma andlimb defects. Teratology 35:253–260.

Webster WS, Lipson AH, Brown-Woodman PDC, Osborn RA. 1988.Moebius unmasked: pathogenesis of Moebius syndrome in an animalmodel. Teratology 38:199; cited from Martinez-Frias ML, GarciaMarario MJ, Caldas CF, et al 2001. High maternal fever during gesta-tion and severe congenital limb disruptions. Am J Med Genet 98:201–203.

192 LEE ET AL.


effects of maternal hyperthermia on myogenesis-related factors in developing upper limb

Documents