Ultrasound in Med. & Biol., Vol. 40, No. 3, pp. 504–512, 2014Copyright � 2014 World Federation for Ultrasound in Medicine & Biology

Printed in the USA. All rights reserved0301-5629/$ - see front matter

/j.ultrasmedbio.2013.10.013

http://dx.doi.org/10.1016

d Original Contribution

INDUCTION OF SKELETAL MUSCLE DIFFERENTIATION IN VITRO BYTHERAPEUTIC ULTRASOUND

VIVIANE MENDES ABRUNHOSA,*y CAROLINA PONTES SOARES,y ANA CLAUDIA BATISTA POSSIDONIO,y

ANDR�EVICTOR ALVARENGA,* RODRIGO P. B. COSTA-FELIX,* MANOEL LUIS COSTA,y

and CLAUDIA MERMELSTEINy

*Laborat�orio de Ultrassom, Diretoria de Metrologia Cient�ıfica e Industrial (DIMCI), Instituto Nacional de Metrologia,Qualidade e Tecnologia (INMETRO), Rio de Janeiro, RJ, Brazil; and y Instituto de Ciencias Biom�edicas, Universidade Federal

do Rio de Janeiro, Rio de Janeiro, RJ, Brazil

(Received 17 July 2013; revised 11 October 2013; in final form 15 October 2013)

AFederaChagasE-mail

Abstract—Therapeutic ultrasound (TU) has been used for the last 50 y in rehabilitation, including treatment of softtissues. Ultrasound waves can be employed in two different modes of operation, continuous and pulsed, which pro-duce both thermal and non-thermal effects. Despite the large-scale use of TU, there are few scientific studies on itsbiologic effects during skeletal muscle differentiation. To better analyze the cellular effects of TU, we decided tofollow cells in vitro. The main purpose of this study was to evaluate the effects of TU in primary chick myogeniccell cultures using phase contrast opticalmicroscopy and immunofluorescencemicroscopy, followed by image anal-ysis and quantification.Our results indicate that TU can stimulate the differentiation of skeletalmuscle cells in vitro,as measured by the thickness of multinucleated myotubes, the ratio of mononucleated cells to multinucleated cellsand expression of themuscle-specific protein desmin. This study is a first step toward ametrologic and science-basedprotocol for cell treatment under different ultrasound field exposures. (E-mail: mermelstein@ufrj.br) � 2014World Federation for Ultrasound in Medicine & Biology.

Key Words: Therapeutic ultrasound, Skeletal muscle differentiation, Desmin, Myogenesis.

INTRODUCTION

Soft tissue lesions are common in athletic and non-athletic populations and can be caused by direct contu-sions, indirect strain injuries or intensive exercise. Theseinjuries affect muscle function, joint mobility and,thereby, quality of life (Rantanen et al. 1999). One ofthe most recommended treatments for a wide variety ofskeletal muscle tissue injuries is therapeutic ultrasound(TU) (Speed 2001). Therapeutic applications of ultra-sound have been widely used in physiotherapy since the1950s (Shaw and Hodnett 2008; Watson 2008).Ultrasound is a form of mechanical energy with a high-frequency pressure wave that can be transmitted intothe body (Khanna et al. 2008). The interaction of the ul-trasound wavewith the tissue can promote biophysical ef-fects that can be classified as thermal (continuous mode)

ddress correspondence to: Claudia Mermelstein, Universidadel do Rio de Janeiro, Instituto de Ciencias Biom�edicas, Av. CarlosFilho 373, Ilha do Fund~ao, 21941-902 Rio de Janeiro, RJ, Brazil.: mermelstein@ufrj.br

504

or non-thermal (pulsed mode) (Dyson 1987). Thermal ef-fects in vivo are derived from the conversion of ultrasonicenergy into elevation of the tissue temperature between40�C and 45�C for at least 5 min. These effects havebeen related to increased blood flow and flexibility ofcollagen fibers, reduction in muscle spasm and a pro-inflammatory response (Speed 2001). Non-thermal ef-fects are the local vibrations that involve stable acousticstreaming and cavitations in tissue fluids and couldimprove tissue healing (Chan et al. 2010). Both effects,thermal and non-thermal, occur simultaneously duringTU treatment.

The administration of TU involves the choice of fre-quency, mode of operation of the equipment, effective in-tensity levels and duration of treatment. TU is broadlyused in the frequency range of 0.5 to 5.0 MHz, but typi-cally TU equipment can be set only at 1 or 3 MHz.Different frequencies are recommended depending onthe injury and subcutaneous fat tissue of the patient. Fre-quencies of 1 MHz are absorbed primarily by tissues at adepth of 3–5 cm, and frequencies of 3 MHz are absorbedat a depth of 1–2 cm. The choice of mode of operation of

US-induced skeletal muscle differentiation d V. M. ABRUNHOSA et al. 505

the equipment depends on the goal of the treatment.Continuous mode could cause a larger temperature in-crease in the tissue than pulsed mode, with other param-eters being held constant. Although themode of operationis related to temperature in the tissue, the intensity level isalso involved. The effective intensity (W/cm2) is acquiredfrom the quotient of maximum ultrasonic output power,delivered through the patient’s body, and the transducer’seffective radiating area. Dyson (1987) found that eleva-tion of temperature in the tissue occurs in both continuousand pulsed mode. The other parameter involved in TU isthe duration of treatment. There is no consensus on theduration of TU treatment, which can vary between2 and 15 min for treatment of disorders in soft tissues.The amount of energy delivered to the tissue is propor-tional to the duration of treatment (Robertson andBaker 2001).

The studies that have examined the effects of TU insoft tissues reported divergent results with respect to theefficacy of TU treatment. The variability in the efficacyof TU treatment could be caused by the lack of standard-ization in TU protocols and/or by the fact that differentbiologic models were used (Fisher et al. 2003; McBrieret al. 2007; Silveira et al. 2012). Although some of thesestudies were conducted in animal models, most wereperformed using the C2C12 mouse myoblast cell line.To better analyze the effects of TU at the cellular level,we studied primary skeletal muscle cell culture. Primarycultures of skeletal muscle cells are more similar to thein vivo system than is any muscle cell lineage. Thus, ourmain goal was to investigate the relationship betweenthe protocols commonly used for TU treatment formuscle lesion recovery and their effects on skeletalmuscle differentiation. The experiments were performedin primary cultures of chick skeletal muscle cells usingTU equipment widely employed in physiotherapy.

The chick myoblast primary culture is a robustin vitro model of myogenesis. Skeletal myogenesis pro-ceeds in these cell cultures through the following mainsequential stages: myoblast proliferation, cell cycle with-drawal, myoblast alignment, and fusion and their subse-quent differentiation into striated multinucleatedmyotubes. The fact that these steps have been well char-acterized makes this an interesting model for analysis ofthe effects of TU during muscle differentiation.

Regeneration of an injured skeletal muscle involvesthe activation of the proliferation of satellite cells, fol-lowed by their differentiation and fusion into myotubes,which contributes to the functional recovery of the mus-cle tissue. Differentiation of satellite cells recapitulatesthe embryonic myogenesis. The molecular and cellularbasis of satellite cell differentiation can therefore be stud-ied in embryonic myoblast cell culture models. One ofthe aims of our study was to determine whether TU treat-

ment can activate the proliferation and differentiation ofembryonic myoblasts, assuming that this process wouldbe equivalent in muscle injury-activated satellite cells,and thus contribute to the comprehension of TU effectson muscle lesion healing.

Our results indicated that skeletal muscle cell differ-entiation is induced in vitro by therapeutic ultrasound.

METHODS

Antibodies and fluorescent probesThe DNA-binding probe 4,6-diamidino-2-

phenylindole dihydrochloride (DAPI), Alexa Fluor 488-conjugated goat anti-mouse serum and Alexa Fluor546-conjugated goat anti-rabbit serum were purchasedfromMolecular Probes (Eugene, OR, USA). Rabbit poly-clonal anti-desmin (D-8281), mouse monoclonal anti-sarcomeric a-actinin (clone EA-53, A-7811) and mousemonoclonal anti-a-tubulin (clone DM1A, T-9026) werepurchased from Sigma-Aldrich (St. Louis, MO, USA).

Ultrasonic pressure field mapping systemA water bath measuring 1700 3 1000 3 800 mm

from Ultrasound Labs (National Institute of Metrology,Quality and Technology, Inmetro, Rio de Janeiro, Brazil)was used for mapping the ultrasound pressure field of thetransducer. The specified positioning system (Newport,Irvine, CA, USA) used to move the transducer (or hydro-phone) in the water bath allows movements of 300 mmalong the x- and y-axes and 600 mm along the z-axis.The transducer is excited using a 20-cycle burst of sinewave generated by a function generator (AFG 3252, Tek-tronix, Beaverton, OR, USA), and waterborne signals areacquired using an oscilloscope (TDS 3032B, Tektronix).In the mapping procedure, a needle hydrophone is usedwith an active element of 0.2 mm (Precision Acoustics,Dorchester, Dorset, UK). To integrate all system compo-nents, and also to provide a user-friendly interface, virtualsoftware was developed in the LabView platform (Na-tional Instruments, Austin, TX, USA). The software auto-matically performs the raster scan necessary to calculatethe effective radiating area (AER) of the physiotherapeuticultrasound transducer (Alvarenga and Costa-Felix 2009).

Primary myogenic cell cultureThis study using chick embryos was approved by the

Ethics Committee for Animal Care and Use in ScientificResearch of the Federal University of Rio de Janeiroand received Approval No. DAHEICB004. All cell cul-ture reagents were purchased from Invitrogen (S~ao Paulo,Brazil). Primary cultures of myogenic cells were preparedfrom breast muscles of 11-d-old chick embryos(Mermelstein et al. 2005). Cells were plated at an initialdensity of 106 cells/35 mm in culture dishes coated with

506 Ultrasound in Medicine and Biology Volume 40, Number 3, 2014

rat tail collagen. Cells were grown with 2 mL of culturemedium (Minimum Essential Medium supplementedwith 10% horse serum, 0.5% chick embryo extract, 1%L-glutamine, and 1% penicillin-streptomycin) under ahumidified 5% CO2 atmosphere at 37�C. Twenty-four-hour cultures were treated with therapeutic ultrasoundwaves (see Table 1). Sodium bicarbonate (1.4 g/L) wasadded to the culture medium to maintain pH in the rangeof 7.2–7.4. Phenol red, a pH indicator, was added to me-dium to colorimetrically monitor changes in pH in bothuntreated and ultrasound-treated cells. After treatment,cultures were grown for the next 48 h and culture mediumwas changed every day.

Treatment with ultrasound waveAn ultrasound device (Model Sonopulse Compact,

Ibramed, Amparo, Brazil) was used in different modesof operation (continuous or pulsed) at different effectiveintensities (0.5 or 1 W/cm2) for different durations oftreatment (5 or 10 min). The frequency of the TU wasthe same for all treatments (1 MHz). Ultrasound wastransmitted through the bottom of the culture dish, witha coupling gel between the ultrasound transducer andthe dish. The ultrasonic attenuation effects within themedium were disregarded, as they were the samethroughout all cell cultures and different treatments,and the free path is markedly short (about 2 mm).Most likely, reflections on the free surface of the mediumwill result in an identical ultrasonic field inside the me-dium, as all culture dishes were prepared in a similarway with respect to the amount (volume) of medium.Both attenuation and surface reflections were consideredinherent characteristics of the ultrasonic pressure fieldand very similar to each other in all tested cultures.The diameter of the ultrasound transducer was 35 mm(the same as the culture dish). All 24-h cell cultureswere exposed to TU treatment only once. Control cul-tures were subjected to the same experimental proce-dures as the treated cultures, except that the ultrasound

Table 1. Protocols for ultrasound wave treatment*

Mode of operation Intensity (W/cm2) Duration of treatment (min)

Continuous 0.5 5Continuous 0.5 10Continuous 1.0 5Continuous 1.0 10Pulsed 0.5 5Pulsed 0.5 10Pulsed 1.0 5Pulsed 1.0 10

* An ultrasound apparatus was used in different modes of operation(continuous or pulsed) at different effective intensities (0.5 or 1 W/cm2) for different durations of treatment (5 or 10 min). The frequencyof therapeutic ultrasound was the same for all treatments (1 MHz).

equipment was not turned on (sham ultrasound). Thedifferent TU treatments used in cells are described inTable 1. All analyses were performed 48 h after treat-ment (which means that the cells were grown in culturefor a total of 72 h). The ultrasound apparatus was certif-icated according to IEC 61689:2007 before its use in thisstudy. The temperature of the culture medium wasmeasured immediately before and after TU treatment.Temperature was measured using an infrared thermom-eter (Model 62MAX1, Fluke, Beijing, China) directlyon the medium of each dish without the cover lid.

Cell viabilityCell viability was measured with the XTT assay

(Scudiero et al. 1988). XTT (2,3-bis (2-methoxy-4-nitro-5-sulfophenyl)-5-[(phenylamino) carbonyl]-2H-tetrazoliumhydroxide, Sigma-Aldrich, USA) is metabolically reducedin viable cells to a water-soluble formazan product. This re-agent allows direct absorbance readings. XTTwas preparedat 1 mg/mL in pre-warmed (37�C) Minimum EssentialMedium without serum. PMS (phenazine methosulfate,Sigma-Aldrich) was prepared at 5 mM (1.53 mg/mL) inphosphate-buffered solution (PBS). Fresh XTT and PMSwere mixed together in the appropriate concentrations. Fora 0.025 mM PMS-XTT solution, 25 mL of the stock 5 mMPMS was added per 5 mL of XTT (1 mg/mL). One hundredmicroliters of this mixture (final concentration, 50 mg XTTand 0.38mg PMS perwell) was added to each of the 96wellson the plate containing 24-h myogenic cell cultures (un-treated and TU-treated). After incubation at 37�C for 24 h,the absorbance at 492 nm was measured with a Sunrise-Basic spectrophotometer (Tecan, Gr€odig, Austria).

Immunofluorescence microscopy and digital imageacquisition

Myogenic cells were rinsed with PBS for 10 min atroom temperature. Cells were then permeabilized with0.5% Triton-X 100 in PBS three times for 10 min. Thesame solution was used for all subsequent washing steps.Cells were incubated with primary antibodies for 1 h at37�C in a humidity chamber. After incubation, cellswere washed for 30 min and incubated with Alexa Fluor488 and/or Alexa Fluor 546-conjugated secondary anti-bodies for 1 h at 37�C in a humidity chamber. After a10-min wash with 0.9% NaCl, DAPI (0.1 mg/mL in0.9% NaCl) was added for 5 min. Cells were mountedin ProLong Gold anti-fade reagent (Molecular Probes)and examined with an Axiovert 100 microscope (CarlZeiss, Jena, Germany) by using filter sets that were selec-tive for each fluorochrome wavelength channel. Imageswere acquired with a DP72 camera (Olympus, Tokyo,Japan). All nuclei labeled with DAPI were counted inmononucleated andmultinucleated desmin-positive cells.

Fig. 1. Cell viability of untreated and therapeutic ultrasound(TU)-treated myogenic cell cultures. Myogenic cells weregrown for 24 h, subjected to sham ultrasound (control) ordifferent TU treatments and grown for the next 24 h. Cellviability was quantified by the XTT assay (Scudiero et al.1988) assay in three independent experiments. Note that cellviability did not change significantly after ultrasound treatment.Treatment parameters are given as continuous (cont) or pulsed

(pul) mode–intensity (W/cm2)–duration (min).

US-induced skeletal muscle differentiation d V. M. ABRUNHOSA et al. 507

Quantification of cultured imagesPhase contrast microscopy images of live cultured

cells were acquired with an Axiovert 100 microscope(Carl Zeiss). Images of 72-h myogenic cultures were ac-quired from several equidistant microscopic fields, repre-sentative of the whole dish for each culture condition.Diameter of myotubes and number of nuclei were ob-tained using the public domain software ImageJ(Schneider et al. 2012). Myotube thickness was measuredin the thicker region of each myotube, for all myotubes ineach microscopic field, in at least 50 different fields foreach experimental condition. The data were collectedfrom three independent experiments.

SDS-PAGE and immunoblottingUntreated and TU-treated myogenic cultures were

grown for 72 h. Cultures were then quickly washed inice-cold PBS. Fifty microliters of ice-cold samplebuffer (4% sodium dodecyl sulfate [SDS], 20% glyc-erol, 0.2 M dithiothreitol, 125 mM Tris-HCl, pH 6.8)was added to the cultures, and then cells were scrapedoff the dish with a plastic cell scraper. Cell extractswere recovered in a tube, boiled for 5 min at 95�Cand centrifuged at 12,000g for 15 min. The pellet wasdiscarded and the supernatant was used in subsequentsteps. The amount of protein in each sample was deter-mined according to the Bradford (1976) method, usingbovine serum albumin as a standard. Equal amountsof protein were loaded on 10% SDS-polyacrylamidegel electrophoresis (SDS-PAGE). Proteins were thentransferred to poly (vinyl difluoride) membranes. Theproteins immobilized on the membranes were immedi-ately blocked for 1 h at room temperature with 5%non-fat dry milk in Tris-buffered saline-Tween 20 solu-tion (0.001%) (TBS-T). Then the membranes wereincubated for 12 h at 37�C with an anti-desmin poly-clonal antibody. After five washes in TBS-T (3 mineach), the membranes were incubated for 1 h at 37�Cwith an anti-rabbit peroxidase-conjugated antibody(dilution 1:10,000 in TBS-T; Amersham, GE Health-care, Pittsburgh, PA, USA) and washed again asdescribed above; the bands were visualized using theECL Plus Western Blotting Detection System (Amer-sham/GE Healthcare). To check sample loading, othermembranes with the same samples were incubatedwith a mouse monoclonal anti-a-tubulin antibody(Sigma-Aldrich, dilution 1:3000 in TBS-T-milk). Afterfive washes in TBS-T (3 min each), membranes wereincubated with anti-mouse peroxidase-conjugated anti-body (Amersham, dilution 1:10,000 in TBS-T) anddeveloped as described above. Quantification of proteinbands was performed using the public domain softwareImageJ (Schneider et al. 2012) with data obtained fromthree independent experiments.

Statistical analysisData are presented as the mean 6 standard devia-

tion. Mean values were calculated from at least three in-dependent experiments. Tests of statistical significancewere performed using IBM SPSS software. Statisticalanalysis of the data related to myotube thickness was per-formed by one-way analysis of variance. An unpairedt-test was used for the data related to the nucleus quanti-fication. Statistical comparisons of the immunoblot datawere performed with Student’s t-test. The values wereconsidered to be statistically different at p-values# 0.05.

RESULTS

Therapeutic ultrasound transducer reveals anon-homogeneous pressure field

First, we established the effectiveness of the ultra-sound system used by evaluating the ultrasound pressurefield of the transducer by mapping the ultrasound fieldin an acoustic tank. The transducer is non-focused; thatis, it has a natural focus. Its last maximum of ultrasonicpressure amplitude was measured at 109.5 6 2.7 mmfrom its surface. The transducer’s mapping disclosednon-homogeneity over its output surface, but the non-uniformity ratio of the ultrasonic beam was less than 8.It leads to non-uniform output, but without marked hotspots. Moreover, the medium depth is short comparedwith the transducer diameter (2 and 35 mm, respectively),which results in near-field behavior throughout the me-dium. In this case, diffraction interference is not expectedto prevail, despite the multiple reflections and attenuationthat might occur. The average value of AER for the physio-therapy transducer used in this study was 2.43 cm2, withan estimated expanded uncertainty of 2.9 3 10 cm2

Fig. 2. Phase contrast microscopy analysis of untreated and ther-apeutic ultrasound (TU)-treated myogenic cell cultures.Myogenic cells were grown for 24 h, subjected to sham ultra-sound (control) or different TU treatments and grown for thenext 48 h. Experimental conditions are detailed in Table 1. Onemyotube in each image is in green. The yellow line in the imageof cells treated with continuousmode ultrasound at 0.5W/cm2 for5min indicates howmyotube thickness was measured. Treatmentparameters are given as continuous (cont) or pulsed (pul) mode–

intensity (W/cm2)–duration (min). Bar 5 50 mm.

Fig. 3. Mean thickness of myotubes in untreated and therapeu-tic ultrasound (TU)-treated myogenic cell cultures. Myogeniccells were grown for 24 h, subjected to sham ultrasound (con-trol) or different TU treatments and grown for the next 48 h.Myotube thickness was measured in at least 50 different micro-scopic fields for each experimental condition from three inde-pendent experiments. Note that some TU treatments inducedsignificant enhancement in the thickness of myotubes.*p , 0.05; one-way analysis of variance, compared with con-trol, n 5 3. Treatment parameters are given as continuous(cont) or pulsed (pul) mode–intensity (W/cm2)–duration (min).

508 Ultrasound in Medicine and Biology Volume 40, Number 3, 2014

(11.9%). To understand the relationship between exposureconditions and any observed biologic effects, a guidelinewas proposed by ter Haar et al. (2011). According to this

guideline, this study was framed in level 1, because theresults were referred to the physiotherapy system’s nomi-nal ultrasonic intensity instead of the actual acousticoutput.

TU does not affect cell viability in myogenic culturesNext, we tested the effects of the continuous and

pulsed modes of operation, two different intensities (0.5and 1.0 W/cm2) and two different durations of treatment(5 and 10 min) in a primary culture of chick embryonicskeletal muscle cells. The treatments tested here werechosen by taking into consideration the protocols mostcommonly used in clinical practice. To determinewhether TU treatment has any effect on cell viability, ex-periments were performed with untreated and TU-treatedmyogenic cultures grown 24 h after TU exposure, using aXTT-based method (Scudiero et al. 1988). No significanteffect on cell viability was noted in the myogenic cells24 h after TU treatment compared with untreated cells(Fig. 1). We also measured the temperature of the culturemedium after TU treatment and did not observe an in-crease in temperature.

TU induces differentiation of embryonic skeletal musclecells

Phase contrast microscopy revealed that some TUtreatments enhanced muscle differentiation, as suggestedby an increase in myotube thickness (Fig. 2). TU

Fig. 4. a-Actinin distribution in untreated and therapeutic ultrasound (TU)-treated myogenic cell cultures. Myogenic cellswere grown for 24 h, subjected to sham ultrasound (a, c) or continuous ultrasoundmode at 0.5W/cm2 for 5min (cont–0.5–5)and grown for the next 48 h (b, d). Cells were fixed and double-stained with an anti-sarcomeric a-actinin antibody (green,a–d), and the nuclear dye 4,6-diamidino-2-phenylindole dihydrochloride (DAPI) (blue, a and b). A merged image is pro-vided in (a) and (b). Note that cultures treated with cont–0.5–5 (d) had thicker myotubes than untreated cells (c). a-Actinin(in green) in localized in the Z-bands of striated myofibrils in both untreated and ultrasound-treated myotubes. Myotube

thickness is represented by white line in (d). Bar 5 100 mm (a, b) and 20 mm (c, d).

US-induced skeletal muscle differentiation d V. M. ABRUNHOSA et al. 509

treatment in continuous mode at 0.5 W/cm2 for 5 or10 min (cont–0.5–5 and cont–0.5–10), TU in continuousmode at 1.0 W/cm2 for 5 min (cont–1.0–5) and TU inpulsed mode at 1.0 W/cm2 for 5 min (pul–1.0–5) differedstatistically significantly (95% confidence interval), asmeasured by myotube thickness, compared with no treat-ment (Fig. 3).

To confirm whether or not the cont–0.5–5 treatmentenhanced muscle differentiation, we analyzed the expres-sion and distribution of the muscle-specific marker a-ac-tinin. We decided to test only the cont–0.5–5 treatmentbecause our previous results (Figs. 2 and 3) had indicatedthat this TU treatment promoted major changes in myo-tube thickness. a-Actinin is a myofibrillar protein that lo-calizes in Z-bands in sarcomeres. Immunofluorescenceimages revealed the presence of a-actinin in striatedmyo-fibrils in multinucleated myotubes in both untreated andultrasound-treated cultures (Fig. 4). Cell cultures sub-jected to cont–0.5–5 displayed more and thicker myo-tubes, confirming the enhancement in skeletal muscledifferentiation. It is possible to observe a larger numberof nuclei (labeled with the nuclear dye DAPI) in theTU-treated condition (Fig. 4a, b), which could be inter-

preted as an increase in cell proliferation after TU expo-sure. These results indicated once again that cont–0.5–5induced the formation of thicker myotubes comparedwith no treatment.

We also quantified the number of nuclei in mononu-cleated and multinucleated (myotubes) cells in untreatedcultures and cultures treated with cont–0.5–5. Our resultsindicate that cont–0.5–5 induced a 30% increase in the to-tal number of nuclei per field and a 60% increase in thetotal number of nuclei per myotube (Fig. 5). Note alsothat the proportion of nuclei in mononucleated cells,compared with nuclei in myotubes, is even higher inTU-treated cultures (from 1.5 to 3.1). The increase inthe total number of cells in TU-treated cultures couldindicate enhancement of cell proliferation and/or sur-vival. The increase in the ratio of mononucleated cellsto multinucleated cells suggests enhancement of cellfusion.

Next we decided to quantify the amount of desminprotein expression in untreated cells and cells treatedwith cont–0.5–5. Cells were grown for 24 h, treated withTU and grown for the next 48 h. Untreated and TU-treated cell extracts were analyzed by 10% SDS-PAGE

Fig. 5. Average number of nuclei in untreated and therapeuticultrasound (TU)-treated myogenic cell cultures. Myogeniccells were grown for 24 h, subjected to sham ultrasound (con-trol) or continuous ultrasound mode at 0.5 W/cm2 for 5 min(cont–0.5–5) and grown for the next 48 h. The nuclei presentin mononucleated cells and multinucleated cells (myotubes)were quantified in at least 50 different microscopic fields foreach experimental condition from three independent experi-ments. Note that cont–0.5–5 induced a 30% increase in the totalnumber of nuclei per field and a 60% increase in the total num-ber of nuclei per myotube. *p, 0.05, unpaired t-test, compared

with control, n 5 3.

Fig. 6. Therapeutic ultrasound (TU) enhances the expressionof desmin in myogenic cell cultures. (a) Total cell extractsfrom primary cultures of chicken myogenic cells were analyzedby Western blotting using an anti-desmin antibody. Cellswere grown for 24 h, subjected to sham ultrasound (control)or continuous ultrasound mode at 0.5 W/cm2 for 5 min(cont–0.5–5) and grown for the next 48 h. (b) Quantificationof immunoblots revealed a 60% increase in the levels of desminexpression (52 kDa) in cells after TU treatment compared withcontrol cells. The semi-quantitative analysis represents theaverage of three independent experiments. *p, 0.05, Student’s

t-test, compared with control, n 5 3.

510 Ultrasound in Medicine and Biology Volume 40, Number 3, 2014

followed by immunoblotting. Quantification of immuno-blots revealed a 60% increase in the expression of desmin(52 kDa) in TU-treated cells compared with control cells(Fig. 6).

DISCUSSION

In the work described here, we studied the effects ofcommonly used TU treatment protocols for muscle lesionrecovery in chick skeletal muscle cell culture to deter-mine the relationship between TU protocols and theireffects on muscle differentiation. We found that severalof these protocols, most of them in continuous mode,enhance skeletal muscle fiber formation. Muscle differ-entiation was reflected in the presence of more myotubeswith more well-organized striated myofibrils. The extentof differentiation was analyzed by measurement of thethickness of multinucleated myotubes and the ratio ofmononucleated cells to multinucleated cells.

These findings argue that the chick in vitromodel ofmyoblast cell culture is a valuable tool for studies of thecellular andmolecular effects of TU duringmuscle differ-entiation. In vitromodels such as primary chick myogenicculture begin with a population of replicating mononucle-ated myoblasts and some fibroblastic cells (Holtzer et al.1991). After a few rounds of replication, these myoblastswithdraw from the cell cycle and begin to change in

shape, becoming elongated and bipolar cells. Bipolarmyoblasts then enter a cell-cell recognition phase inwhich their plasma membranes adhere and fuse. Fusiongives rise to multinucleated and striated myotubes.

The enhancement in thickness of myotubes andnumber of nuclei per myotube after TU treatment couldbe caused by an increase in myoblast fusion. Myoblastfusion can be enhanced in different ways, such as an in-crease in membrane fluidity and/or permeability, anincrease in myoblast proliferation (which occurs beforefusion) and exposure or removal of membrane moleculesthat are necessary for the fusion process. It has been pro-posed that TU treatment induces changes in cell mem-brane permeability, a property that is widely used forimproving gene transfer protocols (Li et al. 2012).

Few studies have examined the effect of TU in thetreatment of skeletal muscle (Fisher et al. 2003;McBrier et al. 2007; Plentz et al. 2008), and none ofthem found significant differences using continuousmode treatment. Our results contradict those articles,because we found that all continuous mode TUtreatments, with the exception of cont–1.0–10, induced

US-induced skeletal muscle differentiation d V. M. ABRUNHOSA et al. 511

enhancement of myotube thickness. Several explanationsfor these contradictory findings can be hypothesized. Onepossible explanation might be the difference betweentheir in vivo approach and our in vitro approach. Fewervariables are involved in myogenesis in skeletal musclecell cultures than in animal myogenesis. In an animal,there are hundreds of cell types interacting with eachother and with the extracellular matrix, through theexchange of chemical signals in a 3-D environment. Incontrast, cell cultures respond to a reduced range ofstimuli in a controlled environment. Therefore, cellculture models are usually less variable and providehigher reproducibility than animal models. In addition,the results obtained in cell cultures are produceddirectly by the cells and their controlled environment,whereas in animals, the results could arise fromcomplex interactions between tissues and/or systems.

Other possible explanations for the contradictoryfindings between our results and those of other authorsare: (i) differences between species (birds vs. mammals);(ii) differences between a muscle damage recovery modeland our normal myogenesis model; and (iii) differencesin TU conditions used, such as, frequency, intensity ofthe equipment and duration of treatment, as suggestedby Schuster et al. (2013).

The enhancement in myogenesis we observed isprobably not related to the thermal effect of the contin-uous or pulsed mode condition used, as no increase intemperature was detected in the myogenic cell culturesafter TU treatments.

Most of the cell cultures we treated with pulsed TUdid not exhibit an increase in myotube thickness, with theexception of pul–1.0–5, whereas myotube thicknessincreased in most continuous TU cultures. This findingmay be explained by the fact that the energy deliveredto the culture by pulsed mode at 1.0 W/cm2 is the sameas that delivered by continuous mode at 0.5W/cm2 duringthe 5min in which we observed positive effects onmusclethickness. Interestingly, we could not observe any corre-lation between the overall energy delivered by TU to cellsand their differentiation. Our results are in agreementwith those of Fisher et al. (2003), who reported a signif-icant increase in synthesis of contractile proteins ininjured gastrocnemius of male Sprague-Dawley rats aftertreatment with the same of mode of operation (pulsed),intensity (1.0 W/cm2) and duration of treatment (5 min)used in our study. Studies using TU conditions similarto those described above have suggested that these resultsare relative to an increase in satellite cell proliferation(Rantanen et al. 1999).

In contrast to our results, Ikeda et al. (2006) reportedthat low-intensity pulsed ultrasound (LIPUS, 1.5 MHz,70 mW/cm2 for 20 min) induces a significant decreasein expression of the myogenic transcription factor

MyoD in the myoblast cell line C2C12. In agreementwith our data, Nagata et al. (2013) described that treat-ment of C2C12 cells with TU (LIPUS, 3 MHz, 30 mW/cm2 for 15 min) up-regulated expression of the myogenictranscription factor myogenin, and treatment of mousemuscle with TU induced an increase in the cross-sectional area of myofibers. Park et al. (2010) reportedthat C2C12 cells subjected to LIPUS (30 mW/cm2 for3 or 10 min daily for 6 d) exhibited a significant increasein expression of tropomyosin. Thus, we can concludefrom the results presented here with primary cultures ofchick myoblasts and from the data previously describedby other groups with the C2C12 cell line (Nagata et al.2013; Park et al. 2010) that some TU treatments canpromote myogenesis and, thus, can be a valuable tool intreatment of muscle injuries. It is difficult to drawfurther conclusions from these studies because ofdifferences in the test systems.

Furthermore, inadequate calibration of TU equip-ment can also interfere with treatment. It has been re-ported that performance of TU equipment that is morethan 10 y old and equipped with dual-frequency treatmentheads can be compromised (Speed 2001). To avoid theseproblems, we sought to certify our TU equipment.

We acknowledge the differences between treatingmuscle cells in vitro with TU and treating human muscletissues in vivo. In addition to the fact that cell cultureslacks the complexity of muscle tissues, as mentionedabove, there is also the problem of artificial interactionsbetween the acoustic field and the plastic culture dish.In the present work, we used a coupling gel betweenthe TU transducer and the plastic culture dish to minimizethese unwanted interactions. On the other hand, as weillustrated the direct effects of TU on cell differentiation,cell cultures can be used as an important strategy in study-ing the biologic mechanisms of TU action.

CONCLUSIONS

We have provided novel evidence indicating that TUenhances the differentiation of skeletal muscle cells.However, more research is needed to explore the molec-ular and cellular basis of the observed TU effects duringskeletal muscle differentiation. These results could havean impact on the development of new strategies and pro-tocols for practical TU treatments.

Acknowledgments—This work was supported by Brazilian grants fromthe Conselho Nacional de Desenvolvimento Cient�ıfico e Tecnol�ogico(CNPq), Programa de Capacitac~ao Cient�ıfica e Industrial do Inmetro-CNPq-PROMETRO No. 563089/2010-5, the Fundac~ao Carlos ChagasFilho de Apoio �a Pesquisa do Estado do Rio de Janeiro (FAPERJ) andthe Programas de Apoio aos N�ucleos de Excelencia (Pronex). We thankJuliana Lourenco Abrantes for technical assistance.

512 Ultrasound in Medicine and Biology Volume 40, Number 3, 2014

REFERENCES

Alvarenga AV, Costa-Felix R. Uncertainty assessment of effective radiationarea and beam non-uniformity ratio of ultrasound transducers deter-mined according to IEC 61689:2007. Metrologia 2009;46:367–374.

Bradford MM. A rapid and sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72:248–254.

Chan YS, Hsu KY, Kuo CH, Lee SD, Chen SC, ChenWJ, Ueng SW. Us-ing low-intensity pulsed ultrasound to improve muscle healing afterlaceration injury: An in vitro and in vivo study. Ultrasound Med Biol2010;36:743–751.

Dyson M. Mechanisms involved in therapeutic ultrasound. Physio-therapy 1987;73:116–120.

Fisher BD, Hiller CM, Rennier SGA. A comparison of continuous ultra-sound and pulsed ultrasound on soft tissue injury markers in the rat.J Phys Ther Sci 2003;15:65–70.

Holtzer H, Dillulo C, Costa ML, Lu M, Choi J, Mermelstein CS,Scultheiss T, Holtzer S. Striated myoblasts and multinucleated my-otubes induced in non-muscle cells by MyoD are similar to normalin vivo and in vitro counterparts. In: Ozawa E, Masaki T,Nabeshima Y, (eds). Frontiers in muscle research. Amsterdam:Elsevier; 1991. p. 187–207.

Ikeda K, Takayama T, Suzuki N, Shimada K, Otsuka K, Ito K. Effects oflow-intensity pulsed ultrasound on the differentiation of C2C12cells. Life Sci 2006;79:1936–1943.

Khanna A, Nelmes RT, Gougoulias N, Maffulli N, Gray J. The effects ofLIPUS on soft-tissue healing: A review of literature. Br Med Bull2008;89:169–182.

Li Y, Wang J, Satterle A, Wu Q, Wang J, Liu F. Gene transfer to skeletalmuscle by site-specific delivery of electroporation and ultrasound.Biochem Biophys Res Commun 2012;424:203–207.

McBrier NM, Lekan JM, Druhan LJ, Devor ST, MerrickMA. Therapeu-tic ultrasound decreases mechano-growth factor messenger ribonu-cleic acid expression after muscle contusion injury. Arch PhysMed Rehabil 2007;88:936–940.

Mermelstein CS, Portilho DM, Medeiros RB, Matos AR,Einicker-Lamas M, Tortelote GG, Vieyra A, Costa ML. Cholesteroldepletion by methyl-b-cyclodextrin enhances myoblast fusion and

induces the formation of myotubes with disorganized nuclei. CellTissue Res 2005;319:289–297.

Nagata K, Nakamura T, Fujihara S, Tanaka E. Ultrasound modulates theinflammatory response and promotes muscle regeneration in injuredmuscles. Ann Biomed Eng 2013;41:1095–1105.

Park H, Yip MC, Chertok B, Kost J, Kobler JB, Langer R, Zeitels SM.Indirect low-intensity ultrasonic stimulation for tissue engineering.J Tissue Eng 2010;1:1–9.

Plentz RDM, Stoffel PB, Kolling GJ. Hematological changes producedby 1 MHz continuous ultrasound, applied during the acute phase ofiatrogenic muscle injury in rats. Rev Bras Fisioter 2008;12:495–501.

Rantanen J, ThorssonO,Wollmer P, Hurme T, KalimoH. Effects of ther-apeutic ultrasound on the regeneration of skeletal myofibers afterexperimental muscle injury. Am J Sports Med 1999;27:54–59.

Robertson VJ, Baker KG. A review of therapeutic ultrasound: Effective-ness studies. Phys Ther 2001;81:1339–1350.

Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25years of image analysis. Nat Methods 2012;9:671–675.

Schuster A, Schwab T, Bischof M, Klotz M, Lemor R, Degel C,Sch€afer KH. Cell specific ultrasound effects are dose and frequencydependent. Ann Anat 2013;195:57–67.

Scudiero DA, Shoemaker RH, Paull KD, Monks A, Tierney S,Nofziger TH, Currens MJ, Seniff D, Boyd MR. Evaluation of a sol-uble tetrazolium/formazan assay for cell growth and drug sensitivityin culture using human and other tumor cell lines. Cancer Res 1988;48:4827–4833.

ShawA, HodnettM. Calibration andmeasurement issues for therapeuticultrasound. Ultrasonics 2008;48:234–252.

Silveira PC, da Silva LA, Tromm PT, Scheffer Dda L, de Souza CT,Pinho RA. Effects of therapeutic pulsed ultrasound and dimethyl-sulfoxide phonophoresis on oxidative stress parameters after injuryinduced by eccentric exercise. Ultrasonics 2012;52:650–654.

Speed CA. Therapeutic ultrasound in soft tissue lesions. Rheumatology2001;40:1331–1336.

Ter Haar G, Shaw A, Pye S, Ward B, Bottomley F, Nolan R, Coady AM.Guidance on reporting ultrasound exposure conditions for bio-effects studies. Ultrasound Med Biol 2011;37:177–183.

Watson T. Ultrasound in contemporary physiotherapy practice.Ultrasonics 2008;48:321–329.