Ei

EGAMa

1b

1

a

ARRAA

KETGCM

1

date(dsb

C5

h0

Tissue and Cell 48 (2016) 224–234

Contents lists available at ScienceDirect

Tissue and Cell

jo ur nal ho me p a ge: www.elsev ier .com/ locate / t i ce

ffects of microcurrent therapy on excisional elastic cartilage defectsn young rats

dson Pereira Tangerino Filhoa, José Luis Fachia, Israel Costa Vasconcelosa,laucia Maria Tech dos Santosa, Fernanda Aparecida Sampaio Mendonc aa,ndrea Aparecida de Aroa, Edson Rosa Pimentelb,arcelo Augusto Marretto Esquisattoa,∗

Programa de Pós-graduac ão em Ciências Biomédicas, Centro Universitário Hermínio Ometto, Av. Dr. Maximiliano Baruto, 500 Jd. Universitário,3607-339 Araras, SP, BrazilDepartamento de Biologia Estrutural e Funcional, Instituto de Biologia, Universidade Estadual de Campinas, Rua Charles Darwin, s/n. CxP 6109,3083-863 Campinas, SP, Brazil

r t i c l e i n f o

rticle history:eceived 2 August 2015eceived in revised form 5 March 2016ccepted 6 March 2016vailable online 9 March 2016

eywords:lectrotherapyissue repairlycosaminoglycansollagenMP

a b s t r a c t

The effects of microcurrent application on the elastic cartilage defects in the outer ear of young animalswere analyzed. Sixty male Wistar rats were divided into a control (CG) and a treated group (TG). Anexcisional lesion was created in the right outer ear of each animal. Daily treatment was started after 24 hand consisted of the application of a low-intensity (20 �A) continuous electrical current to the site ofinjury for 5 min. The animals were euthanized after 7, 14 and 28 days of injury and the samples weresubmitted to analyses. In CG, areas of newly formed cartilage and intense basophilia were seen at 28days, while in TG the same observations were made already at 14 days. The percentage of birefringentcollagen fibers was higher in CG at 28 days. The number of connective tissue cells and granulocytes wassignificantly higher in TG. Ultrastructural analysis revealed the presence of chondrocytes in TG at 14days, while these cells were observed in CG only at 28 days. Cuprolinic blue staining and the amount ofglycosaminoglycans were significantly higher in TG at 14 days and 28 days. The amount of hydroxyproline

was significantly higher in TG at all time points studied. The active isoform of MMP-2 was higher activityin TG at 14 days. Immunoblotting for type II collagen and decorin was positive in both groups and atall time points. The treatment stimulated the proliferation and differentiation of connective tissue cells,the deposition of glycosaminoglycans and collagen, and the structural reorganization of these elementsduring elastic cartilage repair.

. Introduction

Elastic cartilage is a highly specialized connective tissue that iserived from the embryonic mesenchyme. The tissue is avascularnd contains nerve endings in its stroma. The structural organiza-ion of elastic cartilage is characterized by a small number of cellularlements, chondrocytes, and large amounts of extracellular matrixECM) rich in elastin. The chondrocytes are responsible for the pro-

uction, organization and renewal of ECM macromolecules thaturround them, permitting a strong morphofunctional interactionetween these components and with the ECM (Negri et al., 2007).

∗ Corresponding author at: Programa de Pós-graduac ão em Ciências Biomédicas,entro Universitário Hermínio Ometto—FHO|Uniararas, Av. Dr. Maximiliano Baruto,00 Jd. Universitário, 13607-339 Araras, SP, Brazil.

E-mail address: marcelosquisatto@uniararas.br (M.A.M. Esquisatto).

ttp://dx.doi.org/10.1016/j.tice.2016.03.004040-8166/© 2016 Elsevier Ltd. All rights reserved.

© 2016 Elsevier Ltd. All rights reserved.

Furthermore, the chondrocytes found in elastic cartilage are nour-ished by the diffusion of metabolites through the ECM by means ofcapillaries in the perichondrium (Ito et al., 2001).

The composition of cartilaginous ECM includes fibrillar ele-ments such as collagen and elastin. The main types of collagenfound in cartilage are types II, IX and XI, with type II being the mostabundant. The function of collagen fibrils and fibers in the matrixis to withstand tensile forces and to sustain stromal organiza-tion (Myllyharju and Kivirikko, 2001). Proteoglycans (PGs) consistof a central protein covalently bound to extensive and differentpolysaccharide chains, called glycosaminoglycans (GAGs) and canbe classified according to molecular weight or the type of associatedGAG (Vogel, 1994; Iozzo, 1998). Elastic fibers consist of an elastin

core surrounded by a mantle of microfibrils rich in molecules calledfibrillins. The presence of elastic fibers in the ECM of cartilageincreases tensile strength and tissue elasticity (Kielty et al., 2002).

E.P. Tangerino Filho et al. / Tissue and Cell 48 (2016) 224–234 225

F ated inp he tre( r intert

EooeCiytotoe

ipcpae

tm(coisaa(

ttl

ig. 1. Photomicrographs of longitudinal sections of the elastic cartilage defect creH 4. Control group (a–c). Group treated with a microcurrent (20 �A/5 min) (d–f). T*) Blood vessels; (→) connective tissue cells; (�) cartilage areas. Bar = 100 �m. (Fohe web version of this article.)

The organization, composition and concentration of the mainCM components are intimately related to the functional propertiesf the tissue (Brighton et al., 2008). The action of mechanical forcesn tissues, as well as the natural process of aging, directly influ-nces the organization of elastic cartilage (Hyttinen et al., 2001;arrington, 2005; Hennerbichler et al., 2008). Experimental stud-

es have shown that small-diameter cartilage defects induced inoung animals trigger cartilage repair characterized by the forma-ion of new tissue with characteristics and properties similar to theriginal tissue. This fact is attributed to the proliferation and syn-hesis capacity of ECM promoted by the accelerated metabolismf chondrocytes (Newman, 1998; Nixon and Fortier, 2001; Anrakut al., 2009; de Campos Ciccone et al., 2013).

In view of the impact on public health investments, most stud-es on cartilage repair use models of osteochondral injuries thatermit to monitor the incorporation of articular cartilage underonditions of compressive stress. However, these models do notermit to evaluate the characteristics of the newly formed tissuend its incorporation into the surrounding preserved tissue (Khant al., 2008).

The tissue repair responses of non-articular cartilage differ fromhose of articular cartilage since the damage caused to the for-

er are chondral and not osteochondral injuries as in the latterMoyer et al., 2010; Rajnoch et al., 2003). Furthermore, chondro-ytes isolated from non-articular cartilages contain larger amountsf lipid inclusions and glycogen due to the slower metabolismn these tissues (Stockwell, 1967; Souza et al., 2001). Finally, thetructure and size of the keratan sulfate chains present in high-nd low-molecular weight PGs isolated from cartilage ECM differccording to anatomical site and functional demand of the tissueNieduszynski et al., 1990).

Recent studies have opened various treatment options for car-ilage disorders designed to improve the quality of the repairedissue. These options include electrical and electromagnetic stimu-ation and autologous grafts of chondrocytes, mesenchymal cells or

the ear of rats. The sections were stained with Toluidine blue in McIlvaine buffer,atment and observation periods were 7 (a and d), 14 (b and e) and 28 (c and f) days.pretation of the references to colour in this figure legend, the reader is referred to

biocompatible tissues derived from the periosteum and perichon-drium which exhibit a great chondrogenic potential (Hennerbichleret al., 2008; Khubutiva et al., 2008). However, in contrast to electri-cal stimulation, the other techniques require invasive proceduresfor the implantation of these cells and tissues at the site of injury(Johnson et al., 2004; Anraku et al., 2009).

In addition to being a noninvasive and low-risk option, the useof electrical stimuli induced by low-intensity electrical currents hasbeen shown to be an effective treatment for cartilage repair. How-ever, there are only few studies in the literature proposing protocolsand the type of electrical current to be used in order to promote theregeneration of elastic cartilage in mammals (Snyder et al., 2002;Haddad et al., 2007).

Therefore, the objective of the present study was to evaluate thestructural, ultrastructural and biochemical alterations that occurduring the repair of excisional elastic cartilage defects in the outerear of young rats after microcurrent application.

2. Materials and methods

All animal procedures described here were approved by theEthics Committee on Animal Use of FHO, Uniararas (Protocol No.023/2013).

2.1. Experimental groups

Sixty male Wistar rats, 45 days old and weighing on average180 g, were obtained from the Luis Edmundo de Magalhães Centerof Animal Experimentation, Centro Universitário Hermínio Ometto(FHO), Uniararas. The animals were maintained in individual cages

and received commercial chow and water ad libitum. In view oftheir genetic similarity, the animals were divided into two groupsof 30 animals each: a control group (CG) and a group treated witha microcurrent (TG).

226 E.P. Tangerino Filho et al. / Tissue and Cell 48 (2016) 224–234

Fig. 2. Photomicrographs of longitudinal sections of the elastic cartilage defect created in the ear of rats. The sections were stained with picrosirius-hematoxylin and analyzedb ntrol gg fibers

2

(adwttm2s(

y bright-field microscopy (a–c and g–i) and under polarized light (d–f and j–l). Co and j), 14 (b, e, h and k) and 28 (c, f, I and l) days. (*) Cartilage areas; (→) collagen

.2. Defect and experimental treatment

The animals were anesthetized with xylazine hydrochloride0.2 mL/kg) and ketamine hydrochloride (1 mL/kg), followed byntisepsis of their right ears with povidone-iodine. Next, a cylin-rical defect measuring 2 mm in diameter and 1 mm in thicknessas created with a surgical punch in the region of the antihelix of

he right outer ear of each animal. Treatment was started 24 h afterhe surgical intervention and was applied daily for 28 days. The ani-

als were euthanized with an overdose of the anesthetic 7, 14 and

8 days after injury for removal of the outer ear and processing ofamples for structural (n = 5/time point) and biochemical analysisn = 5/time point). All experimental procedures were performed by

roup: a–f. Treated group: g–l. The treatment and observation periods were 7 (a, d,. Bar = 100 �m.

the same researchers and the instruments were adequately treatedto prevent contamination of the animals.

The animals of TG were treated with a biphasic square-pulsemicrogalvanic continuous electrical current (frequency of 0.3 Hzand intensity of 20 �A/5 min). The pulse duration was 10 s, witha 2-s interpulse interval. During application of the current, twometal electrodes with spherical tips (10 mm) were placed onthe left and right side of the defect for 2.5 min. The positionwas then inverted and the current was applied for an additional2.5 min as described by de Campos Ciccone et al. (2013). Subsen-

sory transcutaneous stimulation, which does not excite peripheralnerves, was applied using a Physiotonus Microcurrent apparatus(Bioset®).

E.P. Tangerino Filho et al. / Tissue and Cell 48 (2016) 224–234 227

Fig. 3. Photomicrographs of longitudinal sections of the elastic cartilage defect created in the ear of rats. The sections were stained with acetic orcein. Control group: a–c.Treated group: d–f. The treatment and observation periods were 7 (a and d), 14 (b and e) and 28 (c and f) days. (→) Elastic fibers; (*) cartilage areas. Bar = 100 �m (a and d)and 50 �m (b, c, e and f).

Table 1Morphometric parameters of the repair area in the different groups and at the different time points.

Parameter Time Control group Treated group

Wound reduction (%) 7 days 2.8 ± 1.3 2.6 ± 1.914 days 33.4 ± 10.6 35.5 ± 8.828 days 57.5 ± 14.8 59.2 ± 15.7

Number of connective tissue cells (n/104 �m2) 7 days 24.4 ± 6.5 36.2 ± 6.8*14 days 48.4 ± 6.9 62.1 ± 8.7*28 days 90.7 ± 10.8 108.2 ± 11.4*

Number of granulocytes (n/104 �m2) 7 days 45.4 ± 6.9 84.8 ± 10.1*14 days 37.4 ± 8.1 54.6 ± 8.1*28 days 18.8 ± 3.7 20.1 ± 4.2

Birefringent collagen fiber area (%) 7 days 48.4 ± 10.4 45.2 ± 8.314 days 50.4 ± 8.3 53.6 ± 7.928 days 89.5 ± 8.4* 68.3 ± 7.6

Collagen fibril diameter (nm) 7 days 38.4 ± 9.9 40.2 ± 10.114 days 42.4 ± 12.7 43.6 ± 12.928 days 42.5 ± 10.5 41.9 ± 9.6

Proteoglycan markings (n/25 �m2) 7 days 3.7 ± 2.1 4.4 ± 1.514 days 4.2 ± 1.5 7.8 ± 2.4*28 days 10.2 ± 2.9 14.7 ± 12.2 *

T 28 (28o ampled

2

afniS(p

he measurements were obtained in samples collected 7 (7 days), 14 (14 days) and

f each animal. Values are expressed as the mean and standard deviation of each sifference compared to the control group.

.3. Macroscopic inspection and quantification of the repair area

The ears were removed and immediately photographed with PinePIX 300 digital camera in the frontal view. Anatomicaleatures of the newly formed tissue compared to the origi-al tissue were evaluated. For quantification of the reduction

n the wound area, the digital images were analyzed using the

igma Scan Pro 5.0TM software. The size of the defect area�m2) was quantified and compared between the different timeoints.

days) days after injury. Three samples were collected from the center of the defect and were compared by the Student t-test (5% level of significance). (*) Significant

2.4. Processing for histological and histochemical analysis

After removal of the outer ear at each time point, the woundfragments were immersed in fixative solution containing 10%formaldehyde in Millonig buffer, pH 7.4, for 24 h at room temper-ature. The specimens were then washed in buffer and submittedto standard procedures for embedding in ParaplastTM (Merck).

Longitudinal sections (6 �m) of the specimens were stained withpicrosirius-hematoxylin for subsequent examination by polarizedlight microscopy and observation of collagen fiber organization;

228 E.P. Tangerino Filho et al. / Tissue and Cell 48 (2016) 224–234

F f the

d and f

waqewm

2

sd(rpspDi

2

wttiwmapsa1tf

ig. 4. Immunohistochemical detection of type II collagen in longitudinal sections o–f. The treatment and observation periods were 7 (a and d), 14 (b and e) and 28 (c

ith Toluidine blue in McIlvaine buffer, pH 4.0, for analysis ofcid glycosaminoglycans (tissue basophilia); Dominici stain for theuantification of granulocytes, and acetic orcein for analysis of thelastic fiber system. The samples were analyzed and documentedith a Leica DM2000 photomicroscope at the Laboratory of Micro-orphology, FHO, Uniararas.

.5. Morphometric analysis

Longitudinal sections stained with Toluidine blue, Dominicitain and picrosirius-hematoxylin were used, respectively, toetermine the total number of fibroblasts and granulocytesn/104 �m2) and the area of birefringent collagen fibers (%) in theepair tissue of the two groups studied and at the different timeoints. Measurements were made in triplicate in each of the fiveections obtained from the mid-sections of each animal selecteder group. All images were captured and digitized using a LeicaM2000 photomicroscope. The measurements were made on dig-

tized images using the Sigma Scan Pro 5.0TM program.

.6. Processing for immunohistochemical analysis

Longitudinal deparaffinized sections (6 �m) were incubatedith 1 mg/mL pronase E in PBS for 15 min at room tempera-

ure. Endogenous peroxidase activity was blocked by immersinghe sections in methanol containing 0.3% (v/v) hydrogen perox-de for 30 min. The sections were then washed in PBS, treated

ith fetal bovine serum for 20 min, and incubated with the pri-ary antibody diluted in 0.01 M PBS containing 1% bovine serum

lbumin for 18 h at 4 ◦C in a moisture chamber. The followingrimary antibodies were used: (1) monoclonal anti-chondroitinulfate raised in mice (diluted 1:250, Santa Cruz, Dallas, Texas, USA)

nd (2) polyclonal anti-collagen type II raised in rabbits (diluted:100, Rockland, Limerick, Pennsylvania, USA). After incubation,he sections were washed three times in PBS and treated with theollowing secondary antibodies for 30 min at room temperature:

elastic cartilage defect created in the ear of rats. Control group: a–c. Treated group:) days. Arrow: positive reaction; (*) chondrocytes. Bar = 100 �m.

(1) peroxidase-conjugated anti-mouse IgG antibody (diluted 1:500,Rockland) and (2) peroxidase-conjugated anti-rabbits IgG antibody(diluted 1:500, Rockland). Finally, the sections were washed in0.05 M Tris-HCl, pH 7.4, and treated with 0.05% (v/v) diaminoben-zidine in 0.05 M Tris-HCl, pH 7.4, containing 0.03% (v/v) hydrogenperoxide for 5 min for the detection of peroxidase activity. The sec-tions were then washed in distilled water, dehydrated, cleared, andmounted in Entelan. Sections treated with 1% bovine serum albu-min in 0.01 M PBS without the addition of the primary antibodiesserved as negative control.

2.7. Processing for ultrastructural and cytochemical analysis

Samples of the lesioned tissue obtained from the differentgroups and at the different time points were fixed in a solutioncontaining 2% glutaraldehyde and 0.1% tannic acid dissolved in0.1 M sodium cacodylate buffer, pH 7.3, for 2 h at room temper-ature. The material was then washed in buffer and postfixed in1% osmium tetroxide for 1 h at 4 ◦C. After this step, the fragmentswere washed in glycated saline and treated with 1% uranyl acetatefor 18 h at 4 ◦C, followed by washing in glycated saline and dehy-dration. Fragments obtained from the same samples were stainedwith cuprolinic blue for ultrastructural detection of PGs (Scottet al., 1989). After fixation and postfixation, the different sampleswere dehydrated in an increasing ethanol series and passed twicethrough propylene oxide. The samples were immersed in mix-tures of propylene oxide/resin Epon (1:1, 1:2 and pure) and, finally,embedded in plastic molds in an oven. Sections were cut with aglass and diamond knife in Leica RM2245 and Ultracut microtomesand counterstained with 2% uranyl acetate in water and 0.2% leadcitrate in 0.1 N NaOH (Reynolds, 1963). The samples were observed

and documented with an JEOL transmission electron microscopeat the Center of Electron Microscopy, IB/UNICAMP, for qualitativeanalysis and determination of collagen fibril diameter (nm) and thenumber of positive PG markings (n/25 �m2).

E.P. Tangerino Filho et al. / Tissue and Cell 48 (2016) 224–234 229

Table 2Biochemical parameters evaluated in the repair area of the different groups and at the different time points.

Parameter Time Control group Treated group

GAGs (�g/mg fresh tissue) 7 days 27.8 ± 6.1 29.9 ± 3.814 days 49.8 ± 7.4 66.3 ± 7.8*28 days 143.5 ± 8.3 167.8 ± 11.7*

Hydroxyproline (�g/mg dry tissue) 7 days 28.4 ± 7.1 44.2 ± 6.8*14 days 47.4 ± 4.9 60.6 ± 5.1*28 days 89.5 ± 8.6 109.8 ± 9.9*

Non-collagen proteins (mg/g fresh tissue) 7 days 110.6 ± 38.1 93.1 ± 35.714 days 91.9 ± 35.9 87.2 ± 26.128 days 80.1 ± 21.6 85.5 ± 23.9

n = 5/sample. The measurements were obtained in samples collected 7 (7 days), 14 (14 days) and 28 (28 days) days after injury. Values are expressed as the mean and standarddeviation of each sample and were compared by the Student t-test (5% level of significance). (*) Significant difference compared to the control group.

F tions

g d 28 (

2

pacma2iwGa

2

faptt

ig. 5. Immunohistochemical detection of chondroitin-6-sulfate in longitudinal secroup: d–f. The treatment and observation periods were 7 (a and d), 14 (b and e) an

.8. Quantitative analysis of glycosaminoglycans

The total GAG content (mg/g tissue) of the samples (n = 5/timeoint) was determined based on the release of polysaccharidesfter digestion with pepsin 10 mg/g tissue in 30 mM sodiumitrate buffer, pH 3.5, containing 40 mM EDTA and 80 mM 2-ercaptoethanol for 24 h at 50 ◦C. The samples were centrifuged

nd two volumes of ethanol were added to the supernatant. After4 h at 4 ◦C, the precipitate was collected by centrifugation, washed

n 80% ethanol and acetone, and dried in an oven. The samples wereeighed and diluted in a known volume of water. The content ofAGs was quantified by the DMMB method (Farndale et al., 1986)nd absorbance was read in a visible spectrophotometer at 526 nm.

.9. Quantification of hydroxyproline

Tissue fragments (n = 5/time point) were immersed in acetoneor 48 h, followed by immersion in chloroform:ethanol (2:1) for an

dditional 48 h. The fragments were hydrolyzed in 6 N HCl (1 mLer 10 mg tissue) for 16 h at 110 ◦C. The hydrolysate was neu-ralized in 6 N NaOH and hydroxyproline was quantified usinghe chloramine T method as described by Stegemann and Stalder

of the elastic cartilage defect created in the ear of rats. Control group: a–c. Treatedc and f) days. Arrow: positive reaction; (*) chondrocytes. Bar = 100 �m.

(1967), with some modifications. Hydroxyproline concentrationsof 0.2–6 mg/mL were used for construction of the standard curve.

2.10. Extraction of extracellular matrix molecules

The tissue fragments were separated with a scalpel blade andfragmented in a Polytron® homogenizer immersed in saline. Afterrapid centrifugation, the precipitate was treated with 15 volumesof 50 mM sodium acetate buffer, pH 5.8, containing 4 M Gu-HCl,50 mM EDTA and 1 mM PMSF. Extraction was performed underconstant shaking for 24 h at 4 ◦C. After extraction, the samples werecentrifuged at 18,000 rpm for 20 min at 4 ◦C in a Beckman J2-21 (JA-20 rotor) centrifuge. The supernatant (total extract) was used forthe quantification of non-collagen proteins, Western blot analysis,and zymography for gelatinases.

2.11. Quantification of non-collagen proteins

Samples of the total extract of each group were used for thequantification of non-collagen proteins by the Bradford microassay(1976). The standard curve was constructed using bovine serumalbumin (0.3 to 5 �g/mL). In 96-well plates, 10 �L of each sam-

2 issue a

pAs

2

tsiiHTvbpoo

2

psentT((fti3udmgSdfsa

2

so

3

3i

rAaip

nFwi

30 E.P. Tangerino Filho et al. / T

le was treated with 200 �L reagent per well (Bio-Rad Proteinssay, #500-0006). Non-collagen proteins were quantified in apectrophotometer at 595 nm.

.12. Zymography for metalloproteinases

The supernatant of the total extract of each sample (50 �g pro-ein) was used for the analysis of MMP-2 activity. The samples wereubmitted to electrophoresis on 10% polyacrylamide gel contain-ng 0.1% gelatin at 4 ◦C. After electrophoresis, the gels were washedn 2.5% Triton X-100 and incubated in a solution of 50 mM Tris-Cl, pH 7.4, 0.1 M NaCl and 0.03% sodium azide for 21 h at 37 ◦C.he gels were stained for 1 h with Coomassie blue R-250 for obser-ation of the corresponding gelatinolytic activity-negative proteinands. EDTA (20 mM) in incubation buffer, which inhibits metallo-roteinase activity, was used as positive control. The band intensityf the different isoforms was determined for each group by densit-metry using the Alpha 4.0.3.2 software (Scion Corporation, USA).

.13. Western blotting

Samples of the total extract of each group containing 50 �grotein were fractionated by SDS-polyacrylamide gel electrophore-is under non-reducing conditions (Towbin et al., 1979). Afterlectrophoresis, the proteins were transferred to Hybond-ECLitrocellulose membranes (Amersham, Pharmacia Biotech, Arling-on Heights, IL, USA) using an electrical current of 70 V for 3 h.he membranes were blocked with chemiluminescent blockerWBAVDCH01, Millipore) for 20 s, washed in working solution0.01 M Tris, 0.15 M NaCl, and 0.05% Tween 20), and incubatedor 10 min at 4 ◦C with the following primary antibodies: collagenype II (C2456, Sigma) and decorin (HPA003315, Sigma) dilutedn working solution containing 1% BSA (1:500). Glyceraldehyde-phosphate dehydrogenase (GAPDH: sc-25778, Santa Cruz) wassed as the endogenous control under the same conditions asescribed for the primary antibodies and diluted 1:100. Next, theembranes were incubated for 10 min with HRP-conjugated anti-

oat and anti-mouse antibodies (A0545 and A0412, respectively,igma), diluted 1:500 in working solution plus 1% BSA. Peroxi-ase activity was detected by incubation with diaminobenzidineor 5 min. The band intensity was determined in each group by den-itometry using the Alpha 4.0.3.2 software (Scion Corporation, USA)nd is expressed as pixels.

.14. Statistical analysis

The mean quantitative results were entered into Excel spread-heets (statistics module) and analyzed by the Student t-test. A levelf significance of 5% was adopted.

. Results

.1. Structural analysis, morphometry andmmunohistochemistry

The lesional area maintained the circular macroscopic featureesulting from the injury mechanism in animals of CG and TG.dditionally, no hemorrhage or infectious process was observed atny of the time points studied. There was no significant differencen the percent reduction of the lesional area at any of the timeoints (Table 1).

Samples collected on day 7 from animals of TG exhibited a larger

umber of connective tissue cells (arrows) when compared to CG.urthermore, the presence of these cells organized in central islandsas more intense in TG at 14 days. The newly formed tissue exhib-

ted intense basophilia during the same period. At 28 days, islands

nd Cell 48 (2016) 224–234

of newly formed cartilage (arrowhead) with a strongly basophilicpericellular matrix were observed near the border and in the centerof the defect in both CG and TG, but the findings were more evidentin TG. Few blood vessels (*) were seen in the two groups at all timepoints studied (Fig. 1). The total number of connective tissue cellsin the repair area was significantly higher in TG than CG when thesame time point was compared (Table 1). The total number of gran-ulocytes analyzed by Dominici staining was significantly higher inTG animals at 7 and 14 days after injury, but was similar in bothgroups after 28 days (Table 1).

Collagen fibers (arrows) were detected at all time points in CGand TG. At 7 days, thick bundles arranged in parallel to the majorcartilage axis were observed throughout the lesional area, whilethese bundles were concentrated in the center of the defect at 14and 28 days. Apparently thinner fibers were detected at the bordersof the defect. In both regions, the collagen fibers were arrangedperpendicularly to the major cartilage axis (Fig. 2). The percent areaoccupied by birefringent collagen fibers was similar in CG and TGat 7 and 14 days, while at 28 days the area occupied by these fiberswas significantly greater in CG (Table 1).

Small numbers of elastic fibers (arrows) were detected amidstcollagen fibers (7 days) and in connective tissue areas near the car-tilage islands (*) and in the pericellular matrix of the newly formedcartilage (14 and 28 days) in the two groups (Fig. 3).

Immunohistochemical analysis showed a positive reaction totype II collagen (arrows) in the fibrous matrix of CG and TG animalsat 7 days and in the islands of newly formed cartilage (*) in thetwo groups at 14 and 28 days (Fig. 4). A similar immunohistochem-ical pattern was observed for chondroitin-6-sulfate, with a moreintense reaction (arrows) in the pericellular matrix of chondrocytesof newly formed cartilage in the two groups at 28 days (Fig. 5).

3.2. Ultrastructural and cytochemical analysis

In electron micrographs treated by the traditional method,fibroblast-like cells (*) were observed in CG animals from 7 to14 days. These cells were surrounded by a pericellular matrix ofloose fibrillar organization (arrowhead) located in the territorialmatrix. In samples of the same group after 28 days, the predominantcells assumed a chondroblast-like appearance, with a cytoplasmrich in secretory vesicles (arrow) and abundant rough endoplasmicreticulum (RER). The pericellular matrix exhibited a more compactcollagen fibril network (arrowhead) during this period. In TG ani-mals, most cells in the defect area had an intermediate appearancebetween fibroblasts and chondroblasts at 7 and 14 days. These cellscontained a large number of secretory vesicles in their cytoplasm.Furthermore, a fibrillar pericellular matrix of loose organizationwas observed (arrowhead). In this group, the cells presented a fib-rillar pericellular matrix of loose organization. In samples collectedat 28 days from TG animals, the cells exhibited features of chon-drocytes (*) characterized by a predominance of euchromatin andcytoplasm containing large numbers of RER and secretory vesicles(arrow). The matrix showed a loose network of collagen fibrils inthe pericellular and territorial region (Fig. 6). No significant differ-ence in the mean collagen fibril diameter was observed betweensamples obtained from the defect area of CG and TG at any of thetime points studied (Table 1).

In electron micrographs treated by the cytochemical methodfor the detection of PGs exhibited the typical filament pattern forpositive markings (arrowhead). The thin and small electron-densefilaments were detected in the pericellular and territorial matrixof all samples studied. No differences were observed in the distri-

bution or dimensions of PG-filaments between groups at 7 and 14days. Some thicker PG-filaments were observed in CG and TG at 28days (Fig. 7). There was also no difference in the number of posi-tive PG markings between CG and TG animals 7 days after injury.

E.P. Tangerino Filho et al. / Tissue and Cell 48 (2016) 224–234 231

Fig. 6. Electron micrographs of the elastic cartilage defect area created in the ear of rats. The samples were treated with 0.1% tannic acid. Control group: a–c. Treated group:d–f. The treatment and observation periods were 7 (a and d), 14 (b and e) and 28 (c and f) days. (*) Connective tissue cell; (�) fibrillar matrix; (→) secretory vesicle; (L) lipiddroplet. Bar = 2 �m.

Fig. 7. Electron micrographs of the elastic cartilage defect area created in the ear of rats. The samples were stained with cuprolinic blue for the detection of proteoglycans(PGs). Control group: a–c. Treated group: d–f. The treatment and observation periods were 7 (a and d), 14 (b and e) and 28 (c and f) days. (*) Connective tissue cells; (�)positive PG marking; (→) secretory vesicle; (L) lipid droplet. Bar = 2 �m.

232 E.P. Tangerino Filho et al. / Tissue and Cell 48 (2016) 224–234

Fig. 8. Zymogram of the analysis of the latent (72 kDa), intermediate (68 kDa) andactive (62 kDa) isoforms of MMP-2 in the different groups. Observe the presence ofttg

Aitp

3

psosnT

atoiw

stDafg77cb

4

itncatatccncwd

cech

Fig. 9. Western blotting for collagen type II (A) and decorin (B) using 15 and 20 �gof protein per sample, respectively. Alpha-chain of collagen II (120 kDa); beta-chain

he three isoforms in the two groups and at all time points. The bands seen in thereated group (TG) at 14 days are more intense than those observed in the controlroup (CG) at 14 days.

t 14 and 28 days, a significantly larger number of positive mark-ngs were observed in TG when compared to CG. At the same time,he number of positive markings increased gradually in TG over theeriod studied (Table 1).

.3. Biochemical analysis

The amount of GAGs increased gradually in CG and TG over theeriod studied, but significantly higher levels were observed in TGamples at 14 and 28 days when compared to CG. The amountf hydroxyproline also increased over the study period and wasignificantly higher in TG at 7, 14 and 28 days. No differences inon-collagen protein content were observed in extracts of CG andG samples at any of the time points (Table 2).

Zymography for MMP-2 detected three isoforms in all groupsnd at all time points. No significant difference was observed forhe latent or intermediate isoform in any of the samples. On thether hand, a larger amount of the active isoform was observedn TG at 14 days (CG: 90.2 ± 15.7/TG: 121.9 ± 16.9). No differences

ere detected at the other time points (Fig. 8).In Western blotting analysis, band densitometry indicated no

ignificant difference in the amount of the �- and �-chains ofype II collagen between groups or time points. In relation toecorin, possible differences between chains could not be evalu-ted because of the irregular migration profile of the glycosylatedorm in the gel. However, significantly higher amounts of the non-lycosylated form were observed in CG and TG at 28 days (CG:

days—70.6 ± 6.2/14 days—70.4 ± 6.7/28 days—80.5 ± 3.4; TG: days—63.1 ± 6.8/14 days—71.7 ± 4.9/28 days—79.8 ± 4.2) whenompared to the other time points, but there was no differenceetween groups (Fig. 9).

. Discussion

The repair of non-articular cartilage defects has been little stud-ed and knowledge of this process is important to improve thereatment of these injuries in clinical practice. Studies investigatingon-articular cartilage damage in young and adult animals reportedartilage repair to depend on the size of the injury, anatomical site,ge of the animal, and therapy used. Moyer et al. (2010) studiedhe regeneration of cylindrical defects of different diameters cre-ted in rat xiphoid cartilage. The results found were dependent onhe size of the defect created and on the origin of the chondro-ytes implanted in the center of the defect. However, the resultingartilage contained various chondrocytes aligned in columns andumerous PG deposits. In auricular cartilage, Costa et al. (2009),reating 2-mm cylindrical defects, showed that chondrogenesisas faster in young mice, with the observation of organizationalifferences between the different lines studied.

The model used in the present study was based on the proto-

ol described by Costa et al. (2009). Additionally, we analyzed theffect of microcurrent therapy on the repair dynamics of auricularartilage in rats. Our results did not show full closure of the initialole, but newly formed cartilage tissue was detected in both groups

of collagen II (200 kDa); GF: glycosylated form of decorin (110–60 kDa); NGF: non-glycanated form of decorin (49 kDa). Glyceraldehyde 3-phosphate dehydrogenase(GAPDH) (C) western blotting bands corresponding to control (CG) 7 d (a), 14 days(b), 28 days (c) and treated (TG) groups at 7 (d), 14 (e) and 28 (f) days after injury.

after 28 days of observation. These findings agree with Clark et al.(1998) who used an in vivo model of excisional injuries inducedin the outer ear of mice with 6 weeks, suggesting that electricalstimulation does not interfere with the macroscopic features of therepair of this type of injury.

On the other hand, an important finding of this study was theeffect of microcurrent stimulation, increasing the number of con-nective tissue cells and GAG content and inducing earlier tissueorganization in animals receiving this treatment for the same peri-ods of time. These results suggest a positive effect of electricalstimulation on the production, maintenance and organization ofECM. The findings are supported by the studies of Aaron et al.(2004) and Brighton et al. (2008) which demonstrated a stimula-tory effect of electrical and electromagnetic fields on the expressionof genes involved in the synthesis of ECM proteins by connectivetissue cells, increasing the deposition of cartilage and bone in areasof the regenerating tissue.

Microcurrent stimulation also induces the proliferation and dif-ferentiation of mesenchymal cells, which are abundant in younganimals (Nogami et al., 1982; Aaron et al., 2004; Akanji et al., 2008)and this effect increases when the damaged cartilage contains peri-chondrium, which is found in non-articular cartilage. In the presentstudy, the perichondrium was preserved at the border of the defectand its chondrogenic layer contained a large number of cells withdifferentiation potential. Several in vivo and in vitro studies haveshown acceleration of the repair of cartilage containing perichon-drium and that the newly formed tissue exhibits practically thesame properties as the original tissue, a fact differing from thatobserved in newly formed tissue of articular cartilage (Westers,1982; Buckwalter and Mankin, 1998; Nixon and Fortier, 2001).

The increased presence of granulocytes observed here in ani-mals treated for 7 and 14 days corroborates the most known effectsof treatment with low-frequency and low-amperage electrical cur-rents, i.e., acceleration of the inflammatory process (Davis, 1992;Okihana and Shimomura, 1988), which in fact might be related tothe improved healing and repair of connective tissues. Sonnewend

et al. (2005) observed that application of an electrical current oflow frequency and low intensity led to a five-fold increase in theproduction of adenosine triphosphate (ATP), accompanied by anincrease in protein synthesis and in the velocity of tissue repair.

issue a

T3saipmoi

evcicnlcswaclsesaacN

tssmte

itats

rfcrotHvcmH

ht1fitFseac

E.P. Tangerino Filho et al. / T

hese authors demonstrated the beneficial effect of a current of0 �A after 7 days of treatment in terms of the acceleration of tis-ue repair and reduction in the inflammatory process. However,nimals treated with 160 �A exhibited reduced inflammation onlyn the late stages of the healing process. These data agree with theresent findings and confirm that application of low amperagesodulates the early stages of inflammation, in contrast to the effect

f higher amperages which only act during the late stages of thenflammatory process.

Another finding of our study that can be explained by the earlyffect of microcurrent application during tissue repair is the obser-ation of a larger amount of birefringent fibers and consequently ofompacted collagen fibers in CG at 28 days, which was not observedn TG during the same period. This finding suggests that the appli-ation of an electrical current of low intensity and frequency doesot accelerate the maturation, compaction or reorganization of col-

agen fibers during the repair phase, when a greater degree ofompaction of these fibers is observed. Lee et al. (2007) demon-trated that ultralow doses of 3 �A were able to accelerate skinound closure in a study on humans. In this case, the microcurrents

ccelerated healing by increasing the deposition of collagen fibers,ell proliferation, and the concentration of growth factors and ATPevels. In addition to these observations, it should be noted that theubsensory stimuli generated by the microcurrents are able to pen-trate cells, normalizing natural bioelectricity after injury. Othertudies showed that microcurrents can also stimulate ion exchangecross biomembranes, with a subsequent increase in oxygenationnd nutrient absorption by cells, in addition to the removal ofatabolites and reestablishment of cell polarity (Cheng et al., 1982;ordenstrom, 1983; Kirsch and Mercola, 1995; Chao et al., 2000).

Although the parameters and the type of electrical stimulationhat is safe and effective in promoting connective tissue repair,pecifically that of elastic cartilage, have not yet been established,tudies have shown that low amperages are more effective in pro-oting the repair of different connective tissues since they are able

o restore the cell membrane potential (Nogami et al., 1982; Snydert al., 2002; Aaron et al., 2004; Dodge et al., 2007).

In contrast to collagen fibers, the number of elastic fibers did notndicate organizational changes caused by microcurrent therapy. Inhis study, elastic system elements were observed in all groups andt all time points. Despite their elastic nature, the deposition ofhese components in the cartilage found in the outer ear does noteem to be influenced by the treatment proposed here.

Studies using models of cartilage damage treated by microcur-ent application report that this therapy only tends to accelerate theormation of fibrocartilaginous tissue in chondral defects. In osteo-hondral defects, the vascularization derived from bone favors theepair of cartilage injuries (Poltawski and Watson, 2009). On thether hand, the size of the defects, how they are created and thereatment protocols also influence outcomes (Lippiello et al., 1990).owever, few studies have addressed the application of microgal-anic currents to pure cartilage defects and studies involving elasticartilage are even rarer, a fact highlighting the need to elucidate theechanisms of tissue repair induced by these agents (Black, 1985;addad et al., 2007).

The association between collagen fibrils and fibers and highlyydrated PGs is known to permit the tissue to withstand the func-ional demands during a normal activity (Reichenberger and Olsen,996; Brighton et al., 2008). In the present study, the mean collagenbril diameters did not differ between groups or periods. However,he amount of hydroxyproline was higher in TG at all time points.urthermore, the number of PG deposits and amount of GAGs were

ignificantly higher in TG at 14 and 28 days. This finding may bexplained by the fact that during the repair process collagen fibrilsre responsible for the containment of the regenerating stromalontent. Since the tissue is not submitted to important functional

nd Cell 48 (2016) 224–234 233

stress, these fibrils do not require the presence of elements withvariable calibers. However, despite the relatively uniform diameterin the groups, microcurrent therapy (TG) stimulated the synthesisof more collagen molecules as demonstrated by the measurementof hydroxyproline. This finding might be associated with the accel-eration of tissue reorganization in TG as a whole, which requireslarger amounts of collagen fibrils and fibers of smaller caliber and,as discussed above, less compacted and less birefringent. On theother hand, the number of PG deposits and GAG content seem toindicate a positive effect of microcurrent therapy on the deposi-tion of these molecules in the tissue. A previous study from ourgroup showed that microcurrent application stimulates the depo-sition of PGs in ECM during the repair of hyaline cartilage obtainedfrom non-articular sites (de Campos Ciccone et al., 2013). Simi-larly, Liu et al. (1996) observed an increase in the in vitro synthesisof GAGs using primary cultures of chicken xiphoid cartilage, andHilz et al. (2014) demonstrated that low-intensity electromagneticfields strongly stimulate the deposition of GAGs in bovine articularcartilage explants.

Chondroitin-6-sulfate is the most abundant GAG in the elasticcartilage matrix (Wang et al., 2008). Observation of the distributionof this molecule during repair revealed no important differencesbetween groups. However, since the total amount of GAGs washigher in TG, it remains to be determined in future studies whetherstructural differences exist between these components, similar tothe descriptions of Liu et al. (1996) and Chang et al. (2011).

Within this context, the present results suggest that the syn-thesis of decorin, a small proteoglycan involved in the regulationof collagen fibrillogenesis (Iozzo and Schaefer, 2010), is notinfluenced by the treatment used in this study. Similar findingshave been reported by De Mattei et al. (2007) for bovine articularcartilage explants.

In an environment where the synthesis and deposition of colla-gen molecules are stimulated by microcurrent therapy, evaluationof the activity of gelatinolytic matrix metalloproteinases is impor-tant. The present results demonstrated higher activity of the activeisoform of MMP-2 in TG during the proliferative phase of repair (14days). This finding may be related to the presence of less compactedand birefringent collagen fibers in this group. Similar results havebeen reported by Hembry et al. (2001) who studied articular carti-lage repair in young pigs, and by Alves et al. (2014) who evaluatedthe effect of photodynamic therapy on articular cartilage defects inadult rats.

The predominance of chondroblast-like cells in the repair areaof TG since the early stages of therapy highlights another importantaspect of the effect of microcurrent stimulation on the differenti-ation of mesenchymal cells. These data agree with studies in theliterature (Baker et al., 1974; Okihana and Shimomura, 1988; Takeiand Akai, 1993; Snyder et al., 2002) that demonstrated an increasein the amount of type II collagen and PGs in connective tissue ofyoung animals, particularly cartilaginous tissue, in addition to theearly stimulation of morphofunctional differentiation of connectivetissue cells, using different protocols of electrical stimulation.

5. Conclusion

Our results permit to conclude that microcurrent electrical stim-ulation triggered the proliferation and differentiation of connectivetissue cells, the deposition of GAGs and collagen, and the struc-tural reorganization of these elements during the repair of elasticcartilage in young animals.

References

Aaron, R.K., Boyan, B.D., Ciombor, D.M., Schwartz, Z., Simon, B.J., 2004. Stimulationof growth factor synthesis by electric and electromagnetic fields. Clin. Orthop.Relat. Res. 419, 30–37.

2 issue a

A

A

A

B

B

B

B

B

C

C

C

C

C

C

D

d

D

D

F

H

H

H

H

H

I

I

I

Academic Press New York, USA.Wang, Q.G., El Haj, A.J., Kuiper, N.J., 2008. Glycosaminoglycans in the pericellular

34 E.P. Tangerino Filho et al. / T

kanji, O.O., Lee, D.A., Bader, D.A., 2008. The effects of direct current stimulation onisolated chondrocytes seeded in 3D agarose constructs. Biorheology 45,229–243.

lves, A.C., Albertini, R., dos Santos, S.A., Leal-Junior, E.C., Santana, E., Serra, A.J.,Silva, J.A., de Carvalho, P.T., 2014. Effect of low-level laser therapy onmetalloproteinase MMP-2 and MMP-9 production and percentage of collagentypes I and III in a papain cartilage injury model. Lasers Med. Sci. 29, 911–919.

nraku, Y., Mizuta, H., Sei, A., Kudo, S., Nakamura, E., Senba, K., Hiraki, Y., 2009.Analyses of early events during chondrogenic repair in rat full-thicknessarticular cartilages defects. J. Bone Miner. Metab. 27, 272–286.

aker, B., Spadaro, J., Marino, A., Becker, R.O., 1974. Electrical stimulation ofarticular cartilage regeneration. Ann. N. Y. Acad. Sci. 238, 491–499.

lack, J., 1985. Electrical stimulation of hard and soft tissue in animal models. Clin.Plast. Surg. 12, 243–257.

radford, M.M., 1976. Rapid and sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principle of protein-dye binding.Anal. Biochem. 72, 248–254.

righton, C.T., Wang, W., Clark, C.C., 2008. The effect of electrical fields on gene andprotein expression in human osteoarthritic cartilage explants. J. Bone Jt. Surg.Am. 90, 833–838.

uckwalter, J.A., Mankin, H.J., 1998. Articular cartilage, tissue design andchondrocyte-matrix interactions. Instr. Course Lect. 47, 477–486.

arrington, J.L., 2005. Aging bone and cartilage: cross-cutting issues. Biochem.Biophys. Res. Commun. 18, 700–708.

hang, S.H., Hsiao, Y.W., Lin, H.Y., 2011. Low-frequency electromagnetic fieldexposure accelerates chondrocytic phenotype expression on chitosansubstrate. Orthopedics 34, 20.

hao, P.-H., Roy, R., Mauck, R.L., Liu, W., Valhmu, W.B., Hung, C.T., 2000.Chondrocyte translocation response to direct current electric fields. J. Biomech.Eng. 122, 261–267.

heng, N., Hoof, H.V., Bock, E., Hoogmartens, M.J., Mulier, J.C., Ducker, F.J., Sansen,W.M., Loecker, W., 1982. The effect of electric current on ATP generation,protein synthesis, membrane transport in rat skin. Clin. Orthop. Relat. Res. 171,264–272.

lark, L.D., Clark, R.K., Heber-Katz, E., 1998. A new murine model for mammalianwound repair and regeneration. Clin. Immunol. Immunopathol. 88, 35–45.

osta, R.A., Ruiz-de-Souza, V., Azevedo Jr., G.M., Vaz, N.M., Carvalho, C.R., 2009.Effects of strain and age on ear wound healing and regeneration in mice. Braz.J. Med. Biol. Res. 42, 1143–1149.

avis, P., 1992. Microcurrent: a modern healthcare modality. Rehab. Ther. Prod.Rev. 10, 62–66.

e Campos Ciccone, C., Zuzzi, D.C., Neves, L.M., Mendonc a, J.S., Joazeiro, P.P.,Esquisatto, M.A., 2013. Effects of microcurrent stimulation on hyaline cartilagerepair in immature male rats (Rattus norvegicus). BMC Complement. Altern.Med. 19, 13–17.

e Mattei, M., Fini, M., Setti, S., Ongaro, A., Gemmati, D., Stabellini, G., Pellati, A.,Caruso, A., 2007. Proteoglycan synthesis in bovine articular cartilage explantsexposed to different low-frequency low-energy pulsed electromagnetic fields.Osteoarthr. Cartil. 15, 163–168.

odge, G.R., Bowen, J.R., Chang-Wug, O.H., Tokmakova, K., Simon, B.J., Aroojis, A.,Potter, K., 2007. Electrical stimulation of the growth plate: a potentialapproach to an epiphysiodesis. Bioelectromagnetics 28, 463–470.

arndale, R.W., Buttle, D.J., Barrett, A.J., 1986. Improved quantitation anddiscrimination of sulphated glycosaminoglycans by use of dimethylmethyleneblue. Biochim. Biophys. Acta 883, 173–177.

addad, J.B., Obolensky, A.G., Shinnick, P., 2007. The biology effects and thetherapeutic mechanism of action of electric and electromagnetic fieldstimulation on bone cartilage: new findings and review of earlier work. J.Altern. Complement. Med. 13, 485–490.

embry, R.M., Dyce, J., Driesang, I., Hunziker, E.B., Fosang, A.J., Tyler, J.A., Murphy,G., 2001. Immunolocalization of matrix metalloproteinases in partial-thicknessdefects in pig articular cartilage. A preliminary report. J. Bone Jt. Surg. Am. 83,826–838.

ennerbichler, A., Rosenberger, R., Arora, R., Hennerbichler, D., 2008. Biochemical,biomechanical and histological properties of osteoarthritic porcine kneecartilage: implications for osteochondral transplantation. Arch. Orthop.Trauma Surg. 128, 61–70.

ilz, F.M., Ahrens, P., Grad, S., Stoddart, M.J., Dahmani, C., Wilken, F.L., Sauerschnig,M., Niemeyer, P., Zwingmann, J., Burgkart, R., von Eisenhart-Rothe, R.,Südkamp, N.P., Weyh, T., Imhoff, A.B., Alini, M., Salzmann, G.M., 2014. Influenceof extremely low frequency, low energy electromagnetic fields and combinedmechanical stimulation on chondrocytes in 3-D constructs for cartilage tissueengineering. Bioelectromagnetics 35, 116–128.

yttinen, M.M., Arokoskit, J.P.A., Parkkinen, J.J., Lammi, M.J., Lapvetelainen, T.,Mauranent, K., Király, K., Tammi, M.I., Helminen, H.J., 2001. Age matters:collagen birefringence of superficial articular cartilage is increased in youngguinea-pigs but decreased in older animals after identical physiological type ofjoint loading. Osteoarthr. Cartil. 9, 694–701.

ozzo, R.V., 1998. Matrix proteoglycans: from molecular design to cellular function.Annu. Rev. Biochem. 67, 609–652.

ozzo, R.V., Schaefer, L., 2010. Proteoglycans in health and disease: novel regulatory

signaling mechanisms evoked by the small leucine-rich proteoglycans. FEBS J.277, 3864–3875.

to, I., Imada, M., Ikeda, M., Sueno, K., Arikuni, T., Kida, A., 2001. Morphologicalstudy of age changes in adult human auricular cartilage with special emphasison elastic fibers. Laryngoscope 111, 881–886.

nd Cell 48 (2016) 224–234

Johnson, T.S., Xu, J.-W., Zaporojan, V.V., Mesa, J.M., Weinand, C., Randolph, M.A.,Bonassar, L.J., Winograd, J.M., Yaremchuk, M.J., 2004. Integrative repair ofcartilage with articular and nonarticular chondrocytes. Tissue Eng. 10,1308–1315.

Khan, I.M., Gilbert, S.J., Singhrao, S.K., Duance, V.C., Archer, C.W., 2008. Cartilageintegration: evaluation of the reasons for failure of integration during cartilagerepair. Eur. Cell Mater. 16, 26–33.

Khubutiva, M.S., Yu, I., Kliukvin, L.P., Istranov, V.B., Khvatov, A.B., Shekhter, A., Vaza,I.V., Kanakov, V.S., 2008. Stimulation of regeneration of hyaline cartilage inexperimental osteochondral injury. Bull. Exp. Biol. Med. 146, 658–661.

Kielty, C.M., Sherratt, M.J., Shuttleworth, C.A., 2002. Elastic fibres. J. Cell Sci. 115,2817–2828.

Kirsch, D.L., Mercola, J.M., 1995. The basis for microcurrent electrical therapy inconventional medical practice. J. Adv. Med. 8, 83–97.

Lee, B.Y., Wendell, K., Al-Waili, N., Butler, G., 2007. Ultra-low microcurrenttherapy: a novel approach for treatment of chronic resistant wounds. Adv.Ther. 24, 1202–1209.

Lippiello, L., Chakkalakal, D., Connolly, J.F., 1990. Pulsing direct current-inducedrepair of articular cartilage in rabbit osteochondral defects. J. Orthop. Res. 8,266–275.

Liu, H., Abbott, J., Bee, J.A., 1996. Pulsed electromagnetic fields influence hyalinecartilage extracellular matrix composition without affecting molecularstructure. Osteoarthr. Cartil. 4, 63–76.

Moyer, H.R., Wang, Y., Farooque, T., Wick, T., Singh, K.A., Xie, L., Guldberg, R.E.,Williams, J.K., Boyan, B.D., Schwartz, Z., 2010. A new animal model forassessing cartilage repair and regeneration at a nonarticular site. Tissue Eng.Part A 16, 2321–2330.

Myllyharju, J., Kivirikko, K.I., 2001. Collagens and collagen-related diseases. Ann.Med. 33, 7–21.

Negri, S., Farinato, S., Fila, C., Pagliaro, P., Bellomi, A., 2007. Tissue engineering:chondrocyte cultures on type I collagen support: cytohistological andimmunohistochemical study. J. Orthop. Traumatol. 8, 57–63.

Newman, A.P., 1998. Articular cartilage repair. Am. J. Sports Med. 26, 309–324.Nieduszynski, I.A., Huckerby, T.N., Dickenson, J.M., Brown, G.M., Tai, G.H., Morris,

H.G., Eady, S., 1990. There are two major types of skeletal keratin sulphates.Biochem. J. 271, 243–249.

Nixon, A.J., Fortier, L.A., 2001. New horizons in articular cartilage repair. AAEP Proc.47, 217–226.

Nordenstrom, B., 1983. Biologically Closed Electrical Circuits. Nordic MedicalPublications, Uppsala, Sweden.

Nogami, H., Aoki, H., Okagawa, T., Mimatsu, K., 1982. Effects of electric current onchondrogenesis in vitro. Clin. Orthop. Relat. Res. 163, 243–247.

Okihana, H., Shimomura, Y., 1988. Effect of direct current on cultured growthcartilage cells in vitro. J. Orthop. Res. 6, 690–694.

Poltawski, L., Watson, T., 2009. Bioelectricity and microcurrent therapy for tissuehealing—a narrative review. Phys. Ther. Rev. 14, 104–114.

Rajnoch, C., Ferguson, S., Metcalfe, A.D., Herrick, S.E., Willis, H.S., Ferguson, M.W.,2003. Regeneration of the ear after wounding in different mouse strains isdependent on the severity of wound trauma. Dev. Dyn. 226, 388–397.

Reichenberger, E., Olsen, B.R., 1996. Collagens as organizers of extracellular matrixduring morphogenesis. Semin. Cell Dev. Biol. 7, 631–638.

Reynolds, E.S., 1963. The use of lead citrate at high ph as an electron-opaque stainin electron microscopy. J. Cell Biol. 17, 208–212.

Scott, J.E., Haigh, M., Nusgens, B., Lapière, C.M., 1989. Proteoglycan: collageninteractions in dermatosparactic skin and tendon: an electron histochemicalstudy using cupromeronic blue in a critical electrolyte concentration method.Matrix 9, 437–442.

Snyder, M.J., Wilensky, J.A., Fortin, J.D., 2002. Current applications ofelectrotherapeutics in collagen healing. Pain Physician 5, 172–181.

Sonnewend, D., Oliveira, J.L.R., Nicolau, R.A., Zângaro, R.A., Pacheco, M.T.T., 2005.Avaliac ão do efeito da microterapia celular (microcorrentes) sobre o processoinicial da cicatrizac ão de feridas em ratos. In: XI INIC Encontro LatinoAmericano de Iniciac ão Científica e V EPG Encontro Latino Americano de PósGraduac ão. Martra Editora, Rio de Janeiro, Brazil.

Souza, T.D., Junqueira, R., Del Carlo, R.J., Viloria, M.I.V., 2001. Eletroterapia noprocesso de reparac ão da superfície articular de coelhos. Ciência Rural 31,819–824.

Stegemann, H., Stalder, K., 1967. Determination of hydroxyproline. Clin. Chim. Acta18, 267–273.

Stockwell, R.A., 1967. Lipid content of human costal and articular cartilage. Ann.Rheum. Dis. 26, 481–486.

Takei, N., Akai, M., 1993. Effect of direct current stimulation on triradiate physealcartilage. In vivo study in young rabbits. Arch. Orthop. Trauma Surg. 112,159–162.

Towbin, H., Staehelin, T., Gordon, J., 1979. Electrophoretic transfer of proteins frompolyacrylamide gels to nitrocellulose sheets: procedure and some applications.Proc. Natl. Acad. Sci. U. S. A. 76, 4350–4354.

Vogel, K.G., 1994. Glycosaminoglycans and proteoglycans. In: Yurchenco, P.D., Birk,D.E., Mecham, R.P. (Eds.), Extracellular Matrix Assembly and Structure.

matrix of condrons and chondrocytes. J. Anat. 213, 266–273.Westers, B.M., 1982. Review of the repair of defects in articular cartilage: Part II.

JOSPT 4, 6–15.