Anti-inflammatory and antioxidant activities of Citrus aurantifolia (Christm.) Swingle essential oil in dystrophic muscle cells: implication of the PGC-1α pathway

Abstract The commercial essential oil obtained from the steam distillation of cold-pressed oil from Citrus aurantifolia (Christm.) Swingle fruits (OCA), after chemical characterization by gas chromatography-mass spectrometry (GC-MS), was investigated for antioxidant and anti-inflammatory activities and related signaling pathways in dystrophic muscle cells. The chemical composition of OCA showed 24 compounds with a predominance of monoterpenes, such as limonene (42.6%), cymene (11.8%), α-terpineol (11.8%), terpinolene (5.3%) and γ-terpinene (4.6%). DPHH and ORAC assays showed a significant antioxidant activity of OCA. Because the OCA is rich in limonene, it exhibited marked anti-inflammatory (as indicated by the reduction of TNF- α and NF-κB levels) and antioxidant effects (as indicated by the reduction of 4-HNE-protein adducts, SOD-2, GSR and H2O2 levels) on dystrophic muscle cells. These effects appear to be mediated through the activation of PGC-1α pathway. Therefore, OCA can be considered a promising bioactive for therapeutic applications in dystrophinopathies.


Introduction
Plants from the genus Citrus has shown importance to folk medicine worldwide.Growing interest has flourished, especially in its pharmacological properties.This genus belongs to the Rutaceae family which comprises almost 160 genera and 1700 species 1 .Among the species, one of the best known is Citrus aurantifolia, which is native to the tropical and subtropical regions of Asia and Southeast Asia and was introduced to North Africa, Europe, and worldwide 2 .The C. aurantifolia is also known by other vernacular names such as lime (English); citronnier (French); limone (German); Kagzinimbu (Hindi); and jeruk nipis (Indonesia) 3 .C. aurantifolia is a perennial tree that can grow to a height of 3-5 m, displaying stem being slender branched; having alternate and elliptical to oval leaves; with long and narrowly winged petioles; short flowers and auxiliary racemes, bearing few white and fragrant flowers; and round green fruit, which are yellow when rip 3,4 .C. aurantifolia contains active substances such as essential oil, flavonoids; flavones; flavanones; triterpenoids; and limonoids 5,6 .At least 62 volatile compounds in the fruit peel oils and 59 in the leaf oils of several lime species were reported.In the fruit peel oils, limonene was the major volatile component, followed by terpinene, pinene, and sabinene 7 .The traditional uses of C. aurantifolia included: antibacterial; antidiabetic; antifungal; antihypertensive; anti-inflammatory; HKB anti-lipidemic; antioxidant; anti-parasitic; and antiplatelet activities [8][9][10][11][12] .Essential Citrus oils are complex mixtures comprising many chemical compounds, such as limonene, γ-terpinene, citral and linalool, which have attracted researchers to investigate their pharmacological properties.Limonene is a major constituent of numerous essential citrus oils, such as an oil from Citrus aurantifolia (OCA), commonly extracted from the peel of different citrus fruits.It is a chiral monoterpene, naturally present as two optical isomers, called D-or L-limonene.D-Limonene (1-methyl-4-(1-methylethenyl) cyclohexane) with a molecular weight of 136.24 g/mol, has recognized medicinal properties, such as the reduction of inflammatory and oxidative parameters in several in vitro and in vivo experimental models 13,14 .
The existence of hyperoxidative status and inflammation in dystrophic patients and mdx mice, the experimental model for Duchenne muscular dystrophy (DMD) 15,16 , suggests that the search for new drugs and natural products with anti-inflammatory and antioxidant properties may have beneficial effects in this disease.Based on this view, previous studies have shown that natural products such as bis-indolic alkaloid 15 ; Coenzyme Q10 17 and green tea extract 18 reduces oxidative stress, inflammatory mediators and muscle damage in mdx mice.
DMD is a lethal disorder characterized by the lack of the dystrophin protein that results in sarcolemmal damage and skeletal muscle degeneration, affecting approximately 1 in 6,000 live male births 19 .Currently, corticosteroids are main drugs used to improve and prolong the life of dystrophic patients, but have numerous side effects that limit their long-term use 20 .Based on pharmacological indications of the essential Citrus oils disclosed above, in the present study the OCA, after chemical characterization, was assayed for the antioxidant and antiinflammatory activities on dystrophic muscle cells.In addition, we also evaluated the effects of OCA on the peroxisome proliferator-activated receptor-γ coactivator-1α (PGC-1α), since this factor is associated to the anti-inflammatory and the antioxidant pathways 21 .

Citrus aurantifolia essential oil
Commercial OCA was acquired from Ferquima ® (Lime Distilled Essential Oil -INCL: Citrus aurantifolia Peel Oil Distilled, CAS Number: 90063-52-8, Vargem Grande Paulista, São Paulo-Brazil).According to the manufacturer, the commercial lime distilled oil is the essential oil obtained from the steam distillation of coldpressed oil from the C. aurantifolia fruit.

Gas chromatography-mass spectrometry (GC-MS) analysis
The OCA was analyzed to determine its chemical composition profile by GC-MS (Model GC2010/ QP2010 Plus; Shimadzu Corporation, Kyoto, Japan).The apparatus was equipped with an AOC-20i (Shimadzu) auto sampler and using a mass selective detector and a Rtx-5MS fused capillary column (30 m x 0.25 mm x 0.25 μm film thickness).Helium gas was used as carrier with a flow of 1.2 mL/min.The column temperature program was 50°C/2 minutes, a rate of 4°C/minute to 200°C, then a rate of 15°C/minute to 300°C, and then 300°C/15 minutes 22 .The temperatures of the injector and the ion source were 250°C and 280°C, respectively.The injection solution was prepared by dissolving about 1.0 μL of oil in 100 μL of CH 2 Cl 2 , and 1 µL of the solution was injected using a split ratio of 1:100, v/v.MS Spectra were taken at 70 eV with a scan interval of 0.3 s over the mass range 40-350 Da.The other conditions were similar to the GC analysis 22 .To obtain the retention index a homologous series of normal alkanes (C 8 -C 20 ) were also injected that allowed the calculation of the retention index of each oil constituent, according to Vandendool and Kratz 23 .The identification of each constituent was made by both comparing their mass spectra with those from the spectrometer database (NIST  1998) or found in literature 24 and the retention index relative to the n-alkane.The analyses of the essential oil were performed in triplicate.

Antioxidant activity TLC autographic assay for DPPH radicalscavenging
The antioxidant activity of OCA using TLC autographic assay for DPPH radical-scavenging assay was carried out according to the method described by Costa et al. 22 and used as a qualitative preliminary screening.Ten microliters of a 1:250, v/v dilution of the essential oil of C. aurantifolia in methanol was applied to thinlayer chromatography (TLC) plates (silica gel 60 GF254, Fluka, AG, Switzerland).These plates were sprayed with a 0.2% 2,2-diphenyl-1-pi crylhydrazyl (DPPH; Sigma-Aldrich, St Louis, MO) solution in MeOH and left at room temperature.Active compounds, observed as yellow spots against a purple background, were observed 30 minutes after spraying.Relative radical-scavenging activity was assigned as 'strong' (samples that produced an intense bright yellow zone), 'medium' (samples that produced a clear yellow spot), 'weak' (samples that produced only a weakly visible yellow spot) and 'not active' (samples that produced no sign of any yellow spot).

ORAC FL kinetic assay
The antioxidant capacity of the C. aurantifolia essential oil was assessed quantitatively through the oxygen radical absorbance capacity (ORAC) assay.This assay measures scavenging activity against the peroxyl radical [azobis (2-amidinopropane) dihydrochloride (AAPH), Aldrich, Milwaukee, WI, using fluorescein (Aldrich, Milwaukee, WI) as the fluorescent probe (ORAC FL ).The ORAC assays were carried out on a Synergy 2 (Biotek, Winooski, VT) multidetection microplate reader system.The temperature of the incubator was set at 37°C.The procedure was carried out according to the method established by Ou et al. 25 with minor modifications 26 .The data are expressed as μmol of Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid; Aldrich, Milwaukee, WI) equivalents (TE) per g of oil (μmol of TE/g).In these tests, limonene, caryophyllene oxide and (E)-caryophyllene were used as positive controls.The analyses were performed in triplicate.

Primary muscle cell cultures
C57BL/10 (C57BL/10ScCr/PasUnib) and mdx mice (C57BL/10-Dmdmdx/PasUnib), with 28 days of age, were used in all the experiments.These animals were housed according to institutional guidelines and had access to food and water ad libitum.All animal experiments were conducted in accordance with the guidelines of the Brazilian College for Animal Experimentation (COBEA; process #3168-1) and the guidelines set forth by our institution.Primary culture of skeletal muscle cells was performed as previously described 27,28 .Hind limbs (quadriceps femoris, tibialis anterior, extensor digitorum longus, gastrocnemius, soleus, and plantaris muscles) were removed from C57BL/10 and mdx mice and used to prepare primary muscle cell culture.All muscles were triturated and enzymatically digested with collagenase and trypsin solutions.The satellite cells (5 × 10 4 cells/cm 2 ) were plated in Matrigel TM -coated dishes.The myoblasts were cultured in a proliferation and growth medium containing DMEM with glucose, L-glutamine, fetal bovine serum, horse serum and penicillin/ streptomycin.Myotube differentiation was induced by the addition of a fusion medium that consisted of DMEM with glucose, L-glutamine and horse serum.Differentiated muscle cells were observed at 7-8 days of culture.Throughout the experiment, the cells were maintained at 37°C and 5% CO 2 .

Muscle cells viability
Cell viability was assessed by morphology and by tetrazolium [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] MTT (Sigma) assay.The protocol was based on previous studies 28,29 .After washing with PBS, the skeletal muscle cells were incubated with MTT solution for 4 h at 37°C.Next, the supernatant was discarded and the acid isopropanol was used to dissolve the MTT crystals.The absorbance was measured at 570 nm by a multi-mode microplate reader model Synergy H1M (Bio Tek Instruments, Washington DC, USA).The analyzes were carried out in triplicate.Percentage viability was defined as the relative absorbance of treated versus untreated muscle cells.

Determination of H 2 O 2 level on muscle cells
The Amplex UltraRed reagent and horseradish peroxidase were used to determine musclederived ROS.Amplex reacts with H 2 O 2 , in the presence of horseradish peroxidase, to produce resorufin, a stable red fluorescent.The fluorescence signal was measured at 530 (excitation) and 590 nm wavelength (emission) by a multi-mode microplate reader model Synergy H1M (Bio Tek Instruments, Washington DC, USA).The protocol was based on previous studies 28,30 .

Proteins analysis by western blotting on muscle cells
This protocol was based on previous studies 28,31 .After proteins extraction, the cell extracts were sonicated and the homogenates were centrifuged.Next, the supernatant was used to determine the total protein content by the Bradford method.Total protein content (30 µg) was loaded onto SDS-polyacrylamide gels (6-15%).Proteins were transferred from the gels to a nitrocellulose membrane.After blocking the membranes with 5% skim milk/Tris-HCl buffer saline-Tween buffer for 2 h, these are incubated with the primary antibodies overnight.Afterwards, they were washed with TBST and incubated with the peroxidase-conjugated secondary antibodies for 2 h.To control protein loading, western blot transfer, and nonspecific changes in protein levels, the blots were stripped and re-probed for β-actin.Band intensities were quantified using ImageJ 1.38X (National Institutes of Health, Bethesda, MD, USA) software.The following primary antibodies were used for Western blotting: 4-HNE (BIO-RAD AHP1251); anti-Catalase (Sigma-Aldrich C0979); anti-SOD-2 (Sigma-Aldrich SAB2501676); anti-GSR (Sigma-Aldrich SAB2108147); PGC1α (Calbiochem mAb4C1.3);NF-kB (BIO-RAD AHP1342); TNF-α (Cell Signaling #3707S) and β-actin (Sigma-Aldrich A1978).The secondary antibody used was peroxidase-labeled affinity purified mouse (Promega Corporation W4021) or rabbit IgG antibody (Promega Corporation W4011).

Statistical analysis
Direct comparison between means of two groups was performed by Student's t test.Multiple statistical comparisons between groups, were carried out by ANOVA, followed by the Tukey's test.
All data are expressed as mean ± standard deviation (SD).Values of P< 0.05 were considered statistically significant.
After the chemical composition of OCA was analyzed, the cytotoxicity of this essential oil was examined in primary muscle cells.Cell viability in skeletal muscle cells, supplemented with OCA, is shown in Fig. 1.Because the OCA was dissolved in Tween-80, some skeletal muscle cells received the same amount of Tween-80 to exclude any interference due to the vehicle.OCA treatment did not affect cell viability in muscle cells (Ctrl+OCA and mdx+OCA groups) when compared to untreated muscle cells (Ctrl and mdx groups).In agreement with our cell Compounds are listed in order of their elution from a DB-5 column; the coefficients of variation obtained in these analyses (n=3) were below 5%.a Identification based on mass spectra and RI (retention index) published (Adams,1995) and computer matching of the mass spectra with NIST 1998 library (quality level more than 90%); b retention index calculated on a DB-5 column according to Van Den Dool and Kratz (1963), based on a homologous series of normal alkanes; c retention index published (Adams, 2007)   viability results, 37,14 also found non-toxic effects of limonene within the range of 1-50 μM in 3T3-L1white adipocytes.
To verify the biological protective effect of OCA in dystrophic inflammatory response, we analyzed the levels of both tumor necrosis factor-alpha (TNF-α), a key cytokine that stimulates the inflammatory cell response in mdx mice 38 and nuclear factor kappa-B (NF-κB), a transcription factor that regulates the expression of pro-inflammatory cytokines 39 .In our experimental conditions, immunoblotting analyses demonstrated a significant increase in TNF-α levels in dystrophic muscle cells (mdx group) compared with control muscle cells (Ctrl and Ctrl+OCA groups; Fig. 2A,C).Regarding NF-κB, although there is a tendency for an increase, no statistically significant differences were observed in its levels in dystrophic muscle cells (mdx group) compared with control muscle cells (Ctrl and Ctrl+OCA groups; Fig. 2A,B).The in-vitro treatment of dystrophic muscle cells with OCA was effective in reducing TNF-α and NF-κB levels in dystrophic muscle cells (by 87%, and 36%, respectively; Fig. 2A,B,C).A previous study reported that the anti-inflammatory effects of essential oils can be attributed to the presence the bioactive monoterpenes, such as α-terpineol and limonene 40 .In addition, studies with experimental models and human beings demonstrated the anti-inflammatory effects of oral administered limonene 35,41 .So, it is possible that the anti-inflammatory effect of OCA, verified in our experiments, is due to pharmacological approaches of limonene.In agreement with this suggestion, a lesser responsiveness to TNFαinduced NF-κB translocation was seen in fibroblast cells treated with orange peel extract containing large amounts of limonene, thus supporting the anti-inflammatory effect of this compound as an active process, which implies in well-defined cell-signaling pathways 35 .
Considering that oxidative stress is implicated in dystrophic muscle damage, we also analyzed OCA effects on ROS production (H 2 O 2 production), lipid peroxidation and antioxidant enzymatic system.
Representative immunoblots and quantification of 4-hydroxynonenal (4-HNE)-protein adducts 5 fig 3 (a marker of lipid peroxidation) are shown in Fig. 3A,B.The 4-HNE-protein adduct levels significantly increased in the mdx group when compared to Ctrl and Ctrl+OCA groups (Fig. 3A,B).Levels of 4-HNE-protein adducts were significantly reduced by OCA in the mdx+OCA group when compared to the mdx group (Fig. 3A,B).Similar results were observed in H 2 O 2 production (Fig. 3C).Regarding the antioxidant enzymatic system, the catalase levels showed no significant differences between the muscle cells from the experimental Ctrl, Ctrl+OCA, mdx and mdx+OCA groups (Fig. 4A,B).The SOD-2 levels were significantly higher in the mdx group when compared to the Ctrl and Ctrl+OCA groups (Fig. 4A,C).The treatment with OCA significantly decreased SOD-2 levels in the mdx+OCA group when compared to the Ctrl, Ctrl+OCA and mdx groups (Fig. 4A,C).The GSR levels were significantly higher in the mdx group when compared to the Ctrl+OCA group (Fig. 4A,D).The OCA treatment significantly decreased GSR levels in the Ctrl+OCA group when compared to the Ctrl group (Fig. 4A,D) and in the mdx+OCA group when compared to the Ctrl and mdx groups (Fig. 4A,D).
Lipid peroxidation is an oxidative stress marker and studies from our research group have reported an elevation in the level of 4-HNE, a lipid peroxidation product, in dystrophic muscle fibers 15,16 .The present results agree with the above mentioned previous findings which showed that there is a remarkable increase in the level of 4-HNE in dystrophic muscle cells and that treatment with OCA significantly reduced this increase.Concomitantly, OCA treatment also had an effect on ROS levels, reducing the H 2 O 2 production in mdx muscle cells.Similar results were observed in lymphoid cell suspensions treated with limonene, where it could be seen  42 .In addition, another study also showed that the administration of D-limonene to diabetic rats caused a significant reduction in the levels of lipid peroxidation by-products and an increase in the activities of antioxidant enzymes 43 .
Some studies associate ROS level reduction in dystrophic muscle cells to the antioxidant defense system 44,45 .However, in the present study, we observed a reduction in the levels of SOD and GSR antioxidant enzymes after OCA treatment, suggesting that the beneficial OCA effects could not be directly due to the up-regulation of the antioxidant defense system.In agreement with our results, it has been shown that D-limonene decreased SOD activity in the injured skeletal muscle of rats 46 .This research also reports that in their experiments, D-limonene did not change the catalase activity.In our experimental conditions, we also did not observe alterations in catalase levels after OCA treatment.Based on our results, the reduction of SOD and GSR levels after essential oil treatment indicates that oxidative stress was decreased.
In addition, the antioxidant properties of OCA were confirmed in this study by DPPH and ORAC analyses.The results of the antioxidant activity of the essential oil and quercetin, limonene, caryophyllene and (E)-caryophyllene are summarized in Table 2.In the ORAC-FL assay, the essential oil result was 950.3 μmol TE/g, while TLC produced a yellow spot where the essential oil was applied, due to DPPH reduction, indicating significant antioxidant activity by both methods.The sample is considered active in the ORAC-FL test when antioxidant capacity values are > 800 μmol TE/g of sample 47,22 .
The lower ROS production, observed in our experiments, may be also related to the higher levels of the peroxisome proliferator-activated receptor-γ coactivator-1α (PGC-1α) in the dystrophic muscle cells treated with OCA.The   PGC-1α levels were reduced in the mdx group when compared to the Ctrl+OCA group (Fig. 5A,B).The OCA treatment increased the PGC-1α in the mdx+OCA group when compared to the mdx group (Fig. 5A,B).
The high increase of PGC-1α levels in dystrophic muscle cells treated with OCA is an interesting finding, since PGC-1α is a multifunctional regulatory factor, modulating several processes, such as inflammation and oxidative stress 48 .Although some work shows that PGC-1α overexpression leads to an upregulation of antioxidant enzymes, such as SOD2 in the skeletal muscle 49 , in our experimental conditions we did not observe an increase in the levels of the antioxidant enzyme system.Another possible way for PGC-1α to reduce oxidative stress is the regulation of RNA expression of uncoupling protein-3 (UCP3) 50 , since this protein has been related to oxidative stress reduction 51,52 .In addition, the PGC-1α can reduce oxidative stress through its antiinflammatory action.Elevated PGC-1α levels, linked to inflammatory response and oxidative stress reduction, were reported in the dystrophic experimental model 15,16 .In agreement with our results, other studies showed that the beneficial effects of D-limonene are related to the activation of peroxisome proliferator-activated receptor (PPAR)-α signaling, leading to an increase in the PGC-1α levels 37,53 .Thus, the PGC-1α pathway may be one of the mechanisms by which OCA reduced the oxidative stress and inflammatory response in our experimental conditions.

Conclusion
According to the results of this study, it is possible to conclude that the OCA can be considered a promising source of bioactive molecules for therapeutic applications in dystrophinopathies.The biological assays, reported in this investigation, show that OCA exhibited marked anti-inflammatory (TNF-α and NF-κB levels reduction) and antioxidant effects (4-HNE-protein adducts, SOD-2 and GSR levels and H 2 O 2 production decrease) on dystrophic muscle cells.These effects can be attributed to the presence of the bioactive monoterpenes in the essential oil, such as limonene which has been proven to have anti-inflammatory properties in several in vitro and in vivo experimental models.These anti-inflammatory effects on dystrophic muscle cells appear to be mediated through the activation of the PGC-1α pathway.

Figure 1 .Figure 2 .
Figure 1.Cell viability was assessed by measurement of MTT assay in Ctrl, Ctrl+OCA, mdx, and mdx+OCA groups.The values are means ± SD; n= 3

Figure 3 . 4 Figure 4 .
Figure 3. (A) Immunoblot analysis shows several bands of 4-HNE-protein adducts, ranging from 34 to 180 kDa, in Ctrl, Ctrl+OCA, mdx, and mdx+OCA groups.(B) Graphs showing 4-HNE-protein adducts level in experimental groups.Β-actin was used as a loading control.(C) Graphs show the quantification of H 2 O 2 production in the Ctrl, Ctrl+OCA, mdx and mdx+OCA groups.The values are means ± SD; n= 3. The means in each chart with different letters are significantly different at P< 0.05 according to Tukey's post-hoc test

Figure 5 .
Figure 5. (A) Immunoblot analysis of PGC1-α in Ctrl, Ctrl+OCA, mdx, and mdx+OCA (groups).(B) Graphs showing PGC1-α level in experimental groups.Β-actin was used as a loading control.The values are means ± SD; n= 3. The means in each chart with different letters are significantly different at P< 0.05 according to Tukey's post-hoc test

Table 1 .
Chemical composition of Citrus aurantifolia essential oil (OCA)