**Protective E**ff**ect of Tomato-Oleoresin Supplementation on Oxidative Injury Recoveries Cardiac Function by Improving** β**-Adrenergic Response in a Diet-Obesity Induced Model**

**Artur Junio Togneri Ferron 1,\*, Giancarlo Aldini 2, Fabiane Valentini Francisqueti-Ferron 1, Carol Cristina Vágula de Almeida Silva 1, Silmeia Garcia Zanati Bazan 1, Jéssica Leite Garcia 1, Dijon Henrique Salomé de Campos 1, Luciana Ghiraldeli 1, Koody Andre Hassemi Kitawara 1, Alessandra Altomare 2, Camila Renata Correa 1, Fernando Moreto <sup>1</sup> and Ana Lucia A. Ferreira <sup>1</sup>**


Received: 30 June 2019; Accepted: 17 August 2019; Published: 2 September 2019

**Abstract:** The system redox imbalance is one of the pathways related to obesity-related cardiac dysfunction. Lycopene is considered one of the best antioxidants. The aim of this study was to test if the tomato-oleoresin would be able to recovery cardiac function by improving β-adrenergic response due its antioxidant effect. A total of 40 animals were randomly divided into two experimental groups to receive either the control diet (Control, *n* = 20) or a high sugar-fat diet (HSF, *n* = 20) for 20 weeks. Once cardiac dysfunction was detected by echocardiogram in the HSF group, animals were re- divided to begin the treatment with Tomato-oleoresin or vehicle, performing four groups: Control (*n* = 6); (Control + Ly, *n* = 6); HSF (*n* = 6) and (HSF + Ly, *n* = 6). Tomato oleoresin (10 mg lycopene/kg body weight (BW) per day) was given orally every morning for a 10-week period. The analysis included nutritional and plasma biochemical parameters, systolic blood pressure, oxidative parameters in plasma, heart, and cardiac analyses in vivo and in vitro. A comparison among the groups was performed by two-way analysis of variance (ANOVA). Results: The HSF diet was able to induce obesity, insulin-resistance, cardiac dysfunction, and oxidative damage. However, the tomato-oleoresin supplementation improved insulin-resistance, cardiac remodeling, and dysfunction by improving the β-adrenergic response. It is possible to conclude that tomato-oleoresin is able to reduce the oxidative damage by improving the system's β-adrenergic response, thus recovering cardiac function.

**Keywords:** high sugar-fat diet; obesity; β-adrenergic system; cardiac dysfunction; lycopene; tomato-oleoresin

#### **1. Introduction**

Clinical studies show that the excessive body fat leads to many cardiac abnormalities, among them, morphologic and functional changes [1,2]. Animal studies have demonstrated myocardial dysfunction in obese rodents fed with hypercaloric diets [3–7]. Although it is evident that many cardiac changes and/or impairments in performance occur due to adipose tissue accumulation [3,8], the responsible mechanisms by which these changes are not clarified. The system redox unbalance, characterized by a high production of reactive species and inefficient antioxidant activity, is one of the pathways associated with the obesity-related cardiac dysfunction [9].

The β-adrenergic system is one of the most important mechanisms responsible for myocardial contraction and relaxation [10–12]. However, chronic expositions to reactive species are associated

with sustained adrenergic stimulation, resulting in arrhythmias and heart failure [13]. Considering the redox system's role in the pathogenesis of obesity and cardiac disorders, the use of antioxidants as therapeutic strategies has been tested [14,15].

Lycopene is a carotenoid present in tomato and red fruits and considered a potent antioxidant [16–18]. The tomato, and tomato product consumption are one of the Mediterranean diet's characteristics, which is associated with health benefits [18]. However, there is a lack of studies regarding lycopene dose ingestion in countries with a Mediterranean diet. Moreover, the few studies which bring information about lycopene consumption have a big variability [19] among the results (for example: in Italy the average intake is 7.4 mg per day while in Spain is 1.6mg per day [18]). The effect of lycopene on cardiovascular disease has been evaluated in clinical [20,21] and experimental studies [22]. Although obesity and oxidative stress are able to lead to cardiac dysfunction, no studies have evaluated the cardiac modulation by lycopene due the antioxidant effect. So, this study aimed to test if the tomato-oleoresin would be able to recovery the cardiac function by improving β-adrenergic response due its antioxidant effect.

#### **2. Materials and Methods**

#### *2.1. Animals and Experimental Protocol*

In the present study, male Wistar rats (±187 g) were initially divided into two experimental groups to receive control diet (Control, *n* = 20) or high sugar-fat diet (HSF, *n* = 20) for 20 weeks. The diets and water were provided ad libitum. The diet composition has been described in our previous studies [15,23]. All the animals were housed in an environmental controlled room (22 ◦C ± 3 ◦C, 12 h light-dark cycle and relative humidity of 60 ± 5%). All of the experiments were performed in accordance with the National Institute of Health's Guide for the Care and Use of Laboratory Animals and the procedures were approved by the Animal Ethics Committee of Botucatu Medical School (1196/2016).

At week 20 of this study, the cardiac dysfunction was detected by echocardiogram in the HSF group. Thus, the animals were casually divided to begin the treatment with tomato-oleoresin or vehicle, performing four groups: Control (*n* = 6); Control supplemented with lycopene- tomato oleoresin (Control + Ly, *n* = 6); HSF (*n* = 6) and HSF supplemented with lycopene- tomato oleoresin (HSF + Ly, *n* = 6). Tomato oleoresin was mixed with corn oil correspondent to 10 mg lycopene/kg of body weight (BW) per day and given orally every day, in the morning, for a 10-week period [24,25]. To avoid differences in the energy provided, all the groups received the same corn oil amount (about 2 ml/kg BW per day). The supplementation time and dose were chosen based in previous studies from our research group and others from the literature [24–26].

#### *2.2. Tomato-Oleoresin Preparation*

The tomato-oleoresin (Lyc-O-Mato 6% dewaxed; LycoRed Natural Products Industries, Beersheba, Israel) was mixed with corn oil and kept in the dark, at 4 ◦C, until the moment to be used [27]. The tomato oleoresin-corn oil mixture stayed for 20 min in a water-bath at 54 ◦C before the animals receive. The total amount of lycopene in each solution was 5mg/ml. Lycopene stability was confirmed by diode-array spectra at 450 nm, as previously described [28].

#### *2.3. Nutritional Evaluation*

Nutritional evaluation included: feed consumption (FC)-daily consumed amount in grams of chow feed; final body weight (BW); caloric intake (CI), calculating according to the following formula for the control group: caloric intake (kcal/day) = feed consumption (g) × dietary energy (3.59 kcal/g). For the HSF group, the caloric intake was calculated as following: water volume consumed (mL) × 0.25 (equivalent to 25% fructose) × 4 (calories per gram of carbohydrate) + caloric intake providing by the chow (feed consumption (g) × dietary energy (4.35 kcal/g).

Feed efficiency (FE) is defined as the ability to convert the caloric intake to body weight. It was calculated according to the formula: FE (%) = BW gain (g)/total caloric intake (kcal) × 100 [15,23]. The adiposity index, considered an obesity marker, was calculated as follow: adiposity index = (total body fat (BF)/final body weight) × 100. BF was evaluated considering the sum of the individual fat pad weights: BF = epididymal fat + retroperitoneal fat + visceral fat.

#### *2.4. Metabolic and Hormonal Analysis*

The plasma used for the biochemical analysis was collected after 12 h of fasting. The glucose levels were evaluated by a glucometer (Accu-Chek Performa; Roche Diagnostics, Indianapolis, IN, USA). The insulin levels were analyzed by ELISA assay with commercial kits (Millipore) [23]. The HOMA-IR (homeostatic model of insulin resistance), considered an insulin resistance index, was calculated by the following formula: HOMA-IR = [fasting glucose (mmol/L) × fasting insulin (μU/mL)]/22.5 [15].

#### *2.5. Systolic Blood Pressure (SBP)*

SBP was evaluated by a non-invasive tail-cuff method with a NarcoBioSystems® Electro-Sphygmomanometer (International Biomedical, Austin, TX, USA) with the conscious rats. For this, the animals were heat during 4–5min in a wooden box (50 × 40 cm), with two incandescent lamps and temperature between 38–40 ◦C, to induce arterial vasodilation in the tail. Then, the rats were transferred to an iron cylindrical support specially made to allow the total exposure of the animal's tail [29]. After this procedure, a cuff with a pneumatic pulse sensor was attached to the tail and inflated to 200 mmHg pressure and successively deflated. Blood pressure values were documented on a Gould RS 3200 polygraph (Gould Instrumental Valley View, Cleveland, OH, USA). The final SBP of each animal considered the average of three pressure readings.

#### *2.6. Lycopene Bioavailability Evaluation*

The presence of lycopene was determined in plasma and cardiac tissue homogenate. To extract the carotenoids, samples were incubated with internal standard (equinenone), chloroform/methanol CHCl3/CH3OH (3 mL, 2:1, *v*/*v*) and 500 mL of saline 8.5 g/L. Then the samples were centrifuged at 2000× *g* for 10 min and the supernatant was collected and hexane was added. The chloroform and hexane layers were evaporated under nitrogen and the residue was resuspended in 150 mL of ethanol and sonicated for 30 s. 50 μL of this aliquot was injected into the HPLC. The HPLC system was a Waters Alliance 2695 (Waters, Wilmington, MA, USA) and consisted of pump and chromatography bound to a 2996 programmable photodiode array detector, a C30 carotenoid column (5 μm, 150 × 4.6 mm, YMC-Yamamura Chemical Research, Wilmington, NC, USA), and Empower software (Empower 3, chromatographic data software Milford, MA, USA). The HPLC system programmable photodiode array detector was set at 450 nm for carotenoids. The mobile phase consisted of ethanol/methanol/methyl-tert-butyl ether/water (83:15:2, *v*/*v*/*v*, 15 g/L with ammonium acetate in water, solvent A) and methanol/methyl-tert-butyl ether/water (8:90:2, *v*/*v*/*v*, 10 g/L with ammonium acetate in water, solvent B). The gradient procedure, at a flow rate of 1 mL/min (16 ◦C), was as follows: (1) 100% solvent A was used for 2 min followed by a 6 min linear gradient to 70% solvent A; (2) a 3 min hold followed by a 10 min linear gradient to 45% solvent A; (3) a 2 min hold, then a 10 min linear gradient to 5% solvent A; (4) a 4 min hold, then a 2 min linear gradient back to 100% solvent A. For the quantification of the chromatograms, a comparison was made between the area ratio of the substance and area of the internal standard obtained in the analysis [30].

#### *2.7. Cardiac Malondialdehyde (MDA) Levels*

MDA is the main lipid peroxidation marker. It is considered an oxidative stress index [31] and associated with cardiovascular diseases [32]. Thus, MDA levels were used to evaluate the cardiac lipid oxidation as follow:

Cardiac tissue (±150 mg) was homogenized (ULTRA-TURRAX®T25 basic IKA® Werke Staufen/Germany) with 1.0 mL of cold phosphate buffered saline (PBS) pH 7.4, and centrifuged at 800 g at 4 ◦C for 10 min. Then, 100 μL from the supernatant was mixed with 700 μL of 1% orthophosphoric acid and 200 μL of thiobarbituric acid (42 mM). After this, the samples were kept at 100 ◦C for 60 min in a water bath, and immediately cooled on ice. In a 2 mL tube, 200 μL was mixed with 200 μL sodium hydroxide/methanol (1:12 *v*/*v*). After vortex, the samples were centrifuged for 3 min at 13,000 *g*. 200 μL from the supernatant was transferred to a glass vial and 50μL was injected into the column. The HPLC used was a Shimadzu LC-10AD system (Kyoto, Japan) with a C18 Luna column (5 μm, 150 × 4.60 mm, Phenomenex Inc., Torrance, CA, USA), and a Shimadzu RF-535 fluorescence detector (excitation 525 nm, emission 551 nm), and 0.5mL/min phosphate buffer flow (KH2PO4 1mM, pH 6.8) [25]. The MDA levels considered the peak area determination in the chromatograms relative to the standard curve of known concentrations.

#### *2.8. Circulating Advanced Oxidation Protein Products*

Advanced oxidation protein products (AOPPs) are oxidized plasma proteins resulting from the exposure to oxidation products and are transported by albumin in the circulation [33]. The literature reports that high AOPP circulating levels contribute to cardiac diseases [34].

AOPP determination was based on spectrophotometric detection according to Kalousova et al. [35]. Plasma samples (200 μL) were diluted 1:5 with PBS. It was also used 200 μL of chloramin T (0–100 μmol/L) for calibration curve and the blank was only PBS (200 μL). All the samples were put on a microtiter plate and mixed with 10 μL of KI 1.16 M and 20 μL of acetic acid. The absorbance was measured immediately at 340 nm (spectrophotometer Multiskan Ascent, Labsystems, Vantaa, Finland). The final AOPP concentration is expressed in chloramine units (μmol/L).

#### *2.9. Circulating Carboxymethyl Lysine*

Advanced glycation end products (AGEs) are a group of several molecules generated by both non-enzymatic glycation and protein, lipids and nucleic acids oxidation, able to modify tissue function and mechanical properties [36]. In vivo, CML is the main AGE associated with cardiac pathologies [37]. The plasmatic carboxymethyl lysine (CML) levels were evaluated using an ELISA commercial kit (OxiSelect™ CML, Cell Biolabs Inc., San Diego, CA, USA) following the manufacturer's instructions.

#### *2.10. Echocardiographic Study*

The analyze was performed in the live animals by transthoracic echocardiography, with a Vivid S6 system equipped with multifrequency ultrasonic transducer 5.0 to 11.5 MHz (General Electric Medical Systems, Tirat Carmel, Israel). The animals were lightly anesthetized by intraperitoneal injection with a mixture of ketamine (50 mg/kg) and xylazine (1 mg/kg), put in left decubitus position and only one examiner made all the exams. The heart image structural measurements were obtained in one-dimensional mode (M-mode) guided by the images in two-dimensional mode with the transducer in the parasternal position, minor axis. Left ventricular (LV) evaluation was performed with the cursor M-mode just below the mitral valve plane at the level of the papillary muscles. The aorta and left atrium images were obtained by positioning the M-mode course to plan the aortic valve level [23].

The following cardiac structures were evaluated: diastolic diameter (LVDD); systolic (LVSD) LV; left ventricle diastolic thickness posterior wall (LVPWD); aorta diameter (AD); left atrium (LA). The LV diastolic function was assessed by the transmitral flow early peak velocity (E). The LV systolic function was evaluated by ejection fraction and posterior wall shortening velocity (PWSV). The joint assessment of diastolic and systolic LV function was performed using the Tei index (sum of isovolumetric contraction and IRT time, divided by the left ventricular ejection time). The study was complemented by tissue Doppler evaluation, considering early diastolic (E'), and late (A') of the mitral annulus (arithmetic average travel speeds of lateral and septal walls), and the ratio by the waves E and E' (E/E').

#### *2.11. Myocardial Function by Isolated Papillary Muscle Study*

Besides echocardiographic analysis, myocardial function was also assessed by LV isolated papillary muscles. This procedure has been used by several authors [6,7,29]]. Conventional mechanical parameters at *Lmax* were calculated from isometric contraction: maximum developed tension normalized per cross-sectional area (DT [g/mm2]), resting tension normalized per cross-sectional area (RT [g/mm2]), positive (+d*T*/d*t* [g/mm2/s]) and negative (−d*T*/d*t* [g/mm2/s]) tension derivative normalized per cross-sectional area of papillary muscle (CSA).

#### *2.12.* β*-Adrenergic System Study*

β-adrenergic receptors (βAR) are important to regulate cardiac function in both normal and pathologic conditions [11]. The receptors activity was assessed by the dose-response relationship between the isoproterenol and conventional mechanical parameters of papillary muscle at *Lmax*. After baseline values determination, the isoproterenol was added to the vat in the presence of 1.0 mM [Ca2+] to increase progressively the concentrations for 10<sup>−</sup>8, 10−<sup>7</sup> and 10−<sup>6</sup> mol/L.

The stabilization of contractile response occurs nearly 3–5min after adding each isoproterenol dose. Data were sampled and expressed as the stimulation mean percent (%) [29]. At the end of the study, length (mm), weight (mg), and CSA (mm2) [38] were measured for papillary muscle characterization. The CSA was calculated from papillary muscle length and weight, assuming uniformity and a specific gravity of 1.0. The muscle length at *Lmax* was measured with a cathetometer (Gartner Scientific Corporation, Chicago, IL, USA), and the muscle between the two clips was blotted dry and weighed.

#### *2.13. Statistical Analysis*

The results are expressed in mean ± standard deviation (SD). Two-way analysis of variance (ANOVA) for independent samples was used to determine the differences among the groups. In order to evaluate the positive and negative inotropic effects on myocardial function, it was used a repeated-measures two-way ANOVA. Once detected significant differences (*p* < 0.05), the Tukey post hoc test for multiple comparisons were carried out. All the statistical analyses were performed using SigmaStat for Windows (Version 3.5, San Jose, CA, USA).

#### **3. Results**

The lycopene bioavailability is presented in the Table 1. It is possible to verify the presence of lycopene in both groups, which were supplemented (Control + Ly and HSF + Ly).

The HSF group presented increased caloric intake (kcal/d), final body weight (g), adiposity index, glucose levels, HOMA-IR and systolic blood pressure values compared to the control group. The HSF + Ly showed the same changes observed in HSF group when compared to control + Ly, except for HOMA-IR. Tomato-oleoresin suppressed the insulin resistance in HSF + Ly compared to HSF (Figure 1). No effect was observed of tomato-oleoresin on the other parameters.


#### **Table 1.** Lycopene Bioavailability.

Data are expressed in mean ± standard deviation (*n* = 4 animals/group). ND: Not detectable.

**Figure 1.** Nutritional and cardio- metabolic parameters. **A**—caloric intake (kcal/day); **B**—adiposity index (%); **C**—final body weight (g); **D**—glucose (mg/dL); **E**—HOMA-IR; **F**—systolic blood pressure (mmHg). Data are expressed in mean ± standard deviation (*n* = 6 animals/group). Comparison by Two-way ANOVA with Tukey post-hoc (*p* < 0.05): \* HSF *vs* Control; # HSF *vs* HSF + Ly; \$ HSF + Ly *vs* Control + Ly.

The HSF group presented cardiac remodeling (increased LVDS, LVPWD and reduced LVDD), and deterioration of both systolic (decreased ejection fraction, Tei-a and Tei-b) and diastolic (increased E/E- and decreased Tei-a and Tei-b) functions compared to control group. Regarding the tomato-oleoresin supplementation effect, HSF + Ly group showed improvement in some remodeling, systolic and diastolic parameters compared to HSF (Table 2).


**Table 2.** Echocardiographic study.

Data are expressed in mean ± standard deviation (*n* = 6 animals/group). Comparison by Two-way ANOVA with Tukey post-hoc (*p* < 0.05): \* HSF *vs* Control; # HSF *vs* HSF+Ly. LVDD, left ventricular diastolic diameter; LVSD, left ventricular systolic diameter; LVPWD, diastolic thickness posterior wall of the left ventricle; AD, aorta diameter; LA, left atrium diameter during ventricular systole; HR, heart rate; E, E-wave peak transmitral early diastolic inflow velocity; PWSV, posterior wall shortening velocity; Dec. time, deceleration time; Transmitral flow, Tei-a and Tei-b; EF, ejection fraction; E/E'.

The myocardial papillary muscle study at baseline condition with 2.5 mM Ca2<sup>+</sup> is presented in the Table 3. HSF group showed functional impairment in the maximum developed tension (DT) compared

to control group. Tomato-oleoresin supplementation was effective to recovery the DT capacity in HSF + Ly group compared to HSF (Table 3).


**Table 3.** Isolated papillary muscle at baseline condition (2.5 mM Ca2<sup>+</sup>).

Data are expressed in mean ± standard deviation (*n* = 6 animals/group). Comparison by Two-way ANOVA with Tukey post-hoc (*p* < 0.05): \* HSF *vs* Control; # HSF *vs* HSF+Ly. DT, Maximum developed tension normalized per cross-sectional area of the papillary muscle; RT, Resting tension normalized per cross-sectional area of the papillary muscle; peak of the positive, +d*T*/d*t* and negative, −d*T*/d*t* tension derivatives normalized per cross-sectional area of the papillary muscle; CSA, cross-sectional area.

Figure 2 shows the β-adrenergic stimulation on the papillary muscle function. The isoproterenol stimulation demonstrated that the HSF group presented functional impairment in DT (10−<sup>6</sup> M) and <sup>−</sup>d*T*/d*<sup>t</sup>* (10−<sup>7</sup> and 10−<sup>6</sup> M) compared to control group. Tomato-oleoresin supplementation was effective to recover the <sup>−</sup>d*T*/d*<sup>t</sup>* (10−<sup>7</sup> and 10−<sup>6</sup> M) capacity in HSF <sup>+</sup> Ly group compared to HSF.

**Figure 2.** β-adrenergic stimulation in papillary muscles. Data are expressed in mean ± standard deviation (*n* = 6 animals/group). Baseline calcium concentration (1.0 mM) is presented as 100%. A, Maximum developed tension normalized per cross-sectional area [DT, g/mm2]. B, positive [+d*T*/d*t*, <sup>g</sup>/mm2/s] and C, negative [−d*T*/d*t*, g/mm2/s] tension derivative normalized per cross-sectional area of the papillary muscle. Two-way ANOVA repeated-measures with Tukey post-hoc was used to compare the groups (*p* < 0.05); \* HSF *vs* Control; # HSF *vs* HSF + Ly.

Figure 3 shows the oxidative stress parameters in plasma and cardiac tissue. All the parameters increased in HSF group compared to control group. By contrast, it is possible to note that CML, AOPP and cardiac MDA plasma levels reduced in HSF + Ly group in respect to HSF to demonstrate a positive effect of tomato-oleoresin.

**Figure 3.** Plasma and cardiac tissue redox state parameters. **A**—Carboxymethyl lysine (CML-pg/mg protein); **B**—Cholaramine T equivalents %; **C**—Malondyhaldeide (MDA-nmol/μg protein). Data are expressed in mean ± standard deviation (*n* = 6 animals/group). Comparison by Two-way ANOVA with Tukey post-hoc (*p* < 0.05), \* HSF *vs* Control; # HSF *vs* HSF + Ly.

#### **4. Discussion**

This study aimed to test if the tomato-oleoresin would be able to recovery the cardiac function by improving β-adrenergic response due its antioxidant effect. The results show that the HSF groups presented with obesity, characterized by the higher values of body weight and adiposity index, and metabolic syndrome, with insulin resistance, dyslipidemia, and hypertension, all diseases usually associated with obesity [39]. These findings confirm that the diet model used in this study was efficient to lead obesity and related disorders, corroborating the literature [6,7,15,23]. Regarding the lycopene effect on obesity and related disorders, it was observed a positive action only on insulin resistance in the HSF + Ly group, represented by the reduction in HOMA-IR. The literature attributes the tomato-oleoresin benefic effects on diabetes to the lycopene antioxidant potential [40]. Another explanation for this amelioration is the anti-inflammatory effect of tomato-oleoresin. Since insulin resistance and type 2 diabetes are conditions closely related with inflammation and studies already showed that tomato-oleoresin ameliorates the inflammation, this property may explain the beneficial effect on glucose metabolism [41]. The antioxidant and anti-inflammatory effect of lycopene can also be explained by considering its well-established ability, through electrophilic metabolites, to activate Nrf2 pathway thus inducing phase II detoxifying/antioxidant enzymes and inhibiting NF-κB activation [42–44].

Obesity is also associated with cardiac abnormalities, among them morphological, hemodynamic and functional alterations [1,8,23]. Considering the lycopene absence effect on obesity and hypertension in the HSF + Ly group, should both HSF groups present cardiac damage. However, the echocardiographic analysis showed cardiac remodeling and impairment in ventricular systolic and diastolic function only in HSF group after 30 weeks. In opposition, the HSF group supplemented with tomato-oleoresin showed a cardiac remodeling and function recovery.

Several mechanisms could explain the obesity-induced cardiac dysfunction, among them is the β-adrenergic system responsiveness. The myocardial β-adrenergic mechanism is the main responsible by regulating the cardiac performance, especially by intracellular Ca2<sup>+</sup> handling [6,7]. Although functional studies using isolated papillary muscle have showed that obesity is able to lead to impairment in cardiac contractile [3,5,7], a small number of studies have evaluated the β-adrenergic response in high sugar-fat diet obesity-induced experimental models [29,45–49]. Our results demonstrated that the isoproterenol stimulation leaded to negative responses in both systolic (DT) and diastolic (−d*T*/d*t*) response in the HSF group while the HSF + Ly group showed an improvement in the β-adrenergic response. However, it is still unclear how the high sugar-fat diet obesity-induced leads to a reduction in the β-adrenergic response.

One hypothesis for the β-adrenergic response impairment is the chronic exposition to reactive oxygen species (ROS) promoted by obesity [9,50]. The literature reports that the direct contact with ROS exerts the same action of isoproterenol on β-adrenergic response, increasing the calcium transient amplitude, therefore, exerting a modulator role in the myocardial contractility [13]. However, this continues exposition to ROS may result in deleterious effects and contribute the development of cardiac arrhythmias and failure [9]. Considering the lycopene antioxidant effect, the amelioration in the β-adrenergic responsiveness of the HSF + Ly group can be attributed to this carotenoid property [14].

Another hypothesis is that the redox system imbalance in obesity conditions may lead to damage to lipids and proteins, generating such biomarkers as MDA, CML and AOPP, which were evaluated in this study [33,51–53]. These oxidative products can damage directly the cardiac tissue by altering its geometry and functionality, or indirectly by the carbonylation of proteins involved in the myocardial contractility regulatory response, as the β-adrenergic pathway [13,54,55]. While the HSF group presented higher levels of MDA, CML and AOPP and cardiac function deterioration, the tomato-oleoresin antioxidant effect is confirmed by reduced levels of these markers and cardiac function recovery in the HSF + Ly group.

#### **5. Conclusions**

In summary, this study found that the HSF diet induced obesity-related cardiac dysfunction and the tomato-oleoresin was able to attenuate this condition. Therefore, it is possible to conclude that tomato-oleoresin is able to reduce oxidative damage, thereby improving the system's β-adrenergic response and recovering cardiac function.

**Author Contributions:** Conceptualization, A.J.T.F., F.V.F.-F., G.A., F.M. and A.L.A.F.; methodology, A.J.T.F., F.V.F.-F., J.L.G., L.G., S.G.Z.B., K.A.H.K., C.C.V.d.A.S.; D.H.S.d.C. and A.A.; data curation, A.J.T.F., F.V.F.-F., S.G.Z.B., and A.A.; writing—original draft preparation, A.J.T.F.; F.V.F.-F.; C.R.C.; A.L.A.F.; writing—review and editing, F.V.F.-F., G.A., F.M. and A.L.A.F.; supervision, A.L.A.F.; project administration, A.J.T.F.

**Funding:** This work was funded by Conselho Nacional de Desenvolvimento Científico e Tecnológico CNPq (424209/2016-0) and Fundação de Amparo à Pesquisa do Estado de São Paulo FAPESP and Universidade Estadual Paulista "Julio de Mesquita Filho".

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Review* **Dietary Antioxidants and Parkinson's Disease**

#### **Han-A Park \* and Amy C. Ellis**

Department of Human Nutrition and Hospitality Management, College of Human Environmental Sciences, The University of Alabama, Tuscaloosa, AL 35487, USA; aellis@ches.ua.edu

**\*** Correspondence: hpark36@ches.ua.edu; Tel.: +1-205-348-8051

Received: 26 May 2020; Accepted: 26 June 2020; Published: 1 July 2020

**Abstract:** Parkinson's disease (PD) is a neurodegenerative disorder caused by the depletion of dopaminergic neurons in the basal ganglia, the movement center of the brain. Approximately 60,000 people are diagnosed with PD in the United States each year. Although the direct cause of PD can vary, accumulation of oxidative stress-induced neuronal damage due to increased production of reactive oxygen species (ROS) or impaired intracellular antioxidant defenses invariably occurs at the cellular levels. Pharmaceuticals such as dopaminergic prodrugs and agonists can alleviate some of the symptoms of PD. Currently, however, there is no treatment to halt the progression of PD pathology. Due to the nature of PD, a long and progressive neurodegenerative process, strategies to prevent or delay PD pathology may be well suited to lifestyle changes like dietary modification with antioxidant-rich foods to improve intracellular redox homeostasis. In this review, we discuss cellular and genetic factors that increase oxidative stress in PD. We also discuss neuroprotective roles of dietary antioxidants including vitamin C, vitamin E, carotenoids, selenium, and polyphenols along with their potential mechanisms to alleviate PD pathology.

**Keywords:** antioxidant; mitochondria; neurodegeneration; nutrient; apoptosis

#### **1. Introduction**

Parkinson's disease (PD) is a neurodegenerative disorder characterized by inadequate levels of dopamine that is caused by loss of dopaminergic neurons in the substantia nigra pars compacta (SNpc) of the basal ganglia. Dopamine also acts in other regions of the brain like the striatum, a substructure of the forebrain that regulates the motor system. Patients with PD exhibit motor symptoms including tremor, bradykinesia, rigidity, and speech difficulties, and also frequently suffer from nonmotor symptoms including depression and insomnia [1,2]. The incidence of sporadic PD is influenced by many factors including lifestyle, environment, age, and pre-existing conditions. Oxidative stress generated by many of these factors has been addressed as a major contributor to the development and progression of neurodegeneration at the cellular levels (Figure 1) [3–5]. In particular, mitochondrial dysfunction is a key finding in reactive oxygen species (ROS)-induced PD pathology [4–7]. Complex I, also known as NADH oxidoreductase of the electron transport chain (ETC) transfers electrons from NADH to ubiquinone and so plays a key role in oxidative phosphorylation. Complex I is vulnerable to oxidative damage, and its inhibition is also strongly associated with the generation of ROS such as superoxide and hydrogen peroxide presenting a positive feedback loop [8–10]. Currently, neurotoxins that target complex I like 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and rotenone are used to induce parkinsonism in both in vitro and in vivo models for research, and treatment with these drugs is known to induce oxidative stress [11,12]. In addition, imbalances in dopamine metabolism contribute to ROS generation, thus damaging dopaminergic neurons. Under normal physiological conditions, dopamine is synthesized from the amino acids tyrosine and tyramine. Hydroxylation and decarboxylation of tyrosine produce dopamine, and dopamine is further converted to norepinephrine and epinephrine or undergoes degradation. However, dopamine can also undergo metabolism by monoamine oxidase

(MAO) producing the highly reactive metabolite 3,4-dihydroxyphenylacetaldehyde (DOPAL) [13], and dopamine itself can undergo oxidation. Accumulation of DOPAL and oxidized dopamine increases the production of ROS damaging mitochondria [14–18].

**Figure 1.** Summary of the protective roles of dietary antioxidants in Parkinson's disease (PD). Both genetic and nongenetic factors contribute to the accumulation of oxidative stress by enhancing ROS production and impairing cellular antioxidant defense systems. Oxidative stress damages intracellular organelles, most notably the mitochondria, impairing neuronal energy metabolism and thus hindering the energy-demanding process in the brain including neurotransmission and neuritogenesis. Mitochondrial dysfunction primes apoptosis, calcium release, and opening of mtPTP which leads to the death of neurons, including the specific dopaminergic population of the SNpc which produces the signs and symptoms of PD. Illustration by BioRender.

#### **2. Oxidative Stress and** *PARK* **Genes**

Approximately 5-10% of PD cases are associated with abnormalities of *PARK* genes [19–21]. The mutation of *PARK* genes increases oxidative stress in neurons by enhancing the production of ROS or impairing intracellular antioxidant defense systems. ROS predisposes *PARK* genes to abnormal protein production and vice versa (Figure 1). α-synuclein is a protein encoded by the *SNCA (PARK1)* gene. Although α-synuclein supports synaptogenesis and synaptic plasticity during normal physiology [22,23], α-synuclein aggregation-associated neuronal damage is a common finding in PD affected tissue. Application of oligomeric α-synuclein increases ROS production and lipid peroxidation [24,25]. α-synuclein is translocated to mitochondria and directly interacts with mitochondrial proteins including ATP synthase to lower mitochondrial inner membrane potential, thus altering neuronal energy metabolism and opening mitochondrial death channels [24,26]. Increased oxidative stress induced by treatment with hydrogen peroxide or depletion of antioxidant enzymes enhances post-translational modification and aggregation of α-synuclein and worsens the progression of PD [27–29].

Parkin, the ubiquitin E3 ligase encoded by the *PARK2* gene, regulates mitochondrial quality control. Mutations of *PARK2* are the most common cause of early-onset PD. Approximately 77% of early-onset familial PD in patients younger than 30 years old have Parkin mutations [30]. Parkin works

in conjunction with the myocytes lacking PTEN-induced kinase 1 (PINK1), a key enzyme responsible for carrying out autophagy, encoded by the *PARK6* gene. PINK1-mediated phosphorylation of ubiquitin activates Parkin, and this enhances the removal of unwanted mitochondria [31]. Additionally, Parkin-mediated ubiquitination also targets mitofusin and miro, key components of mitochondrial fusion and mitochondrial transport, respectively [32,33]. The deletion of Parkin or PINK1 impairs mitophagy, and failure to remove dysfunctional mitochondria increases ROS production [34]. *PARK2* knockout transgenic mice treated with chronic ethanol, a stimulator of ROS, show abnormally high superoxide accumulation and glutathione depletion [35]. Application of the mitophagy inducer, autophagy and beclin 1 regulator 1 (AMBRA1), restores mitophagy in PINK1 knockout mice and suppresses ROS production [36]. Overall, the Parkin/PINK1 system plays a critical role in regulating intracellular ROS by mitochondrial quality control, removing inefficient and damaged mitochondria.

Protein deglycase DJ-1 is encoded by the *PARK 7* gene. Although the function of DJ-1 is less studied than other *PARK* gene products, it is reported to play an important role in supporting mitochondrial function. DJ-1 binds directly to F1Fo ATP synthase and the antiapoptotic protein Bcl-xL, and this interaction promotes mitochondrial energy metabolism and survival of dopaminergic neurons [37]. The depletion of DJ-1 increases the vulnerability of mitochondria to neurotoxic insults which mimic PD pathology [38,39], whereas overexpression of DJ-1 improves intracellular antioxidants and protects neurons [38,40,41]. DJ-1 plays an important role in sensing intracellular redox status during oxidative stress [42–44]. Under oxidative stress, DJ-1 undergoes post-translational oxidation at its Cys106 residue to form cysteine-sulfonic acid and cysteine-sulfinic acid [42,45,46], and oxidative stress also enhances translocation of DJ-1 to mitochondria. Thus, it is possible that oxidation of Cys106 may act as the signal for DJ-1 to prevent mitochondrial dysfunction during ROS production in PD. In addition, DJ-1 regulates the expression of antioxidant genes by promoting nuclear translocation of Nrf2, a transcription factor that binds to genes containing an antioxidant response element (ARE) [41,47].

LRRK2, also known as dardarin, is a kinase with guanosine triphosphatase (GTPase) and scaffolding domains [48]. LRRK2 is found in the mitochondrial membrane and interacts with other *PARK* gene products including Parkin, PINK1, and DJ-1 [49,50]. LRRK2 is encoded by the *LRRK2 (PARK8)* gene. Mutation of LRRK2 is associated with the gain of kinase activity, and this is common among patients with late-onset autosomal-dominant PD [51]. Mutation of the kinase domain of LRRK2, G2019S, exacerbates ROS-induced dopaminergic neuronal death, and application of truncated LRRK2 reverses ROS accumulation and prevents morphological alteration of these neurons [52]. In the same way, the depletion of LRRK2 or application of LRRK2 inhibitors decreases ROS, restores mitochondrial function, prevents mitochondrial fragmentation, and blocks increases in proapoptotic proteins including caspase 3, Bax, and apoptotic-inducing factor [53–55].

#### **3. Oxidative Stress and Mitochondrial Dysfunction**

Oxidative stress and mitochondrial dysfunction eventually lead to neuronal apoptosis during PD. Neurotoxic stimulation and ROS exposure increase the abundance of proapoptotic Bcl-2 protein Bax and Bak in the mitochondrial membrane (Figure 1). Oligomerization of proapoptotic proteins increases the permeability of the mitochondrial membrane causing the release of cytochrome c. Cytochrome c forms apoptosomes and activates executor caspases like caspase 3. Antiapoptotic proteins Bcl-2 and Bcl-xL are reported to block apoptosis by directly binding proapoptotic Bcl-2 proteins. Upregulation of proapoptotic proteins such as Bax and Bim as well as of other mechanisms including caspase activation and cytoplasmic release of cytochrome c have been reported in various PD models [54,56–60]. Transgenic mice lacking Bax are resistant to MPTP-induced neuronal death in the SNpc [56], and application of microRNA (miR) including miR216a and miR7 targeting Bax are protective against MPTP treatment in an in vitro and in vivo PD models [61,62]. Bcl-xL is an antiapoptotic protein that binds to DJ-1 and regulates energy metabolism in dopaminergic neurons [37,63]. A recent study shows that Bcl-xL undergoes post-translational cleavage during oxidative stress, and the accumulation of truncated Bcl-xL leads to mitochondrial dysfunction [64]. Approaches that inhibit proteolytic

cleavage of Bcl-xL are reported to be protective against neurotoxicity. Treatment with antioxidants prevents the accumulation of truncated Bcl-xL and rescues neurons from oxidative stress [64]. SH-SY5Y cells derived from human bone marrow that overexpress PINK1 show decreased proteolytic cleavage of Bcl-xL by enhancing phosphorylation of Bcl-xL [65]. Bcl-xL Cre-lox knockout mice show decreased tyrosine hydroxylase-positive cells indicating loss of dopaminergic neurons in the SNpc [66]; thus, maintaining functional Bcl-xL may be critical in preventing PD-associated neuronal death.

The association between neuronal death and opening of mitochondrial permeability transition pore (mtPTP), a large less-selective mitochondrial inner membrane death channel, has been documented in PD models [24,67,68]. The opening of mtPTP allows the passage of ions and small molecules less than 1.5KDa and depolarizes the mitochondrial inner membrane. mtPTP also enhances calcium release [24] which can trigger apoptosis (Figure 1) [69]. Loss of the mitochondrial electrochemical gradient impairs ATP production by the F1Fo ATP synthase and impairs neuronal energy metabolism [70–72]. The F1Fo ATP synthase plays a key role in ATP production and mPTP formation [72–74]. The F1Fo ATP synthase interacts with *PARK* gene products DJ-1 and α-synuclein [24,37]. Interaction between DJ-1 and F1Fo ATP synthase enhances neuronal energy metabolism and promotes elongation and arborization of dopaminergic neurons [37]. On the other hand, oligomeric α-synuclein co-localizes with the F1Fo ATP synthase and causes oxidative modification of its β subunit, the key subunit that interacts with ADP and ATP [24]. This oxidative modification increases the opening of mtPTP. Similarly, treatment with α-synuclein, known to form insoluble fibrils during PD pathology, favors mtPTP opening in both in vitro and in vivo models, and application of the mPTP inhibitor cyclosporin A reverses α-synuclein-induced mitochondrial dysfunction [60,75]. The depletion of PINK1 decreases mitochondrial inner membrane potential and increases the opening of mtPTP, and this leads to mitophagy and neuronal death [76,77].

#### **4. Neuroprotective Dietary Antioxidants**

Neurodegeneration at the cellular level develops years before patients exhibit clinical manifestations of PD. Therefore, finding strategies that can be applied over a lifetime seems of logical importance in fighting against PD. An increasing number of studies have addressed neuroprotective roles of nutrients and functional foods against neurodegeneration [78–80]. In particular, certain vitamins, minerals, and phytochemicals exhibit their antioxidant properties by directly scavenging ROS, binding to antioxidant enzymes as cofactors, and by regulating genes that control intracellular antioxidant systems (Figure 1). Advancing technologies in liquid chromatography and mass spectrometry such as LC/MS/MS and MALDI-TOF allow quantitative analysis of these nutrients and application of molecular approaches including sequencing, polymerase chain reaction, and electrophoresis to elucidate the association between *PARK* genes and dietary antioxidants. Here, we discuss dietary antioxidants that may potentially prevent or delay the progression of PD (Table 1).


**Table 1.** List of studies investigating the roles of antioxidant nutrients in PD models.

oleuropein (Ole).

#### *4.1. Vitamin C*

Vitamin C, also called ascorbic acid or ascorbate, is abundant in fruits and vegetables. Although most mammals are able to synthesize vitamin C endogenously, humans lack the necessary enzyme L-gulonolactone oxidase, so humans must ingest this essential nutrient in food or supplements [149]. Vitamin C acts as an antioxidant by donating electrons to neutralize the toxic effect of free radicals. Depending on available in vivo concentration, at high doses (≥500 mg/d), vitamin C has been shown to exhibit prooxidant properties [150]. In addition to its role in regulating cellular redox status, vitamin C supports the actions of hydroxylases involved in neurotransmitter synthesis including dopamine β-hydroxylase. Neural tissue including the brain contains high levels of vitamin C relative to other tissues, and neuroprotective roles of vitamin C have been discussed in various neurodegenerative disease models [151]. Treatment with divalent metal cations like copper and iron augment oligomerization of α-synuclein during challenge with DOPAL, a neurotoxic byproduct of dopamine metabolism [81], and treatment with vitamin C prevents α-synuclein oligomerization by inhibiting the oxidation of DOPAL [81] or interaction with copper [82]. A *Drosophila* model of PD shows increased oxidative stress with subsequent loss of dopaminergic neurons and locomotor deficits; treatment with vitamin C increased antioxidant enzyme activity and alleviated the PD-associated phenotype [99,127,152,153]. This model of PD is based on depletion of ubiquitin c-terminal hydrolase (UCH), an antioxidant enzyme, that thus enhances aging-associated degeneration of dopaminergic neurons and decreases dopamine content in the brain. The application of vitamin C (0.5 mM) compensates for these effects of UCH knockdown in *Drosophila* [152]. Vitamin C activates ten-eleven-translocation 1-3 (Tet1-3) enzymes and Jumonji C-domain-containing histone demethylases (Jmjds) [154]. These enzymes catalyze the formation of 5-hydroxymethylcytosine in DNA [155] and demethylation of lysine residues in histone, respectively. Tets and JmJds are required during the early stages of dopaminergic neuron differentiation, and treatment with vitamin C advances the development of neural stem cells derived from the embryonic midbrain [154]. Despite the protective roles of vitamin C found in in vitro and animal models, the efficacy of vitamin C against PD in humans is still controversial. Blood samples collected from PD patients show increased lipid peroxidation coupled with significantly lower levels of vitamin C compared to healthy controls [156], but some studies have also reported negligible effects of vitamin C on PD in human subjects [157]. Despite controversial results in human subjects, vitamin C may improve the therapeutic capacity of levodopa by enhancing its bioavailability and alleviating its toxic side effects [158,159].

#### *4.2. Vitamin E*

Vitamin E encompasses the tocopherols and tocotrienols found in plant sources including grains, legumes, vegetables, and seeds. Both tocopherols and tocotrienols have a chromanol ring and a hydrocarbon chain. Tocopherols have a saturated chain whereas tocotrienols contain double bonds. Vitamin E exhibits strong antioxidant properties by acting as a ROS scavenger, attenuating mitochondrial dysfunction, and preventing neuronal apoptosis during neurotoxic insults that mimic neurodegenerative disease [64,160]. Both tocopherol and tocotrienol bind to α-tocopherol transfer protein (TTP), a critical regulator of vitamin E movement and metabolism. α-tocopherol has an 8.5-fold higher affinity for TTP than α-tocotrienol [161], thus α-tocopherol is generally considered to have better bioavailability. However, studies are increasingly demonstrating that tocopherols and tocotrienols have varying roles in different tissues and microenvironments. For example, tocotrienols exhibit a stronger antioxidant capacity in lipid-rich biological membranes [162], thus tocotrienols may be effective in protecting lipid-rich organs like the brain [64,163]. Long-term intraperitoneal injection of α-tocopherol and the water-soluble analog Trolox improved long-term potentiation (LTP) and long-term depression (LTD) in PINK1 knockout mice [164]. Martella et al. report that chronic treatment with low concentration rotenone does not alter ATP production or viability of dopaminergic neurons in heterozygous PINK1 knockout (PINK1 +/ −) mice [165]. Despite this seemingly insignificant outcome, this treatment also completely impairs both LTP and LTD, and intraperitoneal injection

of α-tocopherol (100 mg/kg) and Trolox (5 mg/kg) reverse this synaptic plasticity impairment [165]. DJ-1 mutant flies show altered redox homeostasis as evidenced by high levels of global ROS and hydrogen peroxide production and decreased activity of catalase and superoxide dismutase [99]. However, supplementation with α-tocopherol decreases global ROS levels in DJ-1 mutant flies [99]. In addition to tocopherols, tocotrienols protect neurons against oxidative stress-associated damage. Primary hippocampal neurons treated with α-tocotrienol show a decrease in total and mitochondrial ROS accumulation, and α-tocotrienol attenuates glutamate-induced post-translational cleavage of Bcl-xL to enhance the functions of antiapoptotic Bcl-xL [166,167]. In this study, α-tocotrienol was suggested to exert its effect by blocking the oligomerization of proapoptotic Bcl-2 proteins [167]. Oral administration of 100 μg/kg δ-tocotrienol prevents the loss of dopaminergic neurons in the SNpc and improves motor behavior in a mouse model of PD [128]. δ-tocotrienol binds to the estrogen receptor β and activates PI3K/Akt signaling pathways including phosphorylation of protein kinase B (PKB, Akt) and extracellular signal-regulated kinase (ERK) 1/2 [128,168]. Akt activates Nrf2 [169,170], and Nrf2-mediated upregulation of antioxidant and prosurvival genes is an important mechanism for the neuroprotective properties of many antioxidant nutrients [171–173]. Clinical studies with PD patients show that higher consumption of dietary vitamin E is inversely related to PD occurrence [157,164,174,175]. However, contrary reports have also been published on PD in human subjects [176,177]. Data from randomized controlled trials with vitamin E are limited. However, in a randomized double-blind placebo-controlled trial, Taghizadeh et al. reported significant improvement in clinical symptoms as assessed by the Unified Parkinson's Disease Rating Scale (UPDRS) among PD patients who received 400 IU of vitamin E in combination with 1000 mg of omega-3 fatty acids [178]. These researchers also reported increases in circulating glutathione and total antioxidant capacity along with decreased high-sensitivity C-reactive protein with treatment compared to placebo. Although promising, further investigation into the specific roles of vitamin E subgroups will be important to clarify the efficacy of vitamin E in clinical disease.

#### *4.3. Vitamin A and Carotenoids*

Vitamin A is a fat-soluble vitamin found in both animal (e.g., liver) and plant sources and can also be produced from provitamin A carotenoids. Vitamin A exists as multiple forms: retinol (alcohol), retinal (aldehyde), retinoic acid (carboxylic acid), and retinyl ester (ester form). Retinal binds to opsin and activates rhodopsin, a G-protein coupled receptor that senses light in the eye. Retinoic acid binds to nuclear receptors including retinoic acid receptor (RAR) and retinoid X receptor (RXR) and regulates transcription of genes that control growth and differentiation [179]. In addition to these roles, vitamin A exhibits neuroprotective properties against neurodegeneration. Retinoic acid promotes differentiation of GABAergic neurons expressing dopamine receptors [132,133,179], and changes in PD include inhibition of retinoic acid-mediated neuronal differentiation [180]. Oral supplementation with retinoic acid upregulates the μ-type opioid receptor (MOR1), a G-protein-coupled receptor that mediates inhibitory signaling, in the dorsal striatum and attenuates repetitive dyskinetic movements in PD mice [181].

Carotenoids include the yellow, orange, and red pigments found in fruits and vegetables like carrots, tomatoes, watermelons, and pumpkins, and are also found in algae, salmon, and shrimp. Examples of carotenoids include carotene, lycopene, lutein, and astaxanthin. Serum α-carotene, β-carotene, and lycopene levels are significantly decreased in PD patients, and decreased serum carotenoid levels are also associated with poorer motor function [174,182]. However, a meta-analysis that examined the association between PD and vitamin A and carotenoids (lutein, α-carotene, β-carotene, lycopene, β-cryptoxanthin, zeaxanthin and canthaxanthin) concluded that the evidence was insufficient to make an epidemiological association between vitamin A/carotenoids and risk of developing PD [183]. In an in vivo animal model, oral administration of lycopene (5–20 mg/kg) attenuates oxidative stress induced by intraperitoneal injection of MPTP in mice, and lycopene also inhibits apoptosis by decreasing Bax and caspases while increasing Bcl-2 [129]. Treatment with lutein prevents MPTP-induced Bax and

caspase increases, and lutein also improves motor function in MPTP challenged mice [130]. Astaxanthin lowers intracellular ROS and improves superoxide dismutase and catalase activity, and treatment with astaxanthin prevents apoptotic death in MPTP challenged SH-SY5Y cells [100]. Astaxanthin attenuates MPTP-induced neuronal injury via the downregulation of α-synuclein [83]. miR-7 directly binds to the 3- UTR of α-synuclein mRNA and decreases the translation of α-synuclein [184]. Treatment with astaxanthin prevents the loss of miR-7 to lower the toxic effects of α-synuclein in SH-SY5Y cells [83]. Although clinical trials are lacking, oral supplementation with astaxanthin prevents loss of neurons in the SNpc and tyrosine hydroxylase-positive cells in the striatum from intraperitoneally injected MPTP in mice [131].

#### *4.4. Selenium*

Selenium is an essential trace mineral-rich in Brazil nuts, seafood, and organ meats and is also found in water and soil. The selenium content of plants is directly related to the selenium content of the soil [185]. Enzymes that regulate intracellular redox status likes glutathione peroxidase and thioredoxin reductase are selenoproteins that require selenium at their active sites, and mutations of the selenocysteine residues impair enzyme activity [186]. Microarray investigation reveals that rotenone treatment downregulates the *SELENBP1* gene which encodes selenium binding protein 1, along with other genes that control apoptosis and mitochondrial function [187]. Neuroprotective functions of the selenium-containing quinoline derivative, 7-chloro-4-(phenylselanyl) quinoline, against the rotenone challenge highly correlates with selenium content in the brain of fruit flies [188]. Intraperitoneal delivery of selenium selenite (0.1, 0.2, and 0.3 mg/kg) increases glutathione peroxidase activity, alleviates lipid peroxidation, and improves motor function of the 6-hydroxydopamine challenged striatum in rats [101]. Interestingly, selenium treatment also shows dose-dependent protection of other antioxidant enzymes including glutathione reductase, glutathione transferase, and catalase [101]. Intraperitoneal injection of selenium partially improves dopamine metabolism during the MPTP challenge [189]. Analysis of soil samples from 4856 sites in the US demonstrates that higher selenium content inversely correlates with mortality from PD [190]. Human studies investigating selenium supplementation for PD are lacking. However, low plasma selenium concentrations are associated with decreased performance in neurological tests among older adults [191]. Conversely, increased levels of selenium in cerebrospinal fluid and plasma have been reported in PD patients [192,193]. Chronic exposure to selenium enhances oxidative stress in the brain and leads to cognitive impairment in animal models [194,195]. The underlying mechanism for these findings is unclear; however, evidence suggests that either a deficiency or excess of selenium may contribute to neurodegeneration or conversely PD pathology may impair mobilization of selenium in neurons. The Recommended Dietary Allowance for selenium is 55 mg/day, and the Institute of Medicine has established a Tolerable Upper Intake Level for selenium at 400 mg/day. Therefore, meeting the RDA without excess may be prudent [196].

#### *4.5. Glutathione*

Glutathione is a tripeptide of glycine, cysteine, and glutamate that is widely present in both plant and animal foods. In particular, avocados, asparagus, spinach, and amino acid-rich meat, fish and poultry are good sources of glutathione. Glutathione is a major intracellular antioxidant that reduces reactive oxygen species by being oxidized to glutathione disulfide. Glutathione is required by glutathione peroxidase during the conversion of hydrogen peroxide to water. The depletion of glutathione leads to oxidative stress-induced mitochondrial dysfunction and degeneration of dopaminergic neurons [125,134,197]. Interestingly, excess of glutathione also causes neuronal damage [134], and this may be due to the overproduction of glutathione disulfide, an oxidized form of glutathione responsible for mitochondrial dysfunction and neuronal death [198]. Strategies to support glutathione homeostasis by preventing loss of glutathione or facilitating clearance of glutathione disulfide protect the brain [163,198]. Treatment with glutathione's precursor N-acetylcysteine (NAC) prevents oxidative stress and calcium overload and rescues neurons and other brain cells during PD-like

stress [102,112,135]. Consistently, a protective effect of intravenous and oral delivery of NAC has been reported in PD patients [199–201]; NAC is naturally found in onions and garlic, and it is available in various dosages as an over-the-counter dietary supplement [202]. However, the best duration and concentration of supplementation to consistently show a therapeutic effect in humans has not been established [200,203]. Therefore, further investigation is required. Additionally, since oral glutathione is less bioavailable, finding nutrients that enhance the body's ability to synthesize glutathione may also be of benefit.

#### **5. Polyphenols**

Polyphenols are characterized by the presence of multiple phenol groups and a six-membered hydrocarbon ring structure. Based on the arrangement of phenol groups, hydrocarbon chain and additional functional groups, polyphenols are further classified into subgroups including flavonoids, isoflavonoids, curcuminoids, tannins, and stilbenoids. There are estimated to be over 8000 different polyphenols present in nature [204]. We will describe four well-investigated polyphenols—curcumin, resveratrol, catechin, and oleuropein—and their role in PD models.

#### *5.1. Curcumin*

Curcumin, 1,7-bis(4-hydroxy-3-methoxyphenyl)-1,6-heptadiene-3,5-dione is a polyphenol found in turmeric. Curcumin scavenges biological radicals including superoxide anion, hydrogen peroxide, 1,1-diphenyl-2-picryl-hydrazyl free radical, 2,2- -azino-bis (3-ethylbenzthiazoline-6-sulfonic acid) radical, and *N*,*N*-dimethyl-*p*-phenylenediamine dihydrochloride radical [205]. In addition, treatment with curcumin (10 μM) decreases oxidation-associated protein modification including carbonylation and nitrotyrosine formation to rescue dopaminergic cells [106]. Curcumin effectively protects mitochondria from oxidative stress-associated damage [206]. Curcumin (2 μM) prevents loss of mitochondrial membrane potential and electron transfer system capacity in SH-SY5Y cells depleted with PINK1 [126]. Similarly, treatment with curcumin monoglucoside (0.25–5 μM) restores mitochondrial complex I and IV activity by decreasing the accumulation of hydroperoxides and increasing glutathione levels [103]. Curcumin exhibits antiapoptotic properties. Treatment with curcumin (5 μM) decreases ROS-induced calcium influx, lowering activation of caspase 3 and caspase 9 [104]. In addition, curcumin interferes with prodeath JNK signaling to prevent downstream apoptotic pathways including the release of cytochrome c and cleavage of procaspase 3 [103,138]. In vivo studies demonstrated antioxidant [105,144] and antiapoptotic [87,136] effects of curcumin to improve PD-associated neurobehavior [103,105,107,143–146]. Intraperitoneal injection of curcumin (200 mg/kg) attenuates rotenone-induced motor impairment in rats [143]. Male Wistar rat orally administered 5–20 mg/kg demethoxycurcumin, a derivative of curcumin, show concentration-dependent protection against rotenone challenge [105]. Demethoxycurcumin attenuates rotenone-induced oxidative stress and prevents loss of dopamine in the brain [105], and animals treated with demethoxycurcumin show improved motor function [105]. Dietary supplementation with 0.5% and 2% curcumin also show similar effects on MPTP-induced mouse PD models [137]. In addition to neuroprotection, curcumin may regulate cell differentiation and proliferation. C57BL mice transplanted with curcumin-activated mesenchymal stem cells have increased antiapoptotic Bcl-2, decreased proapoptotic Bax and caspases, and avoided the loss of dopaminergic neurons during MPTP challenge [136]. Curcumin prevents α-synuclein aggregation [87] and attenuates α-synuclein-induced cytotoxicity [85]. Curcumin derivative increases the nuclear translocation of transcription factor EB, a regulator of autophagy, potentially promoting degradation of α-synuclein [86].

#### *5.2. Resveratrol*

Resveratrol, 3,5,4- -trihydroxy-*trans*-stilbene is a nonflavonoid polyphenol with two aromatic ring structures. Resveratrol is found in grapes and berries, and it is also commonly consumed in red wine. Resveratrol promotes brain cell differentiation and proliferation during normal physiology [207], and it is well-described to attenuate oxidative stress-associated damage during the progression of PD pathology [108–112,114]. Intraperitoneally administered resveratrol (20 mg/kg) decreases lipid peroxidation, increases glutathione levels, and prevents deterioration of rat SNpc against 6-hydroxydopamine, an oxidant that causes degeneration of dopaminergic neurons [108]. Various research groups have shown that resveratrol effectively protects mitochondria by decreasing the accumulation of mitochondrial ROS, preventing mitochondrial inner membrane potential loss, restoring mitochondrial respiratory enzyme activity, regulating mitochondrial fission and fusion, and protecting mitochondrial DNA in *PARK2* mutation [109–111,113]. Wang et al. showed that resveratrol treatment (25 μM) increases phosphorylation of Akt and prevents rotenone-induced death of PC12 cells [111]. Akt upregulates genes containing cAMP response element (CRE) including Bcl-2 [208,209], and it inactivates proapoptotic Bad and proteolytic caspases [210]. Thus resveratrol-mediated Akt phosphorylation may hinder apoptotic death during PD-like challenges. In addition, resveratrol may alleviate *PARK* gene-associated PD pathology. Male C57BL/6 mice subjected to intragastric gavage of 100 mg/kg resveratrol attenuate the loss of dopaminergic neurons and have improved motor behavior during the MPTP challenge [88]. This same study also shows that resveratrol significantly increases protein levels of LC3-II, a key protein found in the membrane of autophagosomes, and thereby facilitates degradation of α-synuclein [88]. Resveratrol also increases microRNA-214 which potentially inhibits translation of α-synuclein [89]. Fibroblasts isolated from patients with *PARK2* mutations have increased production of whole-cell ROS and mitochondrial ROS, and treatment with resveratrol protects mitochondria and improves respiration and ATP production in these cells [109].

#### *5.3. Catechin*

Catechins are flavonoids containing two benzene rings and one dihydropyran heterocycle. Catechins are found in various herbs and fruits. Tea in particular is a good source of catechins. Four major catechins include (−)-epicatechin (EC), (−)-epicatechin-3-gallate (ECG), (−)-epigallocatechin (EGC), and (−)-epigallocatechin-3-gallate (EGCG) [211]. Catechins donate an electron from a phenolic hydroxyl group and to scavenge free radicals and thus exhibit direct antioxidant properties [212–214]. Catechins also improve intracellular redox status by preventing the loss of other antioxidants [116]. Treatment with 10 μM EGCG lowers the accumulation of ROS and prevents activation of caspases during hydrogen peroxide challenge and protects N27 dopaminergic cells from apoptotic death [119]. Koch et al. show that a longer brewing time tends to enhance antiradical activity in teas [215] indicating that catechins retain antioxidant properties after exposure to high temperature. Although further investigation is needed, orally supplemented catechins are shown to be delivered to the brain (0.5 nmol/g) in rats [216] and an in vitro blood–brain barrier system (BBB) shows that <10% of catechins are BBB permeable [217,218]. Various research groups have demonstrated that EGCG prevents neurotoxicity associated with α-synuclein [91,92,94,95]. EGCG chelates metal ions including Cu(II) and Fe(III) to inhibit fibrillation of α-synuclein [90,93]. EGCG (350μM) enhances the formation of stable oligomers (a less-toxic form) thus prevents the accumulation of pathological fibril [95] EGCG immobilizes α-synuclein and interferes with its oligomerization in biological membranes [96], thus EGCG helps to maintain membrane integrity [95,96]. EGCG suppresses fibrillation of γ-synuclein, a type of synuclein also found in Lewy bodies [91]. EGCG improves motor behavior in *Drosophila* by preventing mitochondrial dysfunction caused by abnormalities of LRRK2 and Parkin genes [148]. Chemically induced rodent PD models produced by injection with MPTP and 6-hydroxydopamine demonstrate PD-like symptoms like bradykinesia, and administration of 10-50 mg catechin (both oral and intraperitoneal injection) improves locomotor behavior in these animals [115,116,139]. Intraperitoneal injection of 10 or 30 mg/kg catechin restores glutathione levels and increases dopamine in the rat brain [116]. Oral supplementation with 25 mg EGCG reduces oxidative stress and preserves striatal dopamine in C57BL/6J mice challenged with MPTP [115]. C57BL/6J mice intraperitoneally injected with MPTP demonstrate PD-like symptoms including bradykinesia due to loss of SNpc dopaminergic neurons, and oral administration of EGCG (25 and 50 mg/kg) in these animals

lowers proinflammatory cytokines, rescues dopaminergic neurons from death, and improves motor behavior [139]. In addition to catechins' role inhibiting PD pathology, catechins may also support existing PD treatments. Orally administered EGCG (100 and 400 mg/kg) inhibits methylation of levodopa to improve bioavailability [219].

#### *5.4. Oleuropein*

Oleuropein contains hydroxytyrosol, elenolic acid, and glucose. It is a major phenolic compound found in olive oil. Although oleuropein is predominant, other oleuropein derivatives such as oleuropein aglycon and oleuroside are also found in olive oil [220]. Oleuropein acts as a scavenger of superoxide, nitric oxide, 2,2- -azinobis-(3-ethylbenzothiazoline-6-sulfonic acid, and 2,2-diphenyl-1-picrylhydrazyl radicals [221,222]. Various research groups have demonstrated that treatment with oleuropein and its derivatives inhibit the accumulation of ROS and prevent the progression of PD pathology [97,98,120]. Palazzi et al. demonstrated that in vitro incubation with oleuropein aglycone stabilizes α-synuclein monomers to prevent pathological aggregation [97]. Similarly, Mohammad-Beigi et al. show that olive fruit extracts containing oleuropein and oleuropein aglycone inhibit α-synuclein fibril elongation, decreasing cytotoxicity caused by α-synuclein oligomers [98]. In addition, oleuropein activates redox-sensitive transcription factors like Nrf2 to potentially improve intracellular antioxidant capacity via the upregulation of antioxidant genes [120,223]. Oleuropein protects mitochondria by mitigating mitochondrial superoxide production [121]. PC12 cells treated with 1-50 μM oleuropein retain mitochondrial membrane potential during the 6-hydroxydopamine challenge, and oleuropein also alleviates endoplasmic reticulum stress to protect PC12 cells from apoptotic death [141]. Oleuropein increases mitochondrial antiapoptotic Bcl-2 and decreases proapoptotic Bax and apoptotic-inducing factor [121,142]. Furthermore, oleuropein regulates phosphorylation of dynamin-related protein 1 (Drp1) [142] and LC3-II [121], key proteins that control mitochondrial fission and mitophagy, respectively. Thus, oleuropein potentially supports an optimal mitochondrial population in cells. Oral supplementation with olive leaf extract (75–300 mg/kg) significantly increases antioxidant enzymes including superoxide dismutase and glutathione peroxidase in the rat brain [123]. Rats fed with olive leaf extract are protected from loss of dopaminergic neuron during rotenone-induced mitochondrial damage, and showed improved neurobehavior [123]. Similarly, rats supplemented with extra virgin olive oil extract show decreased lipid peroxidation and increased antioxidant enzyme activities [124]. Oral administration of oleuropein is distributed to the brain 2h after ingestion [224], so oleuropein may be a key component in olive leaf and olive oil-mediated neuroprotection.

#### **6. Conclusions**

Although increasing numbers of studies performed in vitro and using animal models demonstrate a potential role in dietary prevention of PD, the efficacy of nutritional intervention to do so in humans remains controversial. Epidemiological studies examining dietary intake of antioxidant micronutrients and the risk of developing PD have yielded equivocal results, and there is a paucity of data from randomized controlled trials among people with pre-existing PD. Dietary antioxidants exhibit multiple effects rather than targeting a single specific process. Vitamin C, vitamin E, and polyphenols directly interact with ROS and terminate oxidative chain reactions. Other minerals like selenium act as cofactors to support the activity of antioxidant enzymes. Many antioxidant nutrients are involved in signaling transduction and protect downstream targets of oxidative stress to alleviate the damage that promotes the development of PD. Nutrients also regulate genes that control the development, growth, and survival of dopaminergic neurons. Polyphenols like curcumin, resveratrol, catechin, and oleuropein inhibit the formation of Lewy bodies. In this review, we have described the complex cellular and molecular mechanisms of these dietary antioxidants as an important step in developing a therapeutic strategy against PD. Future clinical studies with data safety and monitoring are warranted to determine whether these antioxidant micronutrients may act individually or in synergy as a nonpharmacological means of prevention and treatment.

**Author Contributions:** Writing and editing: H.-A.P. and A.C.E. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Antioxidants* Editorial Office E-mail: antioxidants@mdpi.com www.mdpi.com/journal/antioxidants

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18