**Food Bioactives: Impact on Brain and Cardiometabolic Health—Findings from In Vitro to Human Studies**

Editors

**Nenad Naumovski Domenico Sergi**

MDPI • Basel • Beijing • Wuhan • Barcelona • Belgrade • Manchester • Tokyo • Cluj • Tianjin

*Editors* Nenad Naumovski School of Rehabilitation and Exercise Sciences, Faculty of Health, University of Canberra Australia

Domenico Sergi Department of Translational Medicine and for Romagna, University of Ferrara Italy

*Editorial Office* MDPI St. Alban-Anlage 66 4052 Basel, Switzerland

This is a reprint of articles from the Special Issue published online in the open access journal *Foods* (ISSN 2304-8158) (available at: https://www.mdpi.com/journal/foods/special issues/Food Bioactives Cardiometabolic Implications).

For citation purposes, cite each article independently as indicated on the article page online and as indicated below:

LastName, A.A.; LastName, B.B.; LastName, C.C. Article Title. *Journal Name* **Year**, *Volume Number*, Page Range.

**ISBN 978-3-0365-4041-2 (Hbk) ISBN 978-3-0365-4042-9 (PDF)**

© 2022 by the authors. Articles in this book are Open Access and distributed under the Creative Commons Attribution (CC BY) license, which allows users to download, copy and build upon published articles, as long as the author and publisher are properly credited, which ensures maximum dissemination and a wider impact of our publications.

The book as a whole is distributed by MDPI under the terms and conditions of the Creative Commons license CC BY-NC-ND.

## **Contents**


## **About the Editors**

#### **Nenad Naumovski**

Nedad Naumovskiis a food scientist and molecular nutritionist and works at the University of Canberra (ACT, Australia) as Associate Professor in Food Science and Human Nutrition. He leads a Functional Foods and Nutrition Research Laboratory (FFNR Laboratory) and holds academic conjoint positions as the Visiting Professor at the Harokopio University of Athens (Athens, Greece) and University of Newcastle (NSW, Australia). He has presented the findings of his research group at several academic and scientific societies as an invited guest speaker and provided number of plenary talks including prestigious Australian Academy of Science. Dr Naumovski has a strong research interest in the development of functional foods and the effects of food and nutrients on psycho-cardiological markers associated with healthy ageing. Currently, he leads several projects relating to the flavor augmentation and consumption of different plant bioactives on the reduction of stress and anxiety, improvements in focus, attention and quality of sleep.

#### **Domenico Sergi**

Domenico Sergi is a nutrition scientist with a PhD in Human Nutrition pursued at the Rowett Institute (University of Aberdeen, UK) in 2016 under the supervision of Prof. Lynda M. Williams, Dr Janice Drew, Dr Iain Greig and Dr James Hislop. In February 2017 he moved to Maastricht where he was appointed as a postdoctoral researcher within the Diabetes and Metabolism Research Group (University of Maastricht) to study the effect of pharmacological and nutritional compounds on insulin sensitivity, substrate metabolism and mitochondrial function in human primary myocytes and skeletal muscle cell lines. In September 2017 he moved to Adelaide (Sounth Australian Health and Medical Research Institute) where he was appointed as a postdoctoral research fellow at the Commonwealth Scientific and Industrial Research Organisation (CSIRO) to investigate novel approaches in nutrition for health. After which he moved to Canada to start his first tenure-track position as a professor in the department of medical biology at the "Universite du Qu ´ ebec ´ a` Trois-Rivieres". Domenico now holds a tenure-track position (RTDb) in Nutrition and Dietetics in ` the Department of Translational Medicine at University of Ferrara (Italy). Domenico is interested in the tight relationship between nutrition and health and how nutrients and specific dietary patterns can affect human health, with particular focus upon metabolic health. His main interest lies in the mechanisms underpinning the development of obesity and its comorbidities, with a particular focus on metabolic inflammation and lipotoxicity.

### *Editorial* **Food Bioactives: Impact on Brain and Cardiometabolic Health—Findings from In Vitro to Human Studies**

**Nenad Naumovski 1,2,3,\* and Domenico Sergi 4,5**


Modern society is currently (and probably more than ever) immersed in the changing concept of food, seeking the beneficial functions of foods rather than only as a mean to quench hunger and support basic nutritional needs. In this context, we are facing a change in the expectations that consumers have from food items, accompanied by an increased attention towards food bioactive derivatives with health boosting properties. These emerging perceptions of food as a key discriminant in human health are fueled by the already strong evidence linking unhealthy dietary patterns with the onset and progression of several chronic diseases, ranging from type 2 diabetes mellitus (T2DM) to cancer and neurodegenerative diseases. On the contrary, functional foods and their bioactive components may represent a nutritional cornerstone to improve the quality of diet and ameliorate or prevent (in some cases) nutrition-related diseases. Bioactives are unlike pharmaceuticals (compounds used to alleviate symptoms and cure disease). Nevertheless, the latest findings indicate that the clear gap between the two products (bioactives and pharmaceuticals) is becoming narrower and in some cases, they are becoming interchangeable.

In agreement with the aforementioned considerations, the interest of the general population with respect to functional foods containing bioactive molecules is in constant expansion, which provides an impetus for research in this field. Indeed, several studies, including in vitro investigations, clinical trials and observational studies related to food and dietary patterns have already identified, proposed and in some cases, challenged the mechanisms of action of food bioactive derivatives. Therefore, the main aim of this Special Issue was to provide an opportunity to bring together high-quality manuscripts that showcase the current knowledge in relation to food bioactives and their impact on brain and cardiometabolic health.

The article by Deutch et al. (2019) is a comprehensive work identifying the compositional properties of licorice and the potential impacts on blood pressure and the cardiovascular system [1]. Licorice is the root of the legume *Glycyrrhiza glabra* that is commonly grown in warm climatic areas such as the Middle East, Asia and southern parts of Europe. For several millennia, this root was used in traditional medicines of many countries as an ailment for a number of different diseases and heath conditions, such as gastrointestinal symptoms and respiratory diseases. Nowadays, a broad-spectrum of health-related properties have been ascribed to licorice, including immunostimulatory effects; anti-ulcer, anti-cancer, anti-viral and anti-microbial effects; in addition to the protection of the nervous and cardiovascular systems. Although licorice consists of over 300 potential bioactive compounds, the authors report that the health effects elicited by

**Citation:** Naumovski, N.; Sergi, D. Food Bioactives: Impact on Brain and Cardiometabolic Health—Findings from In Vitro to Human Studies. *Foods* **2021**, *10*, 1045. https://doi.org/ 10.3390/foods10051045

Received: 7 May 2021 Accepted: 10 May 2021 Published: 11 May 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 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 (https:// creativecommons.org/licenses/by/ 4.0/).

licorice mainly rely on the bioactive glycyrrhizin. This molecule is a prodrug that is converted into 3β-monoglucuronyl-18β-glycerrhetinic acid (3MGA) and 18β-glycerrhetinic acid (GA) in the small intestine. Despite both compounds having been associated with a variety of potential health benefits, 3MGA and GA can also inhibit the hydroxysteroid dehydrogenase II enzyme, which is responsible for oxidizing cortisol into cortisone. Therefore, high licorice consumption can also potentially promote hypernatremia, hypokalemia and fluid volume retention. Furthermore, the authors report on the findings from a relatively recent meta-analysis where the increased intake of licorice was associated with significant increases in systolic (5.45 mmHg) and diastolic (3.19 mmHg) blood pressures (BPs). The authors also propose caution against the consumption of large quantities of licorice as some negative health effects may occur.

The incidence and occurrence of neurodegenerative diseases is on a constant increase worldwide, with Alzheimer's Disease (AD) and Parkinson's Disease (PD) being the most prevalent. A pivotal pathophysiological aspect of these diseases is the progressive neuronal loss that can be triggered by oxidative stress, mitochondrial dysfunction and neuroinflammation. In this regard, strategies aimed at counteracting the damaging effects of oxidative stress and neuroinflammation are considered as promising avenues to prevent neurodegeneration. In this context, bioactive molecules may play a role in counteracting the pathophysiological mechanisms linked with neurodegeneration. Of these, ε-viniferin (resveratrol dimer), as reported in the paper by Sergi et al., has shown to share similar effects with resveratrol in relation to neuroprotection in animal models of AD and Huntington's Disease [2]. Nevertheless, the effects of ε-viniferin on oxidative stress and inflammation-induced injury in dopaminegeric neurons remain relatively unexplored. In this study, the authors reported the neuroprotective potential of ε-viniferin in nerve growth factor (NGF)-differentiated PC12 cells, an in vitro dopaminergic model of Parkinson's disease (PD), and assessed the potential anti-inflammatory properties of this nutraceutical in a N9 microglia-neuronal PC12 cell co-culture system. The cells were pretreated with εviniferin, resveratrol and their mixtures before the administration of 6-hydroxydompamine (6-OHDA) that is recognized for inducing PD-like symptoms in animal models. In addition, the authors also investigated the effects of these stilbenes on the potential reduction in lipopolysaccharide-induced inflammation. The findings indicated that ε-viniferin alone or in combination with resveratrol protects the neuronal dopaminergic PC12 cells from 6-OHDA-induced cytotoxicity and apoptosis as well as the neuronal cytotoxicity triggered by microglial activation.

L-Theanine (L-THE) is the most abundant non-proteinogenic amino acid found in green tea. Its consumption has been proposed to be associated with stress-reducing effects and antihypertensive properties as well as improvements in cognitive functioning. In consideration of this, and given the recent commercial availability of relatively pure L-THE, there is a strong potential for the development of functional food products that contain this amino acid as a bioactive ingredient. A study by Williams et al. (2020) investigated the physiological responses, including heart rate (HR), heart rate variability (HRV) and BP in healthy males (*n* = 11), following the acute ingestion of a functional food products (mango sorbet) containing 200 mg of pure L-THE [3]. In this double blind, placebo-controlled cross-over trial, the participants were required to consume the test food or placebo mango sorbet (color and flavor matched) after which their physiological responses were continuously monitored over a period of 90 min. The study reported no significant differences between the intervention and placebo or within the individual groups (all *p*'s > 0.05). Furthermore, the results of this study indicate that there was also no parasympathetic response following L-THE intake (determined via the HR response). The authors have proposed that these findings could potentially be due to the interaction between L-THE and the food matrix it was embedded in, which may have affected L-THE bioavailability. In addition, the authors have acknowledged the limitations of the study, such as the sample size (considering this being the pilot trial) and the selection of healthy participants. Nevertheless, this study highlights the importance of the careful selection

of the food matrix composition in the development of future functional foods containing L-THE as an active ingredient, particularly considering that the matrix itself may affect the bioavailability of bioactive molecules.

The urbanization of sub-Saharan Africa is associated with a dietary shift towards the overconsumption of energy-dense foods, which in turn significantly contributes to the overall increase in cardiovascular disease, obesity and type 2 diabetes observed in this geographical area. A study by Zec et al. (2019) investigated the South African cohort of the Prospective Urban Rural Epidemiology (PURE) study, an international study investigating the health complications associated with urbanization in *low-*, *middle-* and *high-*income countries [4]. In this study, the authors examined data from 300 adults (older than 30 years of age) on the occurrence of hypertension and long-chain polyunsaturated fatty acids (PUFA) status. Data were analyzed from three consecutive examinations (2005, 2010 and 2015). The results revealed that the ten-year hypertension incidence significantly increased among rural participants (+20%, *p* = 0.001) while there was no significant change in urban participants (*p* = 0.253). Moreover, regardless of urbanization, n-6 PUFA status increased while the eicosapentanoic acid (EPA) status decreased over the 10-year period. Furthermore, authors reported an increase in BP (+2.92 systolic and +1.94 mmHg diastolic) and 1.46 higher odds of being hypertensive in the participants in the highest EPA tertile. In black South Africans included in this study sample, individual plasma n-6 PUFA were inversely associated with BP while EPA was associated with increases in BP leading to hypertension. Nonetheless, these findings should be interpreted with caution, especially in consideration of the well-documented positive effects of omega-3 fatty acids, such as EPA, on cardiovascular health.

The study by Leal-Martinez et al. (2020) is an exploratory randomized controlled clinical trial investigating the effects of a nutritional support system for improving motor function in children living with cerebral palsy (CP) in Mexico [5]. CP is one of the most common disabilities in childhood and changes to gross motor function is one of the main characteristics associated with this disease that can contribute to malnutrition in patients with CP. Furthermore, parasitosis is very common in this population sample and can contribute to the impairment of nutrient absorption. In this study, children with CP (*n* = 30) were divided in the three groups (*n* = 10 each): follow-up (monitoring of the diet only), control (dewormed and received nutritional therapy recommended by WHO) and intervention group (dewormed and received nutritional support system (diet and supplements)). All participants received Bobath physical therapy. The supplemental composition included a combination of amino acids (glutamine and arginine), vitamins (folic acid, cholecalciferol, ascorbic acid, nicotinic acid), minerals (zinc and selenium), spirulina, vegetable-based protein and n-3 PUFA. In addition, the participants also consumed probiotics (200 mg of *Saccharomyces Bouladii*) every 12 h for three days in the basal period and at week 7 for correcting malabsorption. The overall findings of the study were that a nutritional intervention consisting of diet and combination of supplements resulted in an improvement in gross motor function and promoted increases in walking ability in children living with CP.

In summary, the field of food bioactives is exceptionally diverse and it is a rapidly growing area of research and development. This is also fueled by the availability of novel analytical techniques, experimental models of human disease, and by an increase in consumer demand, seeking food products supported by solid health claims. Indeed, in order to meet these expectations, there is an ever-increasing need to translate the discoveries related to the health-promoting effects of food bioactives from the bench to the clinic, particularly in relation to brain and cardiometabolic health which represent areas of research with the greatest potential impact of human health. The findings reported in the studies published as part of this Special Issue further confirm the potential beneficial role of food bioactive derivatives and pave the way for future investigations aimed at further dissecting the impact of food bioactive molecules on human health. Nevertheless, we must not overlook the fact that food bioactives should not be considered as a panacea in the battle against human chronic diseases, as their health promoting effects cannot be sustained or even

occur in the absence of a healthy lifestyle characterized by adherence to healthy dietary patterns and physical activity. We (the co-editors of this Special Issue) are thankful to the authors and reviewers who have contributed to this issue by sharing their knowledge, findings and time.

**Author Contributions:** N.N. and D.S. wrote the paper. 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**


### *Article* **Anti-Apoptotic and Anti-Inflammatory Role of Trans** *ε***-Viniferin in a Neuron–Glia Co-Culture Cellular Model of Parkinson's Disease**

**Domenico Sergi 1,2, Alex Gélinas <sup>1</sup> , Jimmy Beaulieu <sup>1</sup> , Justine Renaud <sup>1</sup> , Emilie Tardif-Pellerin <sup>1</sup> , Jérôme Guillard <sup>3</sup> and Maria-Grazia Martinoli 1,4,\***


**Abstract:** The polyphenol trans-ε-viniferin (viniferin) is a dimer of resveratrol, reported to hold antioxidant and anti-inflammatory properties. The aims of our study were to evaluate the neuroprotective potential of viniferin in the nerve growth factor (NGF)-differentiated PC12 cells, a dopaminergic cellular model of Parkinson's disease (PD) and assess its anti-inflammatory properties in a N9 microglia–neuronal PC12 cell co-culture system. The neuronal cells were pre-treated with viniferin, resveratrol or their mixture before the administration of 6-hydroxydopamine (6-OHDA), recognized to induce parkinsonism in rats. Furthermore, N9 microglia cells, in a co-culture system with neuronal PC12, were pre-treated with viniferin, resveratrol or their mixture to investigate whether these polyphenols could reduce lipopolysaccharide (LPS)-induced inflammation. Our results show that viniferin as well as a mixture of viniferin and resveratrol protects neuronal dopaminergic cells from 6-OHDA-induced cytotoxicity and apoptosis. Furthermore, when viniferin, resveratrol or their mixture was used to pre-treat microglia cells in our co-culture system, they reduced neuronal cytotoxicity induced by glial activation. Altogether, our data highlight a novel role for viniferin as a neuroprotective and anti-inflammatory molecule in a dopaminergic cellular model, paving the way for nutraceutical therapeutic avenues in the complementary treatments of PD.

**Keywords:** trans-ε-viniferin; resveratrol; neuroprotection; oxidative stress; dopamine; apotosis; neuroinflammation; Parkinson's disease

#### **1. Introduction**

The incidence of neurodegenerative diseases is increasing worldwide, with Alzheimer's disease (AD) and Parkinson's disease (PD) being the most prevalent. Neurodegenerative diseases are characterized by neuronal loss, which in PD affects the dopaminergic neurons of the substantia nigra pars compacta (SNpc) [1]. Currently, apoptosis triggered by oxidative stress and neuroinflammation appears to be the main driver of dopaminergic neuron loss in PD [2].

Mitochondria are the principal source of reactive oxygen species (ROS), the mandatory bioproducts of oxidative metabolism. Mitochondrial dysfunction, a pivotal pathogenetic feature of PD, exacerbates oxidative stress by fostering ROS production which, in turn, via a feed-forward mechanism worsens mitochondrial dysfunction resulting in the activation of signaling cascades leading to apoptosis [3,4].

Due to their intrinsic low antioxidant activity and to the characteristic dopamine metabolism that generates pro-oxidant byproducts [5], dopaminergic neurons are particu-

**Citation:** Sergi, D.; Gélinas, A.; Beaulieu, J.; Renaud, J.; Tardif-Pellerin, E.; Guillard, J.; Martinoli, M.-G. Anti-Apoptotic and Anti-Inflammatory Role of Trans ε-Viniferin in a Neuron– Glia Co-Culture Cellular Model of Parkinson's Disease. *Foods* **2021**, *10*, 586. https://doi.org/10.3390/ foods10030586

Academic Editor: Annalisa Tassoni

Received: 12 February 2021 Accepted: 7 March 2021 Published: 11 March 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 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 (https:// creativecommons.org/licenses/by/ 4.0/).

larly vulnerable to oxidative stress, which may explain their susceptibility to degeneration typically found in PD.

Besides oxidative stress, neuroinflammation represents another key mechanism responsible for neurodegeneration in PD [6]. In support of this, post mortem investigation in the brain of parkinsonian patients demonstrated elevated levels of activated microglia, and pro-inflammatory cytokines in the substantia nigra and in the striatum [7,8]. However, whether neuroinflammation is a cause or a consequence of neurodegeneration it remains a matter of debate. Nonetheless, in vitro cell co-culture models [9,10], have provided evidence that eliciting a pro-inflammatory phenotype in microglial cells using lipopolysaccharide (LPS), a well know pro-inflammatory molecule, promoted microglia-induced neuronal cell damage [9,11]. To the same extent, stereotaxic LPS injection into the brain of rodents can recapitulate neuroinflammation and promote dopaminergic neuron degeneration [12]. Thus, independently of whether neuroinflammation is involved in the etiology of PD or is the result of nigrostriatal pathway injury, sustained microglia pro-inflammatory activation can contribute to dopaminergic neuron loss.

Nowadays, strategies aimed at tackling oxidative stress and neuroinflammation may represent positive avenues to prevent neurodegeneration. In this regard, bioactive molecules derived from plants have emerged for their neuroprotective potential by countering oxidative stress, inflammation or both. Of these, resveratrol in light of its antioxidant [13] antiinflammatory [11], and synergistic effects with other polyphenols such as quercetin [11,14,15], is considered one of the most promising. Nonetheless, resveratrol containing foods also present oligomeric forms of stilbenes whose biological effects remain to be fully elucidated, especially with regard to the central nervous system. Of these, ε-viniferin (viniferin), a naturally occurring resveratrol dimer with higher activity and stability, may contribute to the total amount of stilbenes consumed with foods [16]. This molecule shares several health promoting effects with resveratrol, including neuroprotective properties, as demonstrated in animal models of Huntington's disease [17] and AD [18,19]. However, the effect of viniferin on oxidative stress and inflammation-induced injury in dopaminergic neurons remains to be elucidated.

Thus, the aim of this study was to determine whether viniferin was able to prevent 6-hydroxydopamine (6-OHDA)-induced cytotoxicity and apoptosis in a cellular model of PD, PC12 dopaminergic neurons. Furthermore, we aimed at elucidating whether this resveratrol dimer could prevent LPS-activated N9 microglial cells from inducing PC12 cell death in an N9 microglia PC12 cell co-culture system. The effect of viniferin on the aforementioned outcomes was also investigated in the presence of resveratrol, and resveratrol itself was used as reference molecule to benchmark the effect of viniferin. Our results show that viniferin can successfully protect PC12 dopaminergic neurons from oxidative stress-induced apoptosis, and is also able to prevent microglia from promoting neuronal death, with a synergetic effect being observed when resveratrol is used in a mixture together with viniferin.

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

#### *2.1. Drugs and Chemicals*

All reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA) unless stated otherwise.

#### *2.2. Extractions and Purifications of Polyphenols*

Grape canes variety "Ugni" (Vitaceae) were collected from Cognac's vineyards in France. Grape canes were extracted in an extruder (BC21 clextral) apparatus with a mixture of ethanol/water (8/2) as a solvent. The solution was concentrated after filtration, at 40 ◦C under a vacuum and lyophilized. Dried extract (4 g) was dissolved in 20 mL of a mixture called Biosolvants 1 with the following composition Makigreen D10–EtOAc–EtOH–H2O, (4:2:3:2) and injected into a centrifugal partition chromatography (CPC) apparatus. The CPC method used to obtain viniferin is the one presented previously [20], but using green

solvents as Biosolvent 1 and Biosolvent 2 (Cyclopentane–EtOAc–EtOH–H2O, 1:2:1:2). Thus, fractions were collected and combined on the basis of HPLC analysis, providing a total of three fractions of interest. Fraction 1 (101 mg) corresponded to 95% pure resveratrol as white solid. Fraction 2 (175 mg) contained the dimer of interest, as with viniferin along with other compounds. Afterwards, fraction 2 was re-purified by semi-preparative HPLC and a yellow solid isolated was identified by 1H and 13C NMR spectroscopy analysis in acetone-d6 and mass spectrometry as with viniferin to 98% purity. The concentration of viniferin after extrusion–extraction is very dependent on the grape variety used but with Ugni's grapecane the total polyphenol yield is 0.3% by weight with a concentration of viniferin in the extract grapine-shoot of between 15 and 20%.

#### *2.3. Cell Culture and Treatments*

A rat pheochromocytoma cell line (PC12 cells) was obtained from ATCC (Rockville, MD, USA). Cells were maintained in a humidified environment at 37 ◦C and 5% CO<sup>2</sup> atmosphere and routinely grown in RPMI 1640 medium supplemented with 10% heatinactivated horse serum (HS), 5% heat inactivated fetal bovine serum (FBS; Corning Cellgro, Manassas, VA, USA) and gentamicin (50 µg/mL). Neuronal differentiation was evoked by nerve growth factor-7S (NGF, 50 ng/mL) in RPMI 1640 supplemented with 1% FBS for 9–10 days, as already described [10,21]. NGF-differentiated PC12 cells (neuronal PC12) displayed a dopaminergic phenotype (Figure 1) as already reported [21,22]. They were pretreated with resveratrol or viniferin (generously provided by Dr. Jérôme Guillard, University of Poitiers, Poitiers, France) at 10–9 M or a mixture of the two molecules at 10–9 M for 3 h. Then 6-OHDA (50 µM) or LPS (2 µg/mL) [9] was added to the medium for 24 h. All experiments were performed in RPMI medium without phenol red, supplemented with 1% charcoal-stripped serum to remove steroids from the medium. Microglial cell line N9 was grown in Dulbecco's modified Eagle's medium nutrient mixture F12-Ham's (DMEM-F12) supplemented with 10% HS. *Foods* **2021**, *10*, x FOR PEER REVIEW 4 of 16

**Figure 1.** Representative microphotograph of nerve growth factor (NGF)-differentiated PC12 for 9 days by immunofluorescence. Nuclei are counterstained in blue with Dapi. NF: neurofilaments revealed with an anti-neurofilaments antibody (green). TH: tyrosine hydroxylase revealed with an anti-tyrosine antibody as a marker of dopamine (red). Scale bar = 10 µm. *2.4. Neuronal–Glial Co-Culture* **Figure 1.** Representative microphotograph of nerve growth factor (NGF)-differentiated PC12 for 9 days by immunofluorescence. Nuclei are counterstained in blue with Dapi. NF: neurofilaments revealed with an anti-neurofilaments antibody (green). TH: tyrosine hydroxylase revealed with an anti-tyrosine antibody as a marker of dopamine (red). Scale bar = 10 µm.

Neuronal PC12 cells and N9 microglia were co-cultured without cellular contact to study the impact of LPS-activated microglia on the survival of neuronal cells, as we have

Cytotoxicity was evaluated using a colorimetric assay based on the quantification of LDH activity released from damaged cells into the cell culture medium, as already described [20]. LDH is a stable cytoplasmic enzyme present in all cells, which is rapidly released into the cell culture supernatant upon plasma membrane damage. Enzyme activity in the cell culture medium is a direct function of lysed cells [24]. Briefly, 100 μL of cell-free culture media served to quantify LDH activity by measuring absorbance at 490 nm using a Thermo Lab Systems (Franklin, MA, USA). Total cellular LDH was

between the two cellular systems [23]. Briefly, N9 microglial cells were grown onto culture inserts (pore size 0.4 μm; BD Falcon, Oakville, ON, Canada), LPS (2 μg/mL) was added. After 24 h, inserts containing N9 cells were washed with PBS and then transferred on neuronal PC12, for another 24 h. The PC12 supernatant was collected for cell death measurements with lactate dehydrogenase (LDH) cytotoxicity tests as described below. For immunofluorescent experiment, N9 microglial cells were grown in culture inserts and treated as described then transferred on neuronal PC12 cells grown previously on

coverslips [9,11].

*2.5. Cytotoxicity Measurements*

#### *2.4. Neuronal–Glial Co-Culture*

Neuronal PC12 cells and N9 microglia were co-cultured without cellular contact to study the impact of LPS-activated microglia on the survival of neuronal cells, as we have already described [9]. In this co-culture system, microglial cells communicate with PC12 neuronal cells through a semipermeable membrane, in the absence of a direct contact between the two cellular systems [23]. Briefly, N9 microglial cells were grown onto culture inserts (pore size 0.4 µm; BD Falcon, Oakville, ON, Canada), LPS (2 µg/mL) was added. After 24 h, inserts containing N9 cells were washed with PBS and then transferred on neuronal PC12, for another 24 h. The PC12 supernatant was collected for cell death measurements with lactate dehydrogenase (LDH) cytotoxicity tests as described below. For immunofluorescent experiment, N9 microglial cells were grown in culture inserts and treated as described then transferred on neuronal PC12 cells grown previously on coverslips [9,11].

#### *2.5. Cytotoxicity Measurements*

Cytotoxicity was evaluated using a colorimetric assay based on the quantification of LDH activity released from damaged cells into the cell culture medium, as already described [20]. LDH is a stable cytoplasmic enzyme present in all cells, which is rapidly released into the cell culture supernatant upon plasma membrane damage. Enzyme activity in the cell culture medium is a direct function of lysed cells [24]. Briefly, 100 µL of cell-free culture media served to quantify LDH activity by measuring absorbance at 490 nm using a Thermo Lab Systems (Franklin, MA, USA). Total cellular LDH was determined cells lysed using 1% Triton X-100 (high control); the assay medium was used as a low control and was subtracted from all absorbance measurements:

Cytotoxicity (%) = (Experimental value-Low control)/(High control-Low control) × 100

#### *2.6. MTT Assay*

The cell metabolic activity was measured using 3-(4,5-dimethyltrazol-2-yl)-2,5 diphenyltetrazolium bromide (MTT) assay [21]. Neuronal PC12 cells, plated in 96-wells plates, were treated as already described and then incubated for 4 h at 37 ◦C with MTT dye (5 mg/mL), followed by solubilization in SDS 10% and the absorbance was measured at the 595 nm with a microplate reader (Thermo Lab Systems, Franklin, MA, USA).

#### *2.7. Apoptosis- Specific DNA Denaturation Detection*

Apoptosis-specific detection of DNA denaturation by formamide was assessed with the single-stranded DNA (ssDNA) apoptosis ELISA kit (Chemicon International, Telok Panglima Garang, Malaysia) as already described [15,22]. This procedure relies the on selective DNA denaturation by formamide in apoptotic cells, which does not occur in necrotic cells nor in cells with DNA damage in the absence of apoptosis [25]. The detection of denatured DNA was performed with a monoclonal antibody highly specific to ssDNA and a peroxidase-labelled secondary antibody on fixed neuronal PC12 cells. Following, ssDNA immune detection, the reaction was then terminated with a hydrochloric acid solution and ssDNA was quantified by measuring absorbance at 405 nm in a microplate reader (Thermolab Systems, Vasai-Virar, India). ssDNA was quantified relative to control conditions. Absorbance of positive (wells coated with provided ssDNA) and negative controls (wells coated with S1 nuclease that digest ssDNA) served as quality controls for the ELISA assay, as previously described [26,27].

#### *2.8. Immunofluorescent Microscopy*

Apoptotic P12 neuronal cells were detected by both terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL, Roche Diagnostics, Basel, Switzerland) and cleaved caspase-3 immunofluorescence. Neuronal PC12 cells were seeded at 25,000 cells/cm<sup>2</sup> , differentiated and treated on collagen-coated coverslips in 24-well plates for immunofluo-

rescent microscopy. Briefly, after viniferin and/or resveratrol and/or 6-OHDA treatments, cells were fixed in 4% paraformaldehyde for 15 min at 37 ◦C, then washed and incubated for 1 h at room temperature in a blocking and permeabilizing solution containing 1% BSA, 0.18% fish skin gelatin, 0.1% Triton X-100 and 0.02% sodium azide [22,26]. Cells were then incubated with anti-cleaved caspase-3 antibody (New England Biolabs, Pickering, ON, Canada) diluted 1:500, for 2 h at room temperature, followed by 90-min incubation with a Cy3-conjugated secondary antibody (Medicorp, Montreal, QC, Canada) diluted 1:500 for 1 h at 4 ◦C. The coverslips were then transferred to the TUNEL reaction mixture in a humidified atmosphere at 37 ◦C. The cells were rinsed with PBS, nuclei were counterstained in blue with DAPI for 10 min at 37 ◦C and mounted with ProLong Antifade kits (Invitrogen). PC12 neuronal cells were considered to be apoptotic when they were positive for cleaved caspase-3 and their nuclei were stained by TUNEL. The number of apoptotic neuronal cells among 300 randomly chosen neuronal cells was counted on 10 different optical fields from 3 slides per group [15], with Pro Express 6.3 software (Media Cybernetics, Rockville, MD, USA).

#### *2.9. Electrophoresis and Immunoblot Analysis*

Neuronal cells were grown and treated in 6-well plates. Total proteins were extracted with a Nuclear Extraction Kit (Active Motif, Carlsbad, CA, USA) and their concentration determined by using the BCA protein assay kit (Pierce Biotechnology Inc., Rockford, AZ, USA). Fifteen µg of protein were loaded onto a 12% SDS-polyacrylamide gel. After electrophoretic separation (125 V, for 1 h 30), proteins were transferred onto PVDF membranes (0.22 µm pore size, BioRad) at 25 V overnight. The membranes were blocked in TBST (TBS + Tween 0.05%) with 5% non-fat dry milk for 1 h at room temperature and incubated overnight at 4 ◦C with primary antibodies: anti-cleaved caspase-3 antibody (1:500) anti-cleaved PARP-1 antibody (1:500) or an anti-β-actin antibody (1:500) (Cell Signaling, Danvers, MA, USA). The blots were then rinsed three times with TBS 0.1% Tween 20 and incubated with peroxidase-conjugated (POD) secondary antibody (1:10,000) for 2 h at room temperature. Blots were washed with TBS 0.1% Tween 20 three times and the signal finally revealed by enhanced chemiluminescence with the AlphaEase FC imaging system (Alpha Innotech, San Leandro, CA, USA) and analyzed with AlphaEase FC software (Alpha Innotech) and ImageJ (imagej.nih.gov).

#### *2.10. ELISA*

Pro-inflammatory cytokines IL-1α and TNF-α were measured by specific ELISA kits (BioLegend, San Diego, CA, USA). Following pre-incubation with resveratrol, viniferin, or a mixture of the two molecules for 3 h, N9 microglial cells were administered with LPS 4 µg/mL for 24 h. Supernatants were collected after 27 h and processed for the presence of selected cytokines by ELISA, according to the protocols supplied by the manufacturer. Briefly, the supernatants were collected following the polyphenols and LPS challenges. 96-well plates were coated with mouse-specific monoclonal antibody (IL-1α and TNF-α) and after an overnight incubation, standards and samples were added to the wells for 2 h. The plates were then incubated in the presence of a biotinylated anti-mouse detection antibody for 1 h, followed by a 30 min incubation with an avidin horseradish peroxidase solution. Finally, a tetramethylbenzidine solution was added to the wells for 15 min in the dark to reveal the presence of the cytokines. The reaction was terminated by the addition of 2N sulfuric acid and resulting absorbance recorded at 450 nm using a microplate reader (Synergy H1, BioTek, Winnosky, VT, USA).

#### *2.11. Statistical Analysis*

Significant differences between groups were ascertained by One-way analysis of variance (ANOVA), followed by Tukey's post-hoc analysis, performed using the Graph-Pad Prism8 for Windows (http://www.graphpad.com/). All data were expressed as means ± SEM from at least 3 independent experiments. A *p*-value < 0.05 was considered

statistically significant. Asterisks (\*) indicate statistical differences between the treatments and their respective controls (\*\*\* *p* < 0.001, \*\* *p* < 0.01, \* *p* < 0.05), plus signs (+) denote statistical differences between the treatments and 6-OHDA (+++ *p* < 0.001, ++ *p* < 0.01, + *p* < 0.05), dollar signs (\$) indicate statistical difference between 6-OHDA+resveratrol and 6-OHDA+mixture of the two polyphenols or 6-OHDA and viniferin (\$\$\$ < 0.001, \$ < 0.05) and number signs (#) indicates statistical difference between 6-OHDA+viniferin and 6-OHDA + mixture of the two polyphenols (### *p* < 0.001, ## *p* < 0.01). The same symbols were used in the experiments where LPS was used instead of 6-OHDA.

#### **3. Results**

#### *3.1. Viniferin Reduced 6-OHDA-Induced Cytotoxicity and Promoted Cell Survival*

To evaluate whether viniferin, alone and in combination with resveratrol protected neuronal PC12 cells from 6-OHDA-induced cytotoxicity, we investigated LDH release and metabolic activity in neuronal cells pre-treated for 3 h with resveratrol, viniferin or their combination, before administration of 6-OHDA neurotoxin for 24 h. When the two polyphenols, both at a concentration of 10–9 M, were administered alone before 6- OHDA treatment, they induced a decrease in LDH release by 19 ± 2.5% and 20.3 ± 1.3%, respectively for resveratrol and viniferin (Figure 2A) and increased cellular metabolic activity (Figure 2B) compared to cells treated with 6-OHDA only, by 15.3 ± 0.9% and 14.3 ± 1.9%, respectively. Moreover, preincubation with a mixture of resveratrol and viniferin at a final concentration of 10 -9 M, induced a 27.7 ± 2.3% decrease in LHD release compared to cells exposed to 6-OHDA only (Figure 2A). The same effect was observed for cellular metabolic activity with an increase of 26 ± 2.1% (Figure 2B), notably the effect of the mix of the two polyphenols administered before 6-OHDA challenge was more marked compared to the single polyphenols (Figure 2B). *Foods* **2021**, *10*, x FOR PEER REVIEW 7 of 16 at a final concentration of 10 -9 M, induced a 27.7 ± 2.3% decrease in LHD release compared to cells exposed to 6-OHDA only (Figure 2A). The same effect was observed for cellular metabolic activity with an increase of 26 ± 2.1% (Figure 2B), notably the effect of the mix of the two polyphenols administered before 6-OHDA challenge was more marked compared to the single polyphenols (Figure 2B).

**Figure 2.** Levels of cytotoxicity (**A**) and metabolic activity (**B**) of PC12 neurons pretreated for three hours with culture medium (control), resveratrol, viniferin or a mixture of both polyphenols (Mix), all at a final concentration of 10 -9. Cells were then treated with 6-hydroxydopamine (6-OHDA) at 50 μM for 24 h, as described in the Material and Methods. The supernatants and the cells were used to perform cell death (lactate dehydrogenase (LDH) assay). Data are expressed as means ± SEM of five independent experiments. For each experiment, each condition was in assessed in sextuplicate. Asterisks (\*) indicate statistical differences between the treatments and their respective controls (\*\*\* *p* < 0.001), plus signs (+) denote statistical differences between 6-OHDA and 6-OHDA in the presence of polyphenols (+++ *p* < 0.001, ++ *p* < 0.01,). **Figure 2.** Levels of cytotoxicity (**A**) and metabolic activity (**B**) of PC12 neurons pretreated for three hours with culture medium (control), resveratrol, viniferin or a mixture of both polyphenols (Mix), all at a final concentration of 10–9. Cells were then treated with 6-hydroxydopamine (6-OHDA) at 50 µM for 24 h, as described in the Material and Methods. The supernatants and the cells were used to perform cell death (lactate dehydrogenase (LDH) assay). Data are expressed as means ± SEM of five independent experiments. For each experiment, each condition was in assessed in sextuplicate. Asterisks (\*) indicate statistical differences between the treatments and their respective controls (\*\*\* *p* < 0.001), plus signs (+) denote statistical differences between 6-OHDA and 6-OHDA in the presence of polyphenols (+++ *p* < 0.001, ++ *p* < 0.01,).

#### *3.2. Viniferin Decreased the Rate of Apoptotic Neuronal PC12 Cells 3.2. Viniferin Decreased the Rate of Apoptotic Neuronal PC12 Cells*

In order to determine whether polyphenols may counteract cell death by inhibiting the apoptotic cascade, we assessed DNA fragmentation using a ssDNA apoptosis ELISA kit. In line with the results illustrated in Figure 2, resveratrol and viniferin alone or in In order to determine whether polyphenols may counteract cell death by inhibiting the apoptotic cascade, we assessed DNA fragmentation using a ssDNA apoptosis ELISA kit. In line with the results illustrated in Figure 2, resveratrol and viniferin alone or in

combination inhibited 6-OHDA-induced DNA fragmentation in neuronal PC12 cells by

confirmed by double immunofluorescence indicating an increase in TUNEL and cleaved caspase-3 positive cells after 6-OHDA treatment (Figure 3B). Notably, this anti-apototic effect was more marked when resveratrol and viniferin were used in combination which induced a 36.3 ± 4.2% decrease in the number of apoptotic cells relative to 6-OHDA treated cells, compare to the 22 ± 4% and 25.7 ± 2.3% of resveratrol and viniferin, respectively

(Figure 3B).

combination inhibited 6-OHDA-induced DNA fragmentation in neuronal PC12 cells by 52.8 ± 9.6%, 52 ± 9.2% and 55.8 ± 10.3%, respectively (Figure 3A). This result was further confirmed by double immunofluorescence indicating an increase in TUNEL and cleaved caspase-3 positive cells after 6-OHDA treatment (Figure 3B). Notably, this anti-apototic effect was more marked when resveratrol and viniferin were used in combination which induced a 36.3 ± 4.2% decrease in the number of apoptotic cells relative to 6-OHDA treated cells, compare to the 22 ± 4% and 25.7 ± 2.3% of resveratrol and viniferin, respectively (Figure 3B). *Foods* **2021**, *10*, x FOR PEER REVIEW 8 of 16

**Figure 3.** Detection of apoptosis by single stranded (ssDNA) fragmentation (**A**) and by caspase-3 and TUNEL double immunofluorescence (**B**). PC12 neurons pre-treated for three hours with culture medium (control), resveratrol, viniferin or a mixture of both polyphenols (Mix) all at a final concentration of 10–9 M. Cells were then treated with 6-OHDA at 50 μM for 24 h, as described in the Material and Methods. Apoptotic neuronal cells among 300 randomly chosen neuronal cells were counted on 10 different optical fields from 3 slides per group, as described in Material and Methods. Data are expressed as means ± SEM of three experiments. For each experiment, each condition was assessed in triplicate. Asterisks (\*) indicate statistical differences between the treatments and their respective controls (\*\*\* *p* < 0.001, \*\* *p* < 0.01), plus signs (+) denote statistical differences between 6-OHDA and 6-OHDA in the presence of polyphenols (+++ *p* < 0.001, ++ *p* < 0.01), dollar sign (\$) indicates statistical difference between 6-OHDA+resveratrol and 6-OHDA+mixture of the two polyphenols (\$\$\$ < 0.001) and number sign (#) indicates statistical difference between 6-OHDA+viniferin and 6-OHDA + mixture of the two polyphenols (### *p* < 0.001). **Figure 3.** Detection of apoptosis by single stranded (ssDNA) fragmentation (**A**) and by caspase-3 and TUNEL double immunofluorescence (**B**). PC12 neurons pre-treated for three hours with culture medium (control), resveratrol, viniferin or a mixture of both polyphenols (Mix) all at a final concentration of 10–9 M. Cells were then treated with 6-OHDA at 50 µM for 24 h, as described in the Material and Methods. Apoptotic neuronal cells among 300 randomly chosen neuronal cells were counted on 10 different optical fields from 3 slides per group, as described in Material and Methods. Data are expressed as means ± SEM of three experiments. For each experiment, each condition was assessed in triplicate. Asterisks (\*) indicate statistical differences between the treatments and their respective controls (\*\*\* *p* < 0.001, \*\* *p* < 0.01), plus signs (+) denote statistical differences between 6-OHDA and 6-OHDA in the presence of polyphenols (+++ *p* < 0.001, ++ *p* < 0.01), dollar sign (\$) indicates statistical difference between 6-OHDA+resveratrol and 6-OHDA+mixture of the two polyphenols (\$\$\$ < 0.001) and number sign (#) indicates statistical difference between 6-OHDA+viniferin and 6-OHDA + mixture of the two polyphenols (### *p* < 0.001).

#### *3.3. Viniferin Used Alone or in Combination with Resveratrol Decreased the Cleavage of Caspase-3 and PARP-1 in Neuronal PC12 3.3. Viniferin Used Alone or in Combination with Resveratrol Decreased the Cleavage of Caspase-3 and PARP-1 in Neuronal PC12*

To further confirm the ability of viniferin to counteract the activation of the apoptotic signaling cascade, we investigated the cleavage of caspase-3 and PARP-1, two dominant markers of early apoptotic signaling, by Western blotting. 6-OHDA increased the cleavage of both caspase-3 and PARP-1 (Figure 4A,B), while these effects were blunted by preincubation of neuronal PC12 cells with either viniferin (20.5 ± 0.8% for caspase-3 and 26.1 ± 1.5% forPARP-1), resveratrol (19.8 ± 2.6% for caspase-3 and 27.7 ± 3.3% for PARP-1) or their combination (34.6 ± 2.6 for caspase-3 and 38.9 ± 1.5% for PARP-1) before administration with 6-OHDA (Figure 4A,B). This effect was more pronounced when the mixture of resveratrol and viniferin was administered. In particular, the cleavage of PARP-1 did not increase in the cells pre-treated with the mix of polyphenols compared to control, despite these cells being exposed to 6-OHDA (Figure 4B). These data strongly suggest that, viniferin or resveratrol, used alone or in combination, exerted anti-apoptotic effects, confirming our results illustrated in Figures 2 and 3. To further confirm the ability of viniferin to counteract the activation of the apoptotic signaling cascade, we investigated the cleavage of caspase-3 and PARP-1, two dominant markers of early apoptotic signaling, by Western blotting. 6-OHDA increased the cleavage of both caspase-3 and PARP-1 (Figure 4A,B), while these effects were blunted by pre-incubation of neuronal PC12 cells with either viniferin (20.5 ± 0.8% for caspase-3 and 26.1 ± 1.5% forPARP-1), resveratrol (19.8 ± 2.6% for caspase-3 and 27.7 ± 3.3% for PARP-1) or their combination (34.6 ± 2.6 for caspase-3 and 38.9 ± 1.5% for PARP-1) before administration with 6-OHDA (Figure 4A,B). This effect was more pronounced when the mixture of resveratrol and viniferin was administered. In particular, the cleavage of PARP-1 did not increase in the cells pre-treated with the mix of polyphenols compared to control, despite these cells being exposed to 6-OHDA (Figure 4B). These data strongly suggest that, viniferin or resveratrol, used alone or in combination, exerted anti-apoptotic effects, confirming our results illustrated in Figures 2 and 3.

**Figure 4.** Expression of apoptotic cleaved caspase-3 (**A**) and cleaved Parp-1 (**B**) in PC12 neurons pretreated for three hours with the culture medium (control), resveratrol, viniferin or a mixture of both polyphenols (Mix) at a final concentration of 10 -9 M. Cells were then treated with 6-OHDA at 50 μM for 24 h, as described in the Material and Methods. Top panel: Data are expressed as the ratio of cleaved caspase-3 (**A**) or cleaved Parp-1 (**B**) to β-actin. Bottom panel: representative Western blot bands of cleaved caspase-3, cleaved Parp-1 or β-actin. Optical densities were measured on the same membrane. Data are expressed as means ± SEM of three experiments performed in triplicate. Asterisks (\*) indicate statistical differences between the treatments and their respective controls (\*\*\* *p* < 0.001, \*\* *p* < 0.01), plus signs (+) denote statistical differences between 6-OHDA and 6-OHDA in the presence of polyphenols (+++ *p* < 0.001, + *p* < 0.05). **Figure 4.** Expression of apoptotic cleaved caspase-3 (**A**) and cleaved Parp-1 (**B**) in PC12 neurons pretreated for three hours with the culture medium (control), resveratrol, viniferin or a mixture of both polyphenols (Mix) at a final concentration of 10–9 M. Cells were then treated with 6-OHDA at 50 µM for 24 h, as described in the Material and Methods. Top panel: Data are expressed as the ratio of cleaved caspase-3 (**A**) or cleaved Parp-1 (**B**) to β-actin. Bottom panel: representative Western blot bands of cleaved caspase-3, cleaved Parp-1 or β-actin. Optical densities were measured on the same membrane. Data are expressed as means ± SEM of three experiments performed in triplicate. Asterisks (\*) indicate statistical differences between the treatments and their respective controls (\*\*\* *p* < 0.001, \*\* *p* < 0.01), plus signs (+) denote statistical differences between 6-OHDA and 6-OHDA in the presence of polyphenols (+++ *p* < 0.001, + *p* < 0.05).

#### *3.4. Viniferin, Resveratrol and Their Combination Decreased Neuronal PC12 Cytotoxicity Induced by Activated Microglia, in a Co-Culture System 3.4. Viniferin, Resveratrol and Their Combination Decreased Neuronal PC12 Cytotoxicity Induced by Activated Microglia, in a Co-Culture System*

We have previously demonstrated that LPS-activated N9 microglia cells induced cytotoxicity in neuronal PC12 in a paracrine fashion in a microglia-neuronal PC12 cells co-culture system [9,11]. In order to evaluate whether viniferin may prevent N9 activation and subsequent neuronal PC12 cytotoxicity, N9 cells were pre-treated with the two polyphenols or their mixture and incubated with LPS for 24 h before being co-cultured with PC12 neurons for additional 24 h (Figure 5A). Viniferin, resveratrol and their combination counteracted the increase in LDH release from PC12 neurons co-cultured with LPS-activated N9 microglia, by 51 ± 7.8%, 31.3 ± 5.4% and 64 ± 4.4%, respectively (Figure 5A). This effect was even more marked for viniferin, and the combination of viniferin and resveratrol compared to resveratrol (Figure 5A), suggesting an antiinflammatory role for viniferin and the mixture of viniferin and resveratrol. Furthermore, to evaluate whether these polyphenols exerted a neuroprotective effect against activatedmicroglia secretions, neuronal PC12 cells were pretreated with either viniferin, resveratrol or their mixture before being incubated with LPS-stimulated microglial cells (Figure 5B). Viniferin and resveratrol, both alone and in combination, decreased microglia-induced neuronal PC12 cytotoxicity as indicated by a 28.3 ± 2.7%, 21.7 ± 1.5% and 51.3 ± 2% decrease in LDH release, respectively (Figure 5B), suggesting a neuroprotective role for these polyphenols. Notably, the release of LDH was significantly lower for neuronal PC12 pre-treated with the combination of viniferin and resveratrol compared to each We have previously demonstrated that LPS-activated N9 microglia cells induced cytotoxicity in neuronal PC12 in a paracrine fashion in a microglia-neuronal PC12 cells co-culture system [9,11]. In order to evaluate whether viniferin may prevent N9 activation and subsequent neuronal PC12 cytotoxicity, N9 cells were pre-treated with the two polyphenols or their mixture and incubated with LPS for 24 h before being co-cultured with PC12 neurons for additional 24 h (Figure 5A). Viniferin, resveratrol and their combination counteracted the increase in LDH release from PC12 neurons co-cultured with LPS-activated N9 microglia, by 51 ± 7.8%, 31.3 ± 5.4% and 64 ± 4.4%, respectively (Figure 5A). This effect was even more marked for viniferin, and the combination of viniferin and resveratrol compared to resveratrol (Figure 5A), suggesting an anti-inflammatory role for viniferin and the mixture of viniferin and resveratrol. Furthermore, to evaluate whether these polyphenols exerted a neuroprotective effect against activated-microglia secretions, neuronal PC12 cells were pretreated with either viniferin, resveratrol or their mixture before being incubated with LPS-stimulated microglial cells (Figure 5B). Viniferin and resveratrol, both alone and in combination, decreased microglia-induced neuronal PC12 cytotoxicity as indicated by a 28.3 ± 2.7%, 21.7 ± 1.5% and 51.3 ± 2% decrease in LDH release, respectively (Figure 5B), suggesting a neuroprotective role for these polyphenols. Notably, the release of LDH was significantly lower for neuronal PC12 pre-treated with the combination of viniferin and resveratrol compared to each independent polyphenol.

independent polyphenol.

*Foods* **2021**, *10*, x FOR PEER REVIEW 10 of 16

**Figure 5.** Co-culture of N9 microglial cells on neuronal PC12 cells. (**A**) N9 cells were pre-treated with polyphenols followed incubation with lipopolysaccharide (LPS), as described in the Materials and Methods. (**B**) N9 cells were treated with LPS and neuronal PC12 were pre-incubated with polyphenols as described in the Materials and Methods. Data are expressed as means ± SEM of three experiments. Asterisks (\*) indicate statistical differences between the treatments and their respective controls (\*\*\* *p* < 0.001), plus signs (+) denote statistical differences between LPS and LPS in the presence of polyphenols. (+++ *p* < 0.001, ++ *p* < 0.01), dollar sign (\$) indicates statistical difference between LPS + resveratrol and LPS + mixture of the two polyphenols (\$\$\$ < 0.001, \$ < 0.05) and number sign (#) indicates statistical difference between LPS + viniferin and LPS + mixture of the two polyphenols (## *p* < 0.01). *3.5. Viniferin Affected the Secretion of Pro-Inflammatory Cytokines from LPS-Activated N9*  **Figure 5.** Co-culture of N9 microglial cells on neuronal PC12 cells. (**A**) N9 cells were pre-treated with polyphenols followed incubation with lipopolysaccharide (LPS), as described in the Materials and Methods. (**B**) N9 cells were treated with LPS and neuronal PC12 were pre-incubated with polyphenols as described in the Materials and Methods. Data are expressed as means ± SEM of three experiments. Asterisks (\*) indicate statistical differences between the treatments and their respective controls (\*\*\* *p* < 0.001), plus signs (+) denote statistical differences between LPS and LPS in the presence of polyphenols. (+++ *p* < 0.001, ++ *p* < 0.01), dollar sign (\$) indicates statistical difference between LPS + resveratrol and LPS + mixture of the two polyphenols (\$\$\$ < 0.001, \$ < 0.05) and number sign (#) indicates statistical difference between LPS + viniferin and LPS + mixture of the two polyphenols (## *p* < 0.01).

*Cells*

#### To further investigate the role of viniferin, resveratrol or their mixture on *3.5. Viniferin Affected the Secretion of Pro-Inflammatory Cytokines from LPS-Activated N9 Cells*

neuroinflammatory mechanisms in N9 microglia-neuronal PC12 cells co-culture, we investigated whether these polyphenols may counteract LPS-induced secretion of proinflammatory cytokines. Figure 6 shows that a 24-h LPS treatment increased the secretion of both IL-1α (Figure 6A) and TNFα (Figure 6B). While neither viniferin, resveratrol nor their combination counteracted the secretion of TNFα triggered by LPS (Figure 6B), pretreating N9 cells with these polyphenols tended to decrease the levels of IL-1α in the cell culture media compared to cells exposed to LPS only (43.3 ± 7.3% for resveratrol; 46.7 ± 11.1% viniferin; 41.5 ± 6.3% for their combination) (Figure 6A). To further investigate the role of viniferin, resveratrol or their mixture on neuroinflammatory mechanisms in N9 microglia-neuronal PC12 cells co-culture, we investigated whether these polyphenols may counteract LPS-induced secretion of pro-inflammatory cytokines. Figure 6 shows that a 24-h LPS treatment increased the secretion of both IL-1α (Figure 6A) and TNFα (Figure 6B). While neither viniferin, resveratrol nor their combination counteracted the secretion of TNFα triggered by LPS (Figure 6B), pre-treating N9 cells with these polyphenols tended to decrease the levels of IL-1α in the cell culture media compared to cells exposed to LPS only (43.3 ± 7.3% for resveratrol; 46.7 ± 11.1% viniferin; 41.5 ± 6.3% for their combination) (Figure 6A). *Foods* **2021**, *10*, x FOR PEER REVIEW 11 of 16

**Figure 6.** Measurements of cytokine secretion from LPS-activated microglial N9 cells by ELISA specific kit for IL-1α (**A**) and TNFα (**B**). Data are expressed as means ± SEM of four experiments. Asterisks (\*) indicate statistical differences between the treatments and their respective controls (\*\*\* *p* < 0.001, \*\* *p* < 0.01). **4. Discussion Figure 6.** Measurements of cytokine secretion from LPS-activated microglial N9 cells by ELISA specific kit for IL-1α (**A**) and TNFα (**B**). Data are expressed as means ± SEM of four experiments. Asterisks (\*) indicate statistical differences between the treatments and their respective controls (\*\*\* *p* < 0.001, \*\* *p* < 0.01).

microglia, than resveratrol or viniferin alone.

cleavage of caspase-3 and PARP, as our present results illustrate.

This study provides evidence on the neuroprotective and anti-inflammatory potential of the polyphenol viniferin in a cellular model of PD, PC12 dopaminergic

6-OHDA, due to its similarities to endogenous catecholamines, is taken up and accumulates in catecholaminergic neurons promoting neurotoxicity mainly via oxidative stress-related mechanisms [28]. In support of this, its administration directly into the brain via stereotactic surgery induces parkinsonism in rodent models by promoting nigrostriatal pathway degeneration [29]. These effects were recapitulated in the present study in which 6-OHDA promoted cytotoxicity, as indicated by an increase in LDH release from PC12 dopaminergic neurons and a decrease in their metabolic activity. Importantly, these neurotoxic effects are dependent on the activation of apoptosis as indicated by an increase in caspase-3/TUNEL-positive cells, DNA fragmentation and the

Nonetheless, despite the deleterious impact of 6-OHDA on cell viability, it was sufficient to administrate the PC12 neuronal cells with resveratrol, viniferin or their combination to inhibit 6-OHDA-induced apoptosis. Of these polyphenols, resveratrol has been widely described for its antioxidant properties [30–33] which are mainly dependent on its ability to induce key transcription factors such as nuclear factor erythroid 2-related factor (Nrf2) and antioxidant enzymes, including glutathione S-transferase (GST) [33,34], all crucial antioxidants defenses [35]. Given the role of oxidative stress in the pathogenesis of neurodegenerative diseases, it represents a promising target to counter neuronal loss [36]. Indeed, resveratrol has been reported to exert neuroprotective effects both in vitro and in vivo [15,37–41], which are in agreement with the results reported herein, most

underlie by the ability of this polyphenol to counteract the activation of the apoptotic cascade induced by 6-OHDA in neuronal cells and inhibit cytotoxicity induced by LPSactivated microglia in a microglia–neuron co-culture system. Although our in vitro model could not perfectly replicate in vivo neuron–microglia physiology, these results enlighten that when viniferin was mixed with its homologous resveratrol, it also proved to be more effective in countering neurotoxicity induced by pro-inflammatory activated N9 microglia. Remarkably, despite the fact that final concentration of the mixture of the two polyphenols matched the concentration of the single polyphenols used in isolation, its effect was more powerful in preventing 6-OHDA-induced increase in caspase-3/TUNELpositive PC12 neurons and inhibiting cytotoxicity promoted by LPS-activated N9

#### **4. Discussion**

This study provides evidence on the neuroprotective and anti-inflammatory potential of the polyphenol viniferin in a cellular model of PD, PC12 dopaminergic neurons, and N9 microglia–PC12 neurons co-culture system [9,10]. These effects were underlie by the ability of this polyphenol to counteract the activation of the apoptotic cascade induced by 6-OHDA in neuronal cells and inhibit cytotoxicity induced by LPS-activated microglia in a microglia– neuron co-culture system. Although our in vitro model could not perfectly replicate in vivo neuron–microglia physiology, these results enlighten that when viniferin was mixed with its homologous resveratrol, it also proved to be more effective in countering neurotoxicity induced by pro-inflammatory activated N9 microglia. Remarkably, despite the fact that final concentration of the mixture of the two polyphenols matched the concentration of the single polyphenols used in isolation, its effect was more powerful in preventing 6-OHDAinduced increase in caspase-3/TUNEL-positive PC12 neurons and inhibiting cytotoxicity promoted by LPS-activated N9 microglia, than resveratrol or viniferin alone.

6-OHDA, due to its similarities to endogenous catecholamines, is taken up and accumulates in catecholaminergic neurons promoting neurotoxicity mainly via oxidative stress-related mechanisms [28]. In support of this, its administration directly into the brain via stereotactic surgery induces parkinsonism in rodent models by promoting nigrostriatal pathway degeneration [29]. These effects were recapitulated in the present study in which 6-OHDA promoted cytotoxicity, as indicated by an increase in LDH release from PC12 dopaminergic neurons and a decrease in their metabolic activity. Importantly, these neurotoxic effects are dependent on the activation of apoptosis as indicated by an increase in caspase-3/TUNEL-positive cells, DNA fragmentation and the cleavage of caspase-3 and PARP, as our present results illustrate.

Nonetheless, despite the deleterious impact of 6-OHDA on cell viability, it was sufficient to administrate the PC12 neuronal cells with resveratrol, viniferin or their combination to inhibit 6-OHDA-induced apoptosis. Of these polyphenols, resveratrol has been widely described for its antioxidant properties [30–33] which are mainly dependent on its ability to induce key transcription factors such as nuclear factor erythroid 2-related factor (Nrf2) and antioxidant enzymes, including glutathione S-transferase (GST) [33,34], all crucial antioxidants defenses [35]. Given the role of oxidative stress in the pathogenesis of neurodegenerative diseases, it represents a promising target to counter neuronal loss [36]. Indeed, resveratrol has been reported to exert neuroprotective effects both in vitro and in vivo [15,37–41], which are in agreement with the results reported herein, most likely dependent on the ability of resveratrol to dampen 6-OHDA-induced oxidative stress triggered by mitochondrial dysfunction [42]. We also investigated in our cell culture system whether a similar effect could be exerted by the naturally occurring resveratrol dimer, viniferin, and if these molecules could act synergically. We report that viniferin was able to inhibit 6-OHDA-induced neurotoxicity and apoptosis to the same extent as resveratrol. These effects, in parallel to those exerted by resveratrol, may be dependent on the antioxidant properties of viniferin [43]. Furthermore, the results reported as part of this study are corroborated by previous evidence supporting the neuroprotective role of viniferin, not only in ADmodels [18,44], but also in PD cellular models [45]. In this regard, in a PD cell model represented by a human neuroblastoma SH-SY5Y cell line exposed to rotenone, viniferin reduced cellular apoptosis, oxidative stress and restored mitochondrial homeostasis, all underlain by SIRT3-mediated fork headbox O3 (FOXO3) deacetylation [45], with FOXO3 playing an important role in PD progression and dopaminergic neuron survival [46]. While this study further supports our results and provide mechanistic insights into the role of viniferin, our data highlight the potency of viniferin considering it was able to inhibit apoptosis at a concentration of 10 nM compared to previous reports using 1 µM for 24 h [45]. Remarkably, the ability of viniferin to counteract apoptosis appeared to be potentiated by resveratrol, with these two molecules acting jointly to increase cellular metabolic activity and decrease the number of caspase-3/TUNEL-positive cells following exposure to 6-OHDA, in our cellular paradigm. Even if the results from cellular paradigms cannot be

directly transferred in vivo, we believe that the combination effect of these two molecules represents a promising nutraceutical combination to investigate in in vivo models of PD.

Inflammation represents another pivotal pathogenetic mechanism in PD, underpinned by microglia over-activation which contributes to dopaminergic neuron loss [8,47–49]. In fact, intranigral injection of LPS, a known pro-inflammatory component of Gram-negative bacteria, promotes dopaminergic neuron death recapitulating the pathogenetic features of PD in rodent models [50,51]. This is corroborated with our co-culture model where LPS-activated N9 microglia cells promoted cytotoxicity in downstream neuronal PC12 cells in the absence of a direct cell-to-cell contact, further supporting the role of inflammation in promoting neurotoxicity and its involvement in PD pathogenesis. Furthermore, in consideration of the lack of a direct contact between N9 cells and neurons in our co-culture system, the neurotoxic effect elicited by microglia is mediated by its secretory milieu. Indeed, the activation of the Toll like receptor 4 (TLR4) by LPS in microglial cells triggers the activation of pro-inflammatory pathways, including nuclear factor κB (NFκB) which promotes the secretion of pro-inflammatory cytokines, ROS and nitric oxide via the activation of NADPH oxidase and inducible nitric oxide synthase, respectively [52]. These cytotoxic factors, may also represent the mediators linking microglia activation with PC12 neurons cytotoxicity [53], as indicated in our data by an increase in IL-1α and TNF-α secretion by N9 microglia cells upon exposure to LPS. Remarkably, we demonstrated that resveratrol, viniferin or their combination was able to counteract microglia-induced neurotoxicity both indirectly when these polyphenols were applied to N9 cells, as well as directly when they were used to treat PC12 neurons. This indicates that resveratrol and viniferin were both able to modulate microglia secretory milieu, reducing the cytotoxic effect of microglia secretome by eliciting adaptive protective responses directly in neurons. In either case, resveratrol and viniferin exert anti-inflammatory effects and trigger molecular mechanisms aimed at protecting cells against inflammation and oxidative stress, possibly relying on the activation of the transcription factor nrf2 and the deacetylating enzyme SIRT-3 [17,34]. Particularly, in microglial cells, the activation of nrf2 by resveratrol may be responsible for the inhibition of NFκB via a cross-talk between these transcription factors [54], leading to the consequent downregulation of the inflammatory response elicited by LPS. Similarly, viniferin has also been shown to exert anti-inflammatory responses by inhibition NFκB signaling [55]. Thus, the ability of these polyphenols to restrain activated-microglia from promoting cytotoxicity in PC12 neurons may be dependent on their capacity to dampen LPS-induced inflammation and the consequent release of neurotoxic factors from N9 cells, including nitric oxide, ROS and cytokines. Nonetheless, despite cytokines being a potential driver of microglia-induced neurotoxicity, they are not significantly downregulated by resveratrol or viniferin in our co-culture system. In light of this, these polyphenols may also modulate N9 secretion of neurotoxic mediators such as ROS, nitric oxide and reactive nitrogen species [6,13,52,56,57] that, however, were not measured in our study. Resveratrol and viniferin also exerted direct neuroprotective effect on PC12 neurons co-cultured with activated microglia. This action may rely on the activation of a transcriptional reprogramming triggered by polyphenols in PC12 neurons in order to induce genes involved in cellular defenses, which may be dependent on nrf2 and SIRT-3 activation [17,34]. However, this paradigm remains to be fully elucidated.

To conclude, these polyphenols not only directly protect PC12 neurons against 6- OHDA, but also exert their neuroprotective properties by inhibiting the neurotoxic effect promoted by LPS-activated microglia. Moreover, the neuroprotective potential of resveratrol and viniferin is potentiated when they are used in combination, suggesting a synergistic effect. Thus, considering that the foodstuff rich in resveratrol also contains other forms of stilbenes, including viniferin, and taking into account the neuroprotective potential of their combination, the consumption of these foods may represent a valuable nutritional intervention in complementary therapies for neurodegenerative diseases.

**Author Contributions:** D.S.: performed the experiments and wrote the manuscript; A.G., J.B. and E.T.-P.: performed the experiments, J.R.: supervised the experiments; J.G.: extracted the polyphenols; M.-G.M. designed, supervised and wrote the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by NSERC (National Science and Engineering Research Council) of Canada, Discovery grant no. 04321 to M.-G.M.

**Data Availability Statement:** All data are available upon reasonable request to M.-G. Martinoli.

**Acknowledgments:** The authors would like to thank Melodie B. Plourde for excellent technical assistance.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


*Article*

## **E**ff**ect of a Nutritional Support System (Diet and Supplements) for Improving Gross Motor Function in Cerebral Palsy: An Exploratory Randomized Controlled Clinical Trial**

**Fernando Leal-Martínez <sup>1</sup> , Denise Franco <sup>1</sup> , Andrea Peña-Ruiz <sup>1</sup> , Fabiola Castro-Silva <sup>2</sup> , Andrea A. Escudero-Espinosa <sup>2</sup> , Oscar G. Rolón-Lacarrier <sup>3</sup> , Mardia López-Alarcón 4 , Ximena De León 1 , Mariana Linares-Eslava <sup>1</sup> and Antonio Ibarra 1,\***


Received: 27 July 2020; Accepted: 1 October 2020; Published: 13 October 2020

**Abstract:** Background: Most patients with cerebral palsy (CP) do not respond to physical therapy due to deterioration in their nutritional status, secondary to gastrointestinal disorders and the catabolic state of the disease itself. However, basic treatments only contemplate the energy requirements and do not consider supplementation with glutamine, zinc, selenium, colecalciferol, spirulina, omega 3 or even vegetal proteins. Objective: In this study, we determined the effect of using a nutritional support system (NSS): diet and supplements, on the gross motor function in children with CP with spastic diparesic and Gross Motor Function Classification System III (GMFCS III). Methods: An exploratory study was performed. Thirty patients (from 4 to 12 years old) were randomly assigned to: (1) dietary surveillance (FG), (2) deworming and WHO diet (CG), or (3) deworming and the NSS (IG). Gross motor function was evaluated using the gross motor function measure (GMFM) scale. Results: The IG-treated group presented a significant improvement in standing and walking parameters analyzed in the GMFM compared with FG and CG groups. Fifty percent of the IG-treated patients managed to walk, while in the other groups, no patients were able to walk. Conclusions: The NSS used in the present work improves gross motor function and promotes walking in patients with CP.

**Keywords:** child; cerebral palsy; diet modification; motor development delay; nutrition disorder; nutritional support

#### **1. Introduction**

Cerebral palsy (CP) is the most common physical disability in childhood. Its prevalence worldwide and in developed countries is approximated to 2–2.5 cases per 1000 live births [1,2]. According to the Center for Disease Control (CDC), the cost per person of CP was estimated at around 921,000 dollars per year in the United States (2003), while the costs for general medical assistance amounted to

11,500 million dollars [3]. CP is a permanent movement disorder characterized by a persistent postural tone, causing limitation of activity. This is attributed to non-progressive damage on a developing and immature brain that is originated in the fetal, perinatal period (greater percentage), or first years of life. CP is accompanied by alterations in sensation, cognition, communication, perception, spasticity, seizures, disorders in swallowing, and malnutrition [4].

Traditional treatment for patients with CP generates a recovery of 2% [5] per year and is based on rehabilitation, botulinum toxin therapy, general care, and in the case of malnutrition or disorders in swallowing, nutritional support which is based on the adaptation of Krick to the Schofield formula [5,6]. Nevertheless, there are no specific recommendations for dietary intake in patients with CP. Moreover, the literature information about nutritional support for improving neurological function is scarce. That is why CP continues to be the focus of numerous investigations. In line with this, a novel strategy that includes nutrition and neurological remodeling support could be of valuable use for patients with CP since, at the same time that it nourishes, it significantly impacts the neurological recovery of the patient, especially motor function. Gross motor dysfunction is one of the main affectations in children with CP [7], and it is associated with spasticity, which reflects the damage in the Pyramidal System. This movement-alteration is also strongly associated with malnutrition in patients with CP [8]. Currently, the use of functional foods, supplements or even the administration of probiotics or prebiotics as a therapeutic strategy has become a very important research area [9]. In this field, numerous studies have tested—separately—the effect of some nutrients in the central nervous system (CNS); however, there are no investigations integrating these elements as a whole treatment in CP. Arginine, for instance, participates in the formation of nitric oxide (NO) which is associated with neuronal regeneration and protection of the CNS [10]. Docosahexaenoic acid (DHA) and sphingosine 1 phosphate (S1P) prevent the early death of newly generated neurons [11]. Omega 3 polyunsaturated fatty acids (PUFAs n-3) participate in brain plasticity, neurogenesis, and memory and brain repair. Neuromuscular alterations after cerebral ischemia, improve after omega-3 treatment [12,13]. On the other hand, supplementation with probiotics has also been proposed as a therapeutic strategy for alleviating CNS pathologies [14]. Probiotics have been used to stimulate the production of neurotransmitters, memory, neuroregeneration, and also for correcting malabsorption [15]. In this work, probiotics were used for the latter purpose.

In order to provide patients with CP with the best therapeutic strategy, it is also important to consider the use of several metabolic rescue pathways to produce energy, including lactate and alanine cycles, where glutamine and glutamic acid are the main substrates [16]. Deficiencies in plasma concentrations of iron, folate, niacin, calcium, vitamins D and E, zinc, selenium, and proteins have been in reported in children with CP, even in children who were being supplemented. Therefore, supplementation of some nutrients and proteins should also be considered [17–19]. Regarding protein supplementation, it is relevant to mention that protein plant supplements such as Spirulina Maxima, in addition to providing a high protein intake, produce compounds such as PUFAs and are also a good source of vitamins [20].

Finally, it is relevant to consider that 3500 million people in the world are parasitized and of them, the majority are children [21]. This, and other factors such as the catabolic increment observed in neurologic patients, difficulties for feeding mechanics, gastrointestinal alterations, and the use of drugs that compete with the transport, absorption, or metabolism of nutrients, increases the risk of malnutrition in children with CP [22,23]. Therefore, parasitosis makes malnutrition an even more significant problem in this group of people.

It can be said, therefore, that there are several factors that need to be addressed in order to establish the best therapeutic strategy for CP. At the moment, there is no scientific evidence of an integrative nutritional support. That is why, in the present study, we tested an integral food system—the nutritional support system (NSS)—that includes nutrition, metabolic support, deworming, and neurological remodeling support.

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

The exploratory study was designed as a controlled clinical trial which was randomized, and blinded; children with CP with spastic diparesis and GMFCS III were enrolled in this study (during a period of 3 years) and treated at the Children's Telethon Rehabilitation Center (CRIT) in Tlalnepantla Estado de México. The duration of the study was 13 weeks for each participant. *Foods* **2020**, *9*, x FOR PEER REVIEW 3 of 15 a period of 3 years) and treated at the Children's Telethon Rehabilitation Center (CRIT) in Tlalnepantla Estado de México. The duration of the study was 13 weeks for each participant.

The trial complied with the principles of the Declaration of Helsinki and the Mexican norm NOM-012-SSA3-2012 for scientific research in humans. All procedures were approved by the Committee of Research of the Faculty of Health Sciences of the Universidad Anáhuac México Campus Norte with the number 2014/03001. The parent or guardian and the patient (in the event that they could sign) signed the informed consent letter voluntarily. The Clinical Trials identifier of this study is: NCT03933709. The trial complied with the principles of the Declaration of Helsinki and the Mexican norm NOM-012-SSA3-2012 for scientific research in humans. All procedures were approved by the Committee of Research of the Faculty of Health Sciences of the Universidad Anáhuac México Campus Norte with the number 2014/03001. The parent or guardian and the patient (in the event that they could sign) signed the informed consent letter voluntarily. The Clinical Trials identifier of this study is: NCT03933709.

#### *2.1. Participants 2.1. Participants*

*2.2. Group Assignments* 

Fifty-three children were interviewed and, from this group, thirty met the inclusion criteria (see Figure 1). Fifty-three children were interviewed and, from this group, thirty met the inclusion criteria (see Figure 1).

**Figure 1.** Consort, transparent reporting of trials. **Figure 1.** Consort, transparent reporting of trials.

Inclusion criteria: Patients with CP with spastic diparesis and GMFCS III (these patients demonstrate elevated functionality in the categories of lying/rolling, sitting, and crawling), of both genders aged between 4 and 12 years old, who had support from a full-time caregiver, and who were able to feed orally. Inclusion criteria: Patients with CP with spastic diparesis and GMFCS III (these patients demonstrate elevated functionality in the categories of lying/rolling, sitting, and crawling), of both genders aged between 4 and 12 years old, who had support from a full-time caregiver, and who were able to feed orally.

Non-inclusion criteria: Patients with CP presenting another catabolic disease which could increase the risk of malnutrition (renal, cardiovascular, pulmonary, hepatic, or immunological disease), and those presenting infections or receiving antibiotic treatment at least 15 days before starting the study; patients who had received therapy with botulinum toxin or muscle relaxants in the last 6 months; patients with CP presenting severe gastroesophageal reflux or with any type of surgery performed in the last 9 months; and patients walking by themselves. Non-inclusion criteria: Patients with CP presenting another catabolic disease which could increase the risk of malnutrition (renal, cardiovascular, pulmonary, hepatic, or immunological disease), and those presenting infections or receiving antibiotic treatment at least 15 days before starting the study; patients who had received therapy with botulinum toxin or muscle relaxants in the last 6 months; patients with CP presenting severe gastroesophageal reflux or with any type of surgery performed in the last 9 months; and patients walking by themselves.

Before recruiting patients, thirty numbers (from 1 to 30) were placed in a tombola. Afterwards, each number was randomly selected from the tombola and sequentially allocated to one of three groups, until reaching 10 numbers per group. Each number represented the time the patient was to

#### *2.2. Group Assignments*

Before recruiting patients, thirty numbers (from 1 to 30) were placed in a tombola. Afterwards, each number was randomly selected from the tombola and sequentially allocated to one of three groups, until reaching 10 numbers per group. Each number represented the time the patient was to be enrolled. This way, once initiated the recruitment, the first patient was assigned number one; the second was number two, and so on. In this fashion, each patient was allocated (according to the time of enrollment) to the group to which, their number corresponded.

Afterwards, the patients were informed and trained according to their corresponding groups: (1) the Follow-up Group (FG, *n* = 10)—only their usual diet was monitored; (2) the Control Group (CG, *n* = 10) was dewormed and received the nutritional therapy recommended by the WHO [23]; (3) the Intervention Group (IG, *n* = 10) was dewormed and received the NSS. As parasitosis is a common variable that could affect the absorption of nutrients among the groups and, as a part of our hypothesis, is related to the positive effect that deworming could have on the absorption of supplements, we decided to contrast the IG group versus a dewormed (CG) group and a non-dewormed (FG) group.

All participants received Bobath physical therapy.

#### *2.3. Procedures*

Once the participants were selected, parents and children with CP were called to explain the protocol and to collect their informed consent forms. Upon entering the study, the nutritional history and gross motor function measure (GMFM) scale qualification was obtained and recorded from each patient. Thereafter, patients with CP were randomly allocated into the 3 groups (FG, CG, and IG). The energy calculation for the CG and IG groups was performed with the Krick formula (50% carbohydrates, 30% lipids, and 20% proteins).

The parents and/or caregivers of the participants were trained on the project, procedures, as well as feeding and supplementation schemes if required. They were also warned to avoid any comment on the treatment that the patient received. At the beginning of the treatment, the CG and IG groups were dewormed with nitazoxanide at a dosage of 7.5 mg/kg every 12 h for 3 days.

#### *2.4. Diet, Shakes, and Supplementation*

Our nutritional support system consists of a shake-based diet with functional ingredients, high levels of vegetables, fruits, cereals, root-vegetables, and fish. Additionally, it is supplemented with glutamine, arginine, folic acid, nicotinic acid, zinc, selenium, cholecalciferol, ascorbic acid, spirulina, vegetal protein, PUFAs n-3, and probiotics (Saccharomyces Boulardii; 200 mg every 12 h for 3 days in the basal period and at week 7, for correcting malabsorption). In order to ensure that patients reached the necessary caloric intake, they were instructed to take two shakes in the morning and one in the evening. Shakes were made at home by the parents and consumed immediately after preparation. A shake-based diet was chosen since it facilitates nutrient absorption in the intestine. Envelopes containing the ingredients for the shakes were provided to encourage consumption of the shakes and to make it easier to adhere to the therapy. The IG supplements were mixed in numbered envelopes and administered in three shakes: Shakes 1 and 3 contained amaranth (1<sup>1</sup> 2 tablespoon), oats (1<sup>1</sup> 2 tablespoon), 1/4 of a medium avocado, 1 medium-sized banana, egg whites (two), cinnamon (1 g) and almond milk (250 mL)—equivalent to 650 kcal, 14 g of protein, 11 g of lipids, and 58 g of carbohydrates. Shake 2 contained celery (1/2 of the stalk), oranges (two), pineapple (100 g), nopal (1/2 of a medium leaf), 1 sprig of parsley, radish (1/4 of a medium radish), and ginger (3 g) equivalent to 148 kcal, 3.5 g of protein, 3.5 g of lipids, and 36 g of carbohydrates and was to be drunk in the morning together with shake 1 (see Figure 2).The ingredients and administration of the envelopes were as follows: Envelope 1 contained 4.9 g of Spirulina Maxima, 100 mg ascorbic acid, 5 mg folic acid, and 10 g of glutamine. This envelope was to be added to shake 1 during the first 10 days of the intervention. Envelope 2 contained 1g PUFAs n-3 and was to be added to shake 2 which was

given throughout the intervention. Envelope 3 contained 4.9 g of Spirulina Maxima, 100 mg ascorbic acid, 5 mg folic acid, 5.2 g vegetable protein, 125 mg nicotinic acid, 50 mg zinc, 100 mcg selenium and 800 UI cholecalciferol. This envelope was to be added to shake 1 from day 11 until the end of week 6, after which it was suspended for 10 days and substituted for envelope 5 and then to be retaken until the end of the intervention. Envelope 4 contained 1 g arginine and was to be added to shake 3 from day 8 until the end of the intervention. Envelope 5 contained the same ingredients as envelope 3 with an additional 10 g glutamine and was to be added to shake 1 from the start of week 7 for 10 days, after which envelope 3 was restarted. Ingredients in envelopes provide diverse beneficial effects [24–27] (see Figure 2). *Foods* **2020**, *9*, x FOR PEER REVIEW 5 of 15 after which envelope 3 was restarted. Ingredients in envelopes provide diverse beneficial effects [24–


**Figure 2.** Schematic overview of ingredients, administration, and expected function of shakes and envelopes. (**a**) Administered only the first 10 days of therapy; (**b**) administered from day 11 until the end of week 6, after which it was suspended and substituted for envelope 5 (10 days), and then, it was retaken until the end of the intervention; (**c**) administered only during 10 days after the 6th week of therapy, after which envelope 3 was restarted; (**d**) administered throughout the intervention; (**e**) administered from day 8 until the end of intervention. **Figure 2.** Schematic overview of ingredients, administration, and expected function of shakes and envelopes. (a) Administered only the first 10 days of therapy; (b) administered from day 11 until the end of week 6, after which it was suspended and substituted for envelope 5 (10 days), and then, it was retaken until the end of the intervention; (c) administered only during 10 days after the 6th week of therapy, after which envelope 3 was restarted; (d) administered throughout the intervention; (e) administered from day 8 until the end of intervention.

The total supplementation provides 287.28 kcal, 39.92 g of protein, 20 g of carbohydrates, and 4.96 g of lipids. The micronutrients that it provides are 3.81 g of fiber, 207.58 mg of sodium, 194 mg of calcium, 155 mg of phosphorus, 76 mg of magnesium, 6 mg of iron, 228.8 mg of potassium, 141.28 mg of zinc, 400 ug of selenium, 600 mg of vitamin A, 420 mg of vitamin C, 1600IU of vitamin D, 2.16 mg of vitamin E, 3.58 mg of vitamin B1, 0.76 mg of vitamin B2, 288.5 mg of vitamin B3, 8.81 mg of vitamin B5, 0.12 mg of vitamin B6, and 0.04 mg of vitamin B12. The total supplementation provides 287.28 kcal, 39.92 g of protein, 20 g of carbohydrates, and 4.96 g of lipids. The micronutrients that it provides are 3.81 g of fiber, 207.58 mg of sodium, 194 mg of calcium, 155 mg of phosphorus, 76 mg of magnesium, 6 mg of iron, 228.8 mg of potassium, 141.28 mg of zinc, 400 µg of selenium, 600 mg of vitamin A, 420 mg of vitamin C, 1600IU of vitamin D, 2.16 mg of vitamin E, 3.58 mg of vitamin B1, 0.76 mg of vitamin B2, 288.5 mg of vitamin B3, 8.81 mg of vitamin B5, 0.12 mg of vitamin B6, and 0.04 mg of vitamin B12.

#### *2.6. Follow up 2.5. Follow up*

*2.7. GMFM Assessment* 

The participants attended the CRIT twice a week to receive physical therapy and once a week for a clinical and nutritional review. In the nutritional review every week with the parents or caregivers, it was checked that the patient had no gastrointestinal problems and that they tolerated the oral route well. Food diaries were also reviewed, and patients handed over empty supplement envelopes to verify adherence. On days the patients attended physical therapy, they were supervised to ensure that they had breakfast, a mid-morning snack, and lunch. The participants attended the CRIT twice a week to receive physical therapy and once a week for a clinical and nutritional review. In the nutritional review every week with the parents or caregivers, it was checked that the patient had no gastrointestinal problems and that they tolerated the oral route well. Food diaries were also reviewed, and patients handed over empty supplement envelopes to verify adherence. On days the patients attended physical therapy, they were supervised to ensure that they had breakfast, a mid-morning snack, and lunch.

blinded, so they were warned to avoid any comment on the treatment that the patient received.

The GMFM scale was performed at baseline time and weeks 7 and 13 after intervention. The

This scale assesses five general parameters: 1. Lying (decubitus) and rolling over (GMFAV), 2. Sitting (GMFB), 3. Crawling and kneeling (GMFC), 4. Standing (GMFD), 5. Walking (GMFE), and one

#### *2.6. GMFM Assessment*

The GMFM scale was performed at baseline time and weeks 7 and 13 after intervention. The evaluators and the CRIT staff were the blinded aspects of the study as they did not have access to any information about the treatment each child was receiving. The patients and their families were not blinded, so they were warned to avoid any comment on the treatment that the patient received.

This scale assesses five general parameters: 1. Lying (decubitus) and rolling over (GMFAV), 2. Sitting (GMFB), 3. Crawling and kneeling (GMFC), 4. Standing (GMFD), 5. Walking (GMFE), and one final total item (GMFF). The scoring system consists of 88 items, and each one is valued based on the following criteria: 0 = No, 1 = start, 2 = Partially Complete, 3 = Complete, NE = Not evaluated [28]. The literature has reported a high convergent validity (0.91) of the GMFCS, on GMFM [29]. Robert Palisano, the author of the scale, attended the CRIT EDO MEX to carry out training to the evaluator of this study.

#### *2.7. Statistical Methods*

In order to know the distribution of the data, the test of Shapiro–Wilk was applied. Analysis of data was performed using the Mann Whitney U test or the Kruskal Wallis followed by the Dunn posthoc test. In order to calculate the size of the effect on standing and walking evaluations, we used Rosenthal's r, an effect size test for data with non-normal distribution. This test can be used alongside Mann Whitney U-test results. The level of significance was considered as <0.05 in all cases.

#### *2.8. Sample Size*

As this investigation was an exploratory study, sample size was determined by the feasibility of recruitment. A sample of 10 children per group was established considering that the number of patients per year at CRIT—with the criteria required for the study—fluctuates between 10 and 15 patients. This sample size allows the detection of an effect size of 0.1 or larger. In order to reach the established sample, we required a three-year period.

#### **3. Results**

Fifty-three patients were interviewed. From this group, only thirty were recruited and studied from January 2015 until February 2018. The trial was stopped at this time to evaluate the preliminary results. The analysis of evaluations was made on the original assigned groups (three groups, 10 patients per group; see Figure 1). The demographic characteristics of the patients studied are summarized in Table 1.

As the adherence to the course of treatment was strictly supervised, more than 90% of patients complied with the intervention therapy. Each patient was requested to keep a food diary, which enabled us to establish their caloric intake per day. The average of kcal consumed at the start of the study in FG was 1292.5 kcal; in GC, 1495 kcal; and in IG, 1447.5 kcal. In week 13, the average kcal consumption was 1305 kcal in the FG group, 1113 kcal in CG, and 2898.8 kcal in IG.

#### *3.1. Baseline Results*

The baseline values of GMFM parameters (lying, sitting, crawling, standing, and walking) were not statistically different among the studied groups: lying (FG: 92.75 ± 3.44; CG: 93.34 ± 2.92; IG: 84.13 ± 8.87;mean ± SEM; *p* = 0.950; Kruskal Wallis followed by Dunn's posthoc test), sitting (FG: 61.34 ± 5.24; CG: 74.15 ± 5.49; IG: 75.16 ± 8.09; *p* = 0.089), crawling (FG: 49.52 ± 5.56; CG: 47.38 ± 8.90; IG: 61.86 ± 7.79; *p* = 0.248), standing (FG: 13.44 ± 3.46; CG: 15.66 ± 3.25; IG: 16.79 ± 4.58; *p* = 0.666), walking (FG: 15.49 ± 1.51; CG: 18.05 ± 1.90; IG: 20.27 ± 3.24; *p* = 0.365).


**Table 1.** Clinical characterization of patients with cerebral palsy (CP).

FG, follow group. CG, control group. IG, intervention group. F, female. M, male. GMFD, gross motor. Function standing. GMFE, gross motor function walking.

#### *3.2. Lying, Sitting, and Crawling Evaluations*

Throughout the follow-up, lying and sitting parameters did not show any relevant difference among the groups. In the seventh week, the groups presented very similar values. Lying: FG: 95.69 ± 4; CG: 94.89 ± 7; IG: 83.92 ± 29; mean ± SEM; *p* = 0.982; Kruskal–Wallis followed by Dunn's test. Sitting: FG: 73.83 ± 2; CG: 94.89 ± 7; IG: 83.92 ± 29; *p* = 0.114. In the 13th week, no significant difference was observed among the groups. Lying: FG: 97.63 ± 2; CG: 95.56 ± 5; IG: 89.2 ± 19; mean ± SEM; *p* = 0.960; Kruskal–Wallis followed by Dunn's test. Sitting: FG: 77.33 ± 12; CG: 83.50 ± 15; IG: 90.60 ± 16; *p* = 0.082. With respect to crawling, there was not any significant difference at the seventh week: FG: 61.30 ± 19; CG: 57.29 ± 31; IG: 76.87 ± 28; *p* = 0.143, Kruskal–Wallis followed by Dunn's test. Nevertheless, at the thirteenth week, evaluations presented a significant improvement in IG patients. Crawling: FG: 57.72 ± 17; CG: 54.38 ± 30; IG: 69.86 ± 26; *p* = 0.03, Kruskal–Wallis followed by Dunn's test.

#### *3.3. Standing and Walking Evaluations*

Figure 3 shows the comparison of the FG, CG, and IG patients in the standing parameter at 7 and 13 weeks after the intervention. A total score was considered for these evaluations. Seven weeks after intervention, IG patients presented a significant improvement in standing (26.19 ± 8.2; mean ± SD) compared with FG (10.1 ± 4.9; *p* = 0.0004, Mann Whitney U test) and CG (12.98 ± 7.2; *p* = 0.003, Mann Whitney U test) groups. In order to strengthen these results, we calculated the Rosenthal's r, a statistical assessment that evaluates the effect size (clinical efficacy of the treatment). Results showed

a large size effect in the standing parameter of the IG group (r = 0.64). Thirteen weeks after intervention, the improvement continued being statistically different in IG-patients (30.90 ± 8.7) compared to FG (13.32 ± 7.4, *p* = 0.0002, Mann Whitney U test) and CG (17.29 ± 8.1, *p* = 0.003 Mann Whitney U test) groups. In this case, we also calculated Rosenthal's r and found a large effect in the IG group (*r* = 0.61). *Foods* **2020**, *9*, x FOR PEER REVIEW 8 of 15

**Figure 3.** GMFD (standing) evaluations of FG, CG, and IG. (**A**) Standing evaluations on the seventhweek show a significant improvement of IG-patients compared to FG and CG ones. Analysis was performed by the Mann Whitney U test, \* *p* = 0.003 \*\* *p* = 0.0004. (**B**) The standing evaluations at 13thweek show a significant improvement of the IG compared to the FG and CG groups. Analysis by the Mann Whitney U test, \* *p* = 0.003 \*\* *p* = 0.0002. Bars represent mean ± SD of 10 patients. Baseline values **Figure 3.** GMFD (standing) evaluations of FG, CG, and IG. (**A**) Standing evaluations on the seventh-week show a significant improvement of IG-patients compared to FG and CG ones. Analysis was performed by the Mann Whitney U test, \* *p* = 0.003 \*\* *p* = 0.0004. (**B**) The standing evaluations at 13th-week show a significant improvement of the IG compared to the FG and CG groups. Analysis by the Mann Whitney U test, \* *p* = 0.003 \*\* *p* = 0.0002. Bars represent mean ± SD of 10 patients. Baseline values are also shown.

are also shown. Figure 4 shows the comparison among FG, CG, and IG patients in walking parameter at 7 and 13 weeks after the intervention. Total score was considered. Seven weeks after intervention, there was a significant difference in walking improvement in IG-patients (24.65 ± 6.1) relative to FG (18.47 ± 5.7; *p* = 0.01, Mann Whitney U test) and CG (19.45 ± 5.6; *p* = 0.03, Mann Whitney U test) ones. Rosenthal's r test showed a large effect size in the IG group (r = 0.37). After thirteen weeks, IG-patients continued presenting a significant improvement (34.68 ± 7.3) as compared to the FG (19.86 ± 6; *p* = 0.0003, Mann Whitney U test) and CG (22.51 ± 5.9; *p* = 0.001, Mann Whitney U test) patients. We also calculated Rosenthal's r for this data set. Results showed a large size effect of treatment in IG-patients on walking parameter at thirteen weeks (r = 0.67). Finally, from this group of patients (IG), five managed to walk by themselves (with no support from anyone else; see Videos S1–S3 provided as Figure 4 shows the comparison among FG, CG, and IG patients in walking parameter at 7 and 13 weeks after the intervention. Total score was considered. Seven weeks after intervention, there was a significant difference in walking improvement in IG-patients (24.65 ± 6.1) relative to FG (18.47 ± 5.7; *p* = 0.01, Mann Whitney U test) and CG (19.45 ± 5.6; *p* = 0.03, Mann Whitney U test) ones. Rosenthal's r test showed a large effect size in the IG group (r = 0.37). After thirteen weeks, IG-patients continued presenting a significant improvement (34.68 ± 7.3) as compared to the FG (19.86 ± 6; *p* = 0.0003, Mann Whitney U test) and CG (22.51 ± 5.9; *p* = 0.001, Mann Whitney U test) patients. We also calculated Rosenthal's r for this data set. Results showed a large size effect of treatment in IG-patients on walking parameter at thirteen weeks (r = 0.67). Finally, from this group of patients (IG), five managed to walk by themselves (with no support from anyone else). No patient from the FG or CG groups achieved walking independently.

Supplementary Material). No patient from the FG or CG groups achieved walking independently.

**Figure 4.** GMFE (walking) evaluations of FG, CG, and IG. (**A**) On the seventh week of evaluation, walking parameters showed a significant improvement in IG-patients. Analysis by the Mann Whitney U test. \* *p* = 0.03 \*\* *p* = 0.01. (**B**) Walking evaluations on the 13th week showed a significant improvement of IG-patients compared with those of FG and CG. Analysis was performed using the **Figure 4.** GMFE (walking) evaluations of FG, CG, and IG. (**A**) On the seventh week of evaluation, walking parameters showed a significant improvement in IG-patients. Analysis by the Mann Whitney U test. \* *p* = 0.03 \*\* *p* = 0.01. (**B**) Walking evaluations on the 13th week showed a significant improvement of IG-patients compared with those of FG and CG. Analysis was performed using the Mann Whitney U test, \* *p* = 0.001 \*\* *p* = 0.0003. Bars represent mean ± SD of 10 patients. Baseline values are also shown.

#### Mann Whitney U test, \* *p* = 0.001 \*\* *p* = 0.0003. Bars represent mean ± SD of 10 patients. Baseline values are also shown. *3.4. Percentage of Evolution*

test).

*3.4. Percentage of Evolution*  Figure 5 shows the comparison of the percentage of evolution in standing among the studied groups. Percentage was calculated according to the formula: final corresponding value (7 or 13 weeks)/baseline value −1 × 100). Therefore, we are reporting the percentage of evolution after 7 or 13 weeks of intervention. IG-patients showed a significant improvement in the percentage of evolution compared to both FG and CG patients (*p* = 0.025, Kruskall Wallis followed by Dunn's posthoc test). Figure 5 shows the comparison of the percentage of evolution in standing among the studied groups. Percentage was calculated according to the formula: final corresponding value (7 or 13 weeks)/baseline value − 1 × 100). Therefore, we are reporting the percentage of evolution after 7 or 13 weeks of intervention. IG-patients showed a significant improvement in the percentage of evolution compared to both FG and CG patients (*p* = 0.025, Kruskall Wallis followed by Dunn's posthoc test). The evolution towards the thirteenth week also revealed a significant difference in the improvement of IG-patients compared with FG and CG ones (*p* = 0.03, Kruskall Wallis followed by Dunn's posthoc test).

The evolution towards the thirteenth week also revealed a significant difference in the improvement of IG-patients compared with FG and CG ones (*p* = 0.03, Kruskall Wallis followed by Dunn's posthoc

Wallis followed by Dunn's test).

**Figure 5.** Percentage of evolution (GMFD, standing) of FG, CG, and IG. (**A**) The evolution from the baseline to the seventh week shows a significant improvement of standing ability in IG in comparison with the one observed in FG and CG patients. \* *p* = 0.025, Kruskal Wallis test. (**B**) The evolution from the baseline towards the 13th week shows significant improvement in standing ability of IG compared **Figure 5.** Percentage of evolution (GMFD, standing) of FG, CG, and IG. (**A**) The evolution from the baseline to the seventh week shows a significant improvement of standing ability in IG in comparison with the one observed in FG and CG patients. \* *p* = 0.025, Kruskal Wallis test. (**B**) The evolution from the baseline towards the 13th week shows significant improvement in standing ability of IG compared to FG and CG patients. \* *p* = 0.03, Kruskal Wallis test. Black line represents the mean of 10 patients.

When the walking parameter was analyzed, the evolution from the baseline to the seventh week did not present any significant advantage for any of the groups (Figure 6; *p* = 0.15, Kruskall Wallis followed by Dunn's test). Nevertheless, in the evolution observed up to the thirteenth week, IG patients presented a significant improvement, compared with FG and CG patients (*p* = 0.03, Kruskall When the walking parameter was analyzed, the evolution from the baseline to the seventh week did not present any significant advantage for any of the groups (Figure 6; *p* = 0.15, Kruskall Wallis followed by Dunn's test). Nevertheless, in the evolution observed up to the thirteenth week, IG patients presented a significant improvement, compared with FG and CG patients (*p* = 0.03, Kruskall Wallis followed by Dunn's test).

to FG and CG patients. \* *p* = 0.03, Kruskal Wallis test. Black line represents the mean of 10 patients.

**Figure 6.** Percentages of evolution (GMFE, walking) of FG, CG, and IG. (**A**) The comparison of the percentages of evolution from the baseline to the seventh week does not show significant improvement in the progress of studied groups. Analysis was performed by using the Kruskal Wallis test. (**B**) The percentage of evolution from the baseline towards the 13th week shows significant improvement of the progress of IG compared to FG and CG patients., \* *p* = 0.029, Kruskal Wallis test. **Figure 6.** Percentages of evolution (GMFE, walking) of FG, CG, and IG. (**A**) The comparison of the percentages of evolution from the baseline to the seventh week does not show significant improvement in the progress of studied groups. Analysis was performed by using the Kruskal Wallis test. (**B**) The percentage of evolution from the baseline towards the 13th week shows significant improvement of the progress of IG compared to FG and CG patients., \* *p* = 0.029, Kruskal Wallis test. Black line represents the mean of 10 patients.

#### Black line represents the mean of 10 patients. **4. Discussion**

**4. Discussion**  In this work, we evaluated the effect of an NSS for improving motor function in patents with CP. Our findings showed that the NSS promoted a significant improvement in standing and walking parameters of the GMFM scale. Moreover, NSS-therapy caused half of the patients studied to walk by themselves. This is a relevant finding since the improvement in motor function was, in a shorter time, above the one observed in patients with CP undergoing only conventional therapy [5]. From the initial evaluation (7 weeks after the onset of therapy), the performance of IG-patients was better than the one observed in CG and FG groups, which presented a performance that was similar to that reported in the literature [30]. The improvement of IG-treated patients was evident thirteen weeks after therapy initiation. The results observed in IG-treated patients went beyond the ones reported before using only physical therapy (2–3% of annual advance) [5]. It is of relevance to mention that even though the baseline values of IG-patients were slightly higher, the difference among the groups was not statistically significant. In order to avoid any confusion on this matter, we performed an analysis of the percentage of improvement among the different groups (this analysis included the baseline values). Therefore, we are reporting the percentage of evolution after 7 or 13 weeks of In this work, we evaluated the effect of an NSS for improving motor function in patents with CP. Our findings showed that the NSS promoted a significant improvement in standing and walking parameters of the GMFM scale. Moreover, NSS-therapy caused half of the patients studied to walk by themselves. This is a relevant finding since the improvement in motor function was, in a shorter time, above the one observed in patients with CP undergoing only conventional therapy [5]. From the initial evaluation (7 weeks after the onset of therapy), the performance of IG-patients was better than the one observed in CG and FG groups, which presented a performance that was similar to that reported in the literature [30]. The improvement of IG-treated patients was evident thirteen weeks after therapy initiation. The results observed in IG-treated patients went beyond the ones reported before using only physical therapy (2–3% of annual advance) [5]. It is of relevance to mention that even though the baseline values of IG-patients were slightly higher, the difference among the groups was not statistically significant. In order to avoid any confusion on this matter, we performed an analysis of the percentage of improvement among the different groups (this analysis included the baseline values). Therefore, we are reporting the percentage of evolution after 7 or 13 weeks of intervention. The analysis confirmed that the results were not influenced by the baseline values.

intervention. The analysis confirmed that the results were not influenced by the baseline values.

The encouraging outcome of IG-treated patients could be due to the administration of diverse nutrients and supplements that are necessary in order to compensate the metabolic, nutritional, and neurological deficiencies presented by patients with CP.

Neurological patients regularly present an increased catabolism [31]; however, in some investigations, it is reported that some patients, especially those with CP, present a lower energy requirement than healthy patients although specific nutritional information, including proteins, is not known [32]. It was observed during this investigation that the energy consumption in the baseline of all the groups was around 1300 to 1400 kcal, and the nutriments intake increased in the IG patients around 2890 kcal per day on the 13th week, being the double of kcal intake in comparison with the other groups. It is likely that after deworming, supplementing, and consuming a specific diet, they were able to increase their calorie consumption. The NSS was the only different variable among the groups. Therefore, NSS could be—at least in part—the cause of this increment. Nevertheless, other factors—non-identified at this moment—could also be participating. Further studies should be designed to corroborate these results, find other possible factors, and to establish new nutritional recommendations in CP.

The observed improvements in standing and walking can be related to a reduction in spasticity. Changes were also observed in hand movements in the majority of IG patients, which resulted in the release of the thumb and pincer grip from the "claw hand". Spasticity is related to a lesion in the Pyramidal System (PS); therefore, any improvement in these areas can be associated with a remodeling of the PS and thereby, of the CNS—in particular, in the areas of motor and premotor cortex as well as brainstem. This remodeling could be related with myelination, neurogenesis, promotion of release in neurotransmitters, or even the generation of ganglia [30]. The PS is made up of approximately one million fibers—mainly myelinated whose origins are in the primary motor and premotor cortex (80% of the fibers). Their function is to control voluntary movements, both fine and gross motor skills, in such a way that any alteration in these fibers can cause, amongst other alterations, spasticity [33].

Therefore, any therapy based on PS remodeling could be contributing to diminished spasticity and thereby, to improve fine and gross motor skills in patients with CP.

The induction or increment of plasticity phenomena, or even of regenerating neurons by providing substrates as nutrients, is extremely viable. A number of different nutrients, such as PUFAs n-3 (EPA and DHA), have been employed separately to stimulate plasticity and new neuronal cell bodies [30]. Nutrients such as zinc, ascorbic acid, spirulina, arginine, PUFAs n-3, and vitamin D have all been used in the repair of neuronal cell bodies and in neuronal regeneration [34,35]. Nevertheless, there is no scientific evidence from studies using an integrative nutritional support as a therapeutic strategy in CNS diseases. Moreover, investigations combining at least 3 or more nutrients are not available in the scientific literature.

On the other hand, it is known that the production of neurotransmitters such as serotonin and GABA can be increased using probiotics, nicotinic acid, cobalamine, pyridoxine, folates, zinc, ascorbic acid, and triptofane [36]. Regulating the production of serotonin and GABA can help to control spasticity and thus, fine and gross motor functions. The use of probiotics has become relevant in improving signaling from the intestine and favoring neurotransmission, metabolism, immunity, and inflammation through stimulating the microbiota-gut-brain axis [37].

Recently, studies on gut microbiota have become relevant in the treatment of various neurological disorders, including autism, alterations in memory, and neuronal regeneration [11]. It is also known that simultaneous supplementation of pro and prebiotics, such as inuline and other nutrients such as glutamine, induce the generation of short chain fatty acids which modulate inflammation and favor regeneration of the nervous system [38] as well as reestablishing enterocytes and muscular synthesis [39].

Finally, studies in mice showed that glutamine together with probiotics inhibit nitric oxide, reducing levels of proinflammatory factors such as TNF alfa, IL6, IL8, and reactive oxygen species

(ROS) [40]. The evidence demonstrates that combining probiotics with other nutrients to create symbiosis induces positive effects on immunity, reestablishing mucose and the nervous system.

Although the results of this investigation are encouraging, it is important to consider the constraints of the study. Perhaps the main limitation of this work is the number of evaluated patients. In the present investigation, we could not gather more than 10 individuals per group. This was mainly because the patients did not meet the inclusion criteria. The use of other therapeutic alternatives (i.e., botulinum toxin or muscle relaxants) is very common in patients with CP, and this did not allow us to include them into the study. Further research needs to be undertaken in a larger sample to support the preliminary findings of this study. Additionally, we will implement evaluations such as anthropometry, electromyography, and gait and movement analysis.

On the other hand, it is important to carry out this investigation in a research center with a controlled system, where children and their caregivers could coexist closely during the 13 weeks of the protocol. Finally, despite the convincing clinical findings, it is important to complement further research with sensitive analytical methods to evaluate neuroregeneration or neurogenesis.

Regardless of all these constraints, our findings deserve to be considered for future investigations, since they are the result of a well-controlled and designed study, which, by nature, can be replicated and carried out in other populations. Additionally, the results of this study indicate a positive relationship between NSS and gross motor skills in CP. This means that children with CP who have high-energy supplementary foods, combined with micronutrients in special doses, will further improve gross motor skills. Ati et al. [41] reported similar observations: they found a significant relationship between nutritional status and gross motor children's development. Finally, it is worth mentioning that in this investigation, nutrition was used as a stimulus to repair the nervous system—specifically, the pyramidal system. This is an innovative strategy with no previous similar studies. This exploratory investigation managed to induce better gross motor recovery in a shorter period of time as compared to conventional treatment.

Future studies should be directed to expand and confirm the results of this study.

#### **5. Conclusions**

Together, the results of this investigation provide insights into the positive effect of NSS on CP. We speculate that the observed results are mainly a consequence of a significant reduction in spasticity, which could be the result, at least in part, of a PS remodeling. In this scenario, the effect of all of the NSS components is crucial; however, we consider that the participation of PUFAs, zinc, ascorbic acid, spirulina, or arginine could be critical to repair neural function.

**Author Contributions:** Conceptualization, F.L.-M., M.L.-A., and A.I.; data curation, F.L.-M.; formal analysis, A.I.; funding acquisition, F.L.-M. and A.I.; investigation, F.L.-M., F.C.-S., and A.A.E.-E.; methodology, F.L.-M., D.F., and A.P.-R.; project administration, F.L.-M., X.D.L., and M.L.-E.; resources, A.I.; supervision, O.G.R.-L.; writing—review and editing, F.L.-M. and A.I. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was partially funded by the Centro de Investigación en Ciencias de la Salud (CICSA), FCS, Universidad Anáhuac Campus Norte, grant number 2014/03001; Valdecasas Laboratorios y Alimentos Esenciales para la Humanidad. The funders had no role in study design, data collection, and analysis, decision to publish, or preparation of the manuscript.

**Acknowledgments:** We would like to thank the authorities at Centro de Rehabilitación e Inclusión Infantil Teletón (CRIT) for lending us their facilities and their staff, the students who gave their time, and Keri Craig for helping to review this article.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **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/).

*Article*

## **The E**ff**ect of L-Theanine Incorporated in a Functional Food Product (Mango Sorbet) on Physiological Responses in Healthy Males: A Pilot Randomised Controlled Trial**

#### **Jackson Williams <sup>1</sup> , Andrew J. McKune 1,2,3, Ekavi N. Georgousopoulou 4,5,6, Jane Kellett <sup>1</sup> , Nathan M. D'Cunha <sup>1</sup> , Domenico Sergi <sup>7</sup> , Duane Mellor <sup>8</sup> and Nenad Naumovski 1,\***


Received: 27 February 2020; Accepted: 18 March 2020; Published: 23 March 2020

**Abstract:** Consumption of L-Theanine (L-THE) has been associated with a sensation of relaxation, as well as a reduction of stress. However, these physiological responses have yet to be elucidated in humans where L-THE is compared alongside food or as a functional ingredient within the food matrix. The aim of this study was to determine the physiological responses of a single intake of a potential functional food product (mango sorbet) containing L-THE (ms-L-THE; 200 mgw/*w*) in comparison to a flavour and colour-matched placebo (ms). Eighteen healthy male participants were recruited in this randomised, double-blind, placebo-controlled trial. The participants were required to consume ms-L-THE or placebo and their blood pressure (BP) (systolic and diastolic), heart rate (HR), and heart rate variability (HRV) were monitored continuously over 90 minutes. Eleven males (age 27.7 ± 10.8 years) completed the study. Changes in area under the curve for systolic and diastolic blood pressure and HRV over the 90 minute observation period indicated no differences between the three conditions (all *p* > 0.05) or within individual groups (all *p* > 0.05). The values for heart rate were also not different in the placebo group (*p* = 0.996) and treatment group (*p* = 0.066), while there was a difference seen at the baseline (*p* = 0.003). Based on the findings of this study, L-THE incorporated in a food matrix (mango sorbet) demonstrated no reduction in BP or HR and showed no significant parasympathetic interaction as determined by HRV high-frequency band and low-frequency/high-frequency ratio. Further studies should be focussed towards the comparison of pure L-THE and incorporation within the food matrix to warrant recommendations of L-THE alongside food consumption.

**Keywords:** L-Theanine; amino acid; green tea; bioactive; functional food; blood pressure; heart rate variability; cardiometabolic effect

#### **1. Introduction**

Stress is a dynamic condition that acts as a stimulus, affecting arousal factors of individuals in response to challenges and aversive situations. The consequences of stress are strongly influenced by how an individual perceives and appraises the situation in addition to the intensity of the event that triggers the stress response [1–3]. Traditionally, consumption of tea is associated with relaxation [3], beneficial health outcomes [4], and an increase in longevity, which is proposed to be due to a number of different bioactive constituents [5,6]. Prior to its consumption, tea leaves undergo processing and based on the type of processing, predominantly leads to seven different types of tea; 'green' (unfermented), 'yellow', 'white', 'oolong' (partially fermented), 'black' (completely fermented), 'aged pu-erh' (drastically fermented and aged), and 'ripened pu-erh' tea [7]. The type of processing also affects the formation of different compounds, including but not limited to the composition and degradation of the molecules found in green tea. In green tea, the steaming (or heating) process inactivates the enzyme polyphenol oxidase, which prevents the further fermentation of tea leaves, consequently producing a stable and dry product. During this process, oxidative polymerisation of monomeric flavan-3-ols is inhibited (unlike in the black and other teas), and green tea typically contains higher levels of these flavan-3-ols. Therefore, it can be anticipated that different processing techniques will also influence the levels of other molecules and putative bioactives, such as amino acids.

Derived from the leaves of the *Camellia sinensis* species, green tea is consumed due to its favourable taste and also for its reported therapeutic properties such as cardiovascular benefits, increased relaxation effects and strong, anti-inflammatory [8] and antioxidant effects [6,7,9]. These effects are linked to its many bioactive constituents, including polyphenols [10], particularly catechins [11], and γ-glutamylethylamide, commonly referred to as L-Theanine (L-THE), which is the most abundant and unique non-proteinogenic amino acid in green tea.

The consumption of L-THE is associated with a sensation of relaxation, and is proposed to elicit acute stress-reducing effects [3] as well as providing the characteristic *'umami'* flavour [6]. In addition, the consumption of L-THE is associated with antihypertensive [12,13], anti-inflammatory properties [8] as well as improvements in cognitive functioning in healthy adults [14] and when consumed in combination with caffeine [10]. Studies report stress lowering effects, expressed physiologically in the form of reductions in heart rate (HR), observed post-completion of a stress-inducing mental arithmetic task after the consumption of L-THE (200 mg) in its relatively pure form [15]. Furthermore, reductions in salivary cortisol post L-THE consumption indicate the putative involvement of L-THE in hypothalamic-pituitary-adrenal (HPA) axis responses have also been reported [15,16].

The autonomic nervous system (ANS), consisting of the sympathetic nervous system (SNS) and parasympathetic nervous systems (PNS) [17,18], are of central importance to the homeostatic stress response. Blood pressure is a physiological biomarker that is highly dependant on catecholaminergic neurons [19]. The HR is also considered to be a vital indicator of the stress response, however, the changes in interval between the highest points (R-R interval) of the heartbeat observed on the electrocardiogram (usually referred to as QRS complex), known as heart rate variability (HRV), is becoming a reliable marker for the responses caused by internal and external stressors. Changes in HRV, representing ANS cardiac regulation, can be directly (*via* ANS innervation of the myocardium) or humorally (catecholamines related hormones) mediated [20]. The high frequency (HF) component of HRV ranges from 0.15-0.4 Hz, and is based on respiratory sinus arrhythmia and mediated predominantly by PNS activity, whereas low frequency (LF) ranges between 0.04-0.15 Hz, is mediated by both PNS and SNS [21]. Furthermore, the state of being stressed affects HRV domains such as HF, LF and LF/HF ratios, which are used to understand the activity of the PNS and further as an index of autonomic balance [22].

It is proposed that the 'healthier' the ANS, the better an individual will be able to exhibit high allostatic resilience [23]. As such, HRV changes are reported to be a potential marker of the relationship between food consumption and disease [24]. These findings were further supported in studies where certain components of dietary intake are associated with higher HRV, for instance the Mediterranean

diet, which consists highly of fish and omega-3 fatty acid consumption, fresh produce, [25] as well as healthy weight maintenance [26] all contribute to the potential increase in HRV. On the other hand, reduced HRV is associated with an increased risk for a range of diseases including but not limited to cardiovascular disease [27] and diabetes [28]. This contributes to the possibility that HRV markers provide a fast and convenient way to measure the potential benefits of health status in response to various dietary interventions.

Regardless of the method of delivery of this non-proteinogenic amino acid, L-THE needs to be easily accessible and bioavailable in concentrations sufficient to produce the desired physiological effects. Human studies in which consumption of pure L-THE was provided in capsule form, indicate the most physiologically relevant doses of L-THE range from 0.05−0.4 g, equivalent to a consumption of 2−15 cups of green tea [6,29,30]. One method to deliver these dosages of L-THE is via the integration of pure L-THE in the food matrix as a potential functional food product. Based on the findings of our previous review [6], we propose that incorporating 200 mg of L-THE in a functional food product will provide a convenient and palatable delivery method of a physiologically relevant dose of L-THE. Therefore, the aim of this study was to determine the physiological responses HR, HRV, and blood pressure (BP) in healthy males following the one-off acute ingestion of a potential functional food product (mango sorbet) containing 200 mg of pure L-THE.

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

#### *2.1. Participants*

Participants were informed about the study protocols, provided written consent, and then screened to determine eligibility for participation. Following previously conducted research using similar design and products [11], inclusion criteria included healthy male participants aged 18−65 years old. Participants were excluded from participating if they consume: any functional foods including stanol or sterol ester-containing margarines; weight loss supplements, or commercial dietary products associated with weight loss; currently had or have had any known active pulmonary, hematologic, hepatic, gastrointestinal, renal, premalignant, malignant illnesses; have diabetes (type 1 and type 2); or any thyroid dysfunction.

#### *2.2. Procedure*

A randomised, double-blind, placebo-controlled, crossover design was used to determine the acute physiological HR, HRV, and BP effects after consumption of mango sorbet containing L-THE (200 mg). This study was approved by the Human Research Ethics Committee of the University of Canberra (HREC-193–2018) and written informed consent was obtained from all participants.

Participants were required to attend four clinic visits (Figure 1), initial screening (clinic 1), a baseline measurements clinic (clinic 2) to familiarise participants, and two blinded food consumption clinics (clinics 3 and 4). In the last two clinics, participants consumed the 100 g of the food products mango sorbet, one a placebo without active ingredient (ms) or the treatment mango sorbet containing 200 mg of L-THE (ms-L-THE). The findings from animal studies suggested that L-THE brain levels are increased 30−60 minutes post-consumption. Thus, assessment of the physiological parameters mentioned below occurred immediately post L-THE food product consumption to capture the period and curve at which L-THE exerts its effects.

Allocated randomised treatment sequences were achieved using 4 blocks of random sequencing numbers per intervention (randomizer.org) to ensure the ms and ms-L-THE sorbets administrations were balanced between the participants. The sequence code was placed in a sealed envelope and revealed only at the end of the trial once all participants finalised all of their visits. A 48 h washout period was used between clinics 3 and 4 ensuring the adequate time for the L-THE clearance from the body [29]. Prior to attending clinics 2, 3, and 4, participants were required to fast overnight (at least 8

h) except for water consumption. Participants were also asked to refrain from alcohol for 24 h, and caffeine 12 h prior the commencement of the clinic. the body [29]. Prior to attending clinics 2, 3, and 4, participants were required to fast overnight (at least 8 h) except for water consumption. Participants were also asked to refrain from alcohol for 24 h, and caffeine 12 h prior the commencement of the clinic.

*Foods* **2020**, *9*, x FOR PEER REVIEW 4 of 14

period was used between clinics 3 and 4 ensuring the adequate time for the L-THE clearance from

**Figure 1.** Consort flowchart of clinics attended in the study for the 11 included participants. **Figure 1.** Consort flowchart of clinics attended in the study for the 11 included participants.

#### *2.3. Materials and Reagents 2.3. Materials and Reagents*

Australia).

The mango sorbet used in this study was developed based on the modified formulation described elsewhere [11]. Common mangos (Mango pieces, Coles Pty Ltd, Peru/Mexico), caster sugar (CSR, Yarraville, VIC, Australia), were purchased from local commercial suppliers. The L-THE (Suntheanine™, Taiyo Kagaku Co., Ltd, Yokkaichi Japan) was purchased from Ingredient Resources Australia and New Zealand Pty/Ltd (Sydney, NSW, Australia) and was the functional additive (200 mg/100 g *w*/*w*) to the active product. The whey protein concentrate (WPC) was purchased from Professional Whey Pty/Ltd (Erina, NSW, Australia) and added to provide structural stability and rigidity to the food product. The sorbet was selected as the food matrix of choice for this study due to the preference for storage stability of L-THE such as low temperature (less than 4 °C) and stability in an acidic environment (pH range 5−6) [31]. The dosage of L-THE (200 mg) was selected based on the findings of previous studies [29]. The nutritional composition of the ms-L-THE (Table 1), including the macro- and micronutrient data were analysed using the Australian Food Composition Database integrated into the FoodWorks Professional Software (v9, Xyris Software, Brisbane, QLD, The mango sorbet used in this study was developed based on the modified formulation described elsewhere [11]. Common mangos (Mango pieces, Coles Pty Ltd, Peru/Mexico), caster sugar (CSR, Yarraville, VIC, Australia), were purchased from local commercial suppliers. The L-THE (Suntheanine™, Taiyo Kagaku Co., Ltd, Yokkaichi Japan) was purchased from Ingredient Resources Australia and New Zealand Pty/Ltd (Sydney, NSW, Australia) and was the functional additive (200 mg/100 g *w*/*w*) to the active product. The whey protein concentrate (WPC) was purchased from Professional Whey Pty/Ltd (Erina, NSW, Australia) and added to provide structural stability and rigidity to the food product. The sorbet was selected as the food matrix of choice for this study due to the preference for storage stability of L-THE such as low temperature (less than 4 ◦C) and stability in an acidic environment (pH range 5−6) [31]. The dosage of L-THE (200 mg) was selected based on the findings of previous studies [29]. The nutritional composition of the ms-L-THE (Table 1), including the macroand micronutrient data were analysed using the Australian Food Composition Database integrated into the FoodWorks Professional Software (v9, Xyris Software, Brisbane, QLD, Australia).

*Foods* **2020**, *9*, x; doi: FOR PEER REVIEW www.mdpi.com/journal/foods


**Table 1.** Nutrient profile of the mango sorbet with L-Theanine per single serve (100 g).

The total nutrient profile for the mango sorbet containing the active ingredient L-Theanine (L-THE) expressed as a per 100 g serve. note the mango sorbet (ms) did not contain 200 mg of L-THE.

#### *2.4. Outcomes*

#### 2.4.1. Blood Pressure

Participants' blood pressure was determined following the guidelines for the 2nd Australian National Blood Pressure study [32]. Blood pressure was determined using the finger cuff (non-invasive BP nano, AD Instruments, Dunedin, New Zealand), using two inter-inflation cuffs that were placed on the participants in a sedentary position. Continuous readings for systolic, diastolic blood pressure were recorded, as well as inter-beat interval, HR, and mean arterial pressure over 90 minutes. The data were smoothed based on the AD instruments smoothing algorithm to remove outlying data points due to finger movement and further segmented into 18 distinct values from each 5 min interval [11].

#### 2.4.2. Heart Rate Variability

Each participant was provided with a fitness-tracking device heart rate belt (Suunto® T6, Vantaa, Finland) that measures HR and R-R intervals of continuous electrocardiogram QRS heartbeat complexes. Participants were instructed how to use the device, ensuring the belt was placed below the chest muscles across the sternum after water was applied to the undersurface of the belt to ensure conduction [33]. Participants were required to be in a sedentary position for 10 min to establish resting measurements. During each clinic, participants were required to wear the heart rate belt in a sedentary position for the entirety of the clinic. Recordings for all clinics began after 10 minutes of participants being in a sedentary position. However, during consumption clinics, recordings began after complete consumption of each respective sorbet.

#### *2.5. Statistical Analysis*

All data obtained were analysed as a 5 min average up to 90 min. The obtained R-R interval data were downloaded from the HR monitors via the Suunto T6 Team Management Software (v2.1, Vantaa, Finland) and exported as a text file for time domain and spectral HRV analysis using the Kubios HRV Standard 3.0.2 diagnostic device software (Kuopio, Finland). All R-R interval artefacts were manually corrected using the methodology described by Sookan and McKune (2012). Using total area under the curve (AUC), the non-parametric independent samples Kruskal–Wallis ANOVA was the primary model for analysis for time effects for between-group time relationships the entire 0−90 min block, and further analysed between each 5 min period up to 90 min also using Kruskal–Wallis analysis. Both are presented as medians, 25th and 75th quartiles with Chi-square and *p*-value. For related samples, Friedman's two-way ANOVA by ranks was applied to test same group relationships for the entirety of 90 min, presented as Chi-square and *p*-value. The HF HRV values were log-transformed prior to analysis. Level of significance was set at α = 0.05. Data were analysed using the SPSS v25 (Armonk, IBM Corp, New York, USA).

#### **3. Results**

Initially, eighteen healthy male adults were recruited with 17 completing all four study visits. Following the completion of the study, data were analysed from 11 participants (age 27.7 ± 10.8 years, weight 90.6 ± 16.8 kg) due to the following; one participant voluntarily discontinued the trial due to lost contact; data of five participants were excluded due to insufficient data points collected due to equipment malfunction, whilst one participant was excluded post-study due to a clinical diagnosis of high blood pressure. All descriptive statistics for systolic and diastolic BP, heart rate, HF HRV including 1st and 3rd quartiles, median, total AUC values described in Table 2.



The non-parametric frequencies median, 1st and 3rd quartiles and total AUC (expressed as a sum of values from the 0−90 minute period) values for systolic and diastolic BP, systolic/diastolic BP, HR, high-frequency HRV, and LF/HF HRV ratio for the 11 included participants. \* Indicates significant *p* < 0.05.

#### *3.1. Systolic and Diastolic Blood Pressure 3.1. Systolic and Diastolic Blood Pressure*

For systolic BP, between-group analysis for AUC did not demonstrate any significant effects for the entire 90 minute period (Figure 2A) (median; 1323 mmHg, 25th; 1157 mmHg, 75th; 1462 mmHg, *p* = 0.839). Further analysis of the individual 18 × 5 minute time points showed no significant differences between the three clinic visits (*p* > 0.05). The within-group analysis provided all non-significant results respectively for the entirety of the 90 minute groups (all *p* > 0.05). For systolic BP, between-group analysis for AUC did not demonstrate any significant effects for the entire 90 minute period (Figure 2A) (median; 1323 mmHg, 25th; 1157 mmHg, 75th; 1462 mmHg, *p* = 0.839)*.* Further analysis of the individual 18 × 5 minute time points showed no significant differences between the three clinic visits (*p* > 0.05). The within-group analysis provided all nonsignificant results respectively for the entirety of the 90 minute groups (all *p* > 0.05). For diastolic BP, between-group AUC analysis showed no significant differences during the

*Foods* **2020**, *9*, x FOR PEER REVIEW 8 of 14

For diastolic BP, between-group AUC analysis showed no significant differences during the entirety of the 90 minutes (Figure 2B) amongst either of the conditions (median; 1323 mmHg, 25th; 1157 mmHg, 75th; 1462 mmHg, *p* = 0.922). Similarly, between-group analysis of the individual 18 × 5 minute time points also showed no significant differences between each of the three clinic visits (all *p* > *0.05*). Same group analysis test indicated no significant differences within each of the conditions (all *p* > 0.05). entirety of the 90 minutes (Figure 2B) amongst either of the conditions (median; 1323 mmHg, 25th; 1157 mmHg, 75th; 1462 mmHg, *p* = 0.922*).* Similarly, between-group analysis of the individual 18 × 5 minute time points also showed no significant differences between each of the three clinic visits (all *p > 0.05*). Same group analysis test indicated no significant differences within each of the conditions (all *p* > 0.05).

**Figure 2.** The changes in physiological parameters: systolic blood pressure (**A**); diastolic blood pressure (**B**), heart rate (**C**) and log of high-frequency heart rate variability (**D**) over the 90 minute period for the 11 included particpants. **Figure 2.** The changes in physiological parameters: systolic blood pressure (**A**); diastolic blood pressure (**B**), heart rate (**C**) and log of high-frequency heart rate variability (**D**) over the 90 minute period for the 11 included particpants.

#### *3.2. Systolic/Diastolic Ratio 3.2. Systolic*/*Diastolic Ratio*

The ratio of systolic/diastolic AUC analysis showed no significant differences during the entirety of 90 min between either of the conditions (median; 32.3, 25th; 31.1, 75th; 35.7, *p* = 0.970)*.* Further, analysis of the individual 18 × 5 min time points showed no significant differences between each of the three conditions (all *p* > 0.05). Same condition analysis showed no significant differences within the baseline group (*p* = 0.761), whereas significant differences were observed in the ms-L-THE group (*p* = 0.003) and the ms group (*p* = 0.003). The ratio of systolic/diastolic AUC analysis showed no significant differences during the entirety of 90 min between either of the conditions (median; 32.3, 25th; 31.1, 75th; 35.7, *p* = 0.970). Further, analysis of the individual 18 × 5 min time points showed no significant differences between each of the three conditions (all *p* > 0.05). Same condition analysis showed no significant differences within the baseline group (*p* = 0.761), whereas significant differences were observed in the ms-L-THE group (*p* = 0.003) and the ms group (*p* = 0.003).

#### *3.3. Heart Rate 3.3. Heart Rate*

Between-group AUC analysis showed no significant differences during the entire 90 minutes between either of the conditions (median; 1325, 25th; 1222, 75th; 1385, *p* = 0.996)*.* Further analysis of the individual 18 × 5 min time points also showed no significant differences between each of the three conditions (all *p* > 0.05), whereas within-group analysis showed significant differences within the Between-group AUC analysis showed no significant differences during the entire 90 minutes between either of the conditions (median; 1325, 25th; 1222, 75th; 1385, *p* = 0.996). Further analysis of the individual 18 × 5 min time points also showed no significant differences between each of the three conditions (all *p* > 0.05), whereas within-group analysis showed significant differences within the baseline group (*p* = 0.003) as well as within the ms-L-THE group (*p* = 0.066) (Figure 2C). No significant differences were observed in the ms group (*p* = 0.060).

*Foods* **2020**, *9*, x; doi: FOR PEER REVIEW www.mdpi.com/journal/foods

#### *3.4. Sympathetic Activity*

#### 3.4.1. High-frequency HRV

AUC HF analysis showed no significant time effects between groups (median; 97.3, 25th; 6.59, 75th; 120, *p* = 0.974). Analysis of the individual 18 × 5 min time points also showed no significant differences between each of the three conditions (all *p* > 0.05). Intergroup analysis showed no significant differences within each of the group conditions (all *p* > 0.05) (Figure 2D).

#### 3.4.2. LF/HF Ratio

LF/HF measurements using for AUC analysis during the 90 minute period showed no significant time effects between groups (median; 50.9, 25th; 35.3, 75th; 16.3, *p* = 0.704). Analysis of the individual 18 × 5 min time points between groups showed no significant differences between each of the three conditions (all *p* > 0.05). Analysis of same group comparisons showed no significant differences within each of the group conditions (all *p* > 0.05).

#### **4. Discussion**

The current study implemented a randomised double-blind, placebo-controlled cross over design to investigate the effects of L-THE incorporated into a whey protein-based mango sorbet as a potential functional food product. To our knowledge, this study is the first to examine the effect of a functional food product containing L-THE to determine its physiological effectiveness using a continuous set of BP and HRV measurements.

The observed results for the ms-L-THE did not produce any significant physiological changes for the parameters of systolic and diastolic BP across the entire 90 minute testing period in comparison to the ms or baseline measurements. Similar to other studies where external factors such as stress or caffeine are used to raise BP, our intervention was designed to cause fluctuations in BP, HR, and HRV due to the carbohydrate response attributed to sorbet consumption [34]. Despite showing initial significant interactions, the systolic/diastolic ratio was a good indicator of the BP effect of both sorbets for within-subject effects. This indicated that consumption of both ms and ms-L-THE caused a postprandial response, in turn, causing the systolic/diastolic to show significant differences as supported by our analysis. Pharmacokinetic clinical data [35] that involved the ingestion of pure L-THE (25−200 mg) suggest peaks in physiological and blood plasma effects occur between 32 and 50 min post oral ingestion (in a fasted state) [29]. Although a similar trend is observed in our current study in terms of a time effect relationship for systolic BP reduction (Figure 2), our study design did not demonstrate any statistically significant physiological effects in comparison to the reported pharmokinetic trends that L-THE exhibits.

Further, as blood plasma concentrations were not studied in our trial, we can only speculate that the observed physiological trends in the current study are in line with the mentioned pharmacokinetic studies [29,35]. The same 200 mg L-THE dosage is clinically reported to acutely attenuate BP (the rise seen in high-stress-response adults), as well as reducing anxiety after stress loading tasks [13]. Additionally, the findings of a relatively recent study have indicated that consumption of L-THE (200 mg) has potential to promote mental health by providing better quality sleep and improvement in cognitive functioning by improving verbal fluency and executive functioning scores [14]. Similarly, a study measuring BP [36] at time points of 20 and 70 min respectively post 97 mg L-THE and 40 mg caffeine intake, highlighted the attenuative effects L-THE has on caffeine. However, this study was limited by reference of only two-time points [36]. Considering previously reported evidence regarding L-THE concentration in the blood post-consumption, it is demonstrated that L-THE potentially antagonises the BP effect of caffeine on participants, without affecting the alertness or mood aspects as well as slowing overall reaction time [37]. In the same study, the authors did not state whether a fasted state was implemented, whilst in comparison, the present study did not implement a cognitive task to

raise BP and may in future benefit from an external stimulus to observe the BP effects of L-THE within the food matrix [37].

The initial results suggested that HR was significantly affected using within-subject contrasts however, further analysis revealed no significant interactions. It can be proposed that consumption of both the mango sorbet interventions stabilised HR in contrast to the significant results observed solely in the baseline clinic, however further testing by comparing L-THE in its pure form against L-THE within the food matrix must be conducted to confirm this. As such, these results fall in line with the study by Dodd et al. [10], who also reported no significant differences in HR when pure L-THE was consumed via capsule intake (50 mg); however, this may have occurred due to the low dose provided to participants. In comparison, our results were not on par with those reported by Kimura et al. [15], where consumption of 200 mg L-THE dissolved in water actively resulted in the reduction of HR in response to an acute mental arithmetic stressor task. In the design of our current study, we did not implement a mental stressor task to alter HR, and in future, may benefit from adding a mentally stressing intervention such as arithmetic task. Based on this, it is safe to say that our study is the first to identify that L-THE does not have any effect on HR when consumed within a functional food product and in the absence of environmental stressors; however, from the literature, 200 mg of pure L-THE may acutely lower HR in response to stressful situations which is thought to reflect attenuation of sympathetic nervous activation [15]. Our results do not indicate a potential for L-THE to buffer HRV physiological markers for both HF and LF/HF ratio in the periods of post-consumption. Nonetheless, the presents study suggests that L-THE did reduce the variance of HF outcomes; however, this was not shown to be statistically significant.

In humans, understanding of the physiological responses attributed to L-THE intake where pure L-THE is consumed alongside everyday foods or integrated as part of a food matrix has only been partially investigated to date. Currently, only two known studies aimed at investigating the effect of L-THE consumption embedded within food have been reported [38,39]. The first investigated a nutrient-based drink containing 200 mg of L-THE as well as alpha glycerylphosphorylcholine (α GPC; 25 mg), phosphatidylserine (1 mg), and micronised chamomile (10 mg), however, no BP or HR data was recorded and this particular study reported a decrease in salivary cortisol that supports the potential anti-stress effects of L-THE [39]. The latter study [39] displayed sympathomimetic inhibitory responses post-consumption of 128 mg L-THE within a cacao (60% *w*/*w*) chocolate that resulted in a decrease in BP associated with the L-THE condition. It has been previously suggested by Kakuda et al. [40] that the L-THE mechanism of action relies on the reduction of glutamate release from pre-synapse to the synaptic cleft by acting as an inhibitor of glutamine re-uptake. Additionally, glutamine was also associated with the replenishment of the neurotransmitter pool. This, in turn, inhibits extracellular glutamine uptake by neurons, and its conversion to glutamate via glutaminase enzyme [41]. The recent review by Yoneda et al [42] proposes a possible novel in vitro neurogenic role of L-THE for brain wellness via increase in neurogenesis. Interestingly, the findings of an animal trial using a stress-sensitive strain of mice (senescence-accelerated mice prone 10) that were exposed to stress, had decreased brain volume. However, the same strain of mice that ingested L-THE (6 mg/kg), under the same experimental conditions had a suppression in brain atrophy [43]. Although this is reported to occur based on common neuroprotective outcomes related to stress [42,43], this mechanism of action and the outcomes specific to glutamate biomarkers assessed in the Kakuda et al. [44] study were not assessed in the present study. As there were no statistically significant interactions between the baseline, ms, and ms-L-THE variables, one of the potential mechanisms that may have affected the outcomes was that the L-THE may have bound to the food matrix of the ingredients used in the sorbet (WPC, sucrose, or the mango pulp). Whilst our previous findings indicate that high-performance liquid chromatography (HPLC) analysis of the L-THE extraction from mango pulp yielded a 98% extraction (unpublished data); thus, a potential rational for the integration with mangoes, the potential for L-THE binding to the WPC or sucrose cannot be overlooked. Despite this, it is equally important to acknowledge that HPLC analysis of L-THE is unlikely to reflect the biological outcomes of L-THE in

the gastrointestinal tract. Furthermore, it is difficult to determine all known bioactive constituents are present in the mango pulp as well as the WPC and sucrose, which all may affect the kinetical aspects of L-THE including but not limited to: absorption, distribution, elimination, and biotransformation. It is essential to acknowledge the unaccounted bioactive ingredients found within the mango sorbet may have lessened the effects of L-THE, however, to our knowledge, no literature suggest L-THE interacts with WPC or sucrose. Therefore, a similar study design that investigates the effects of L-THE as a mono supplement against other food matrixes is required to determine the physiological outcomes, as well as implementing techniques to monitor L-THE in the circulation as a point of reference for its bioavailability. Based on previous literature, we propose that from the results of this study, L-THE in its pure encapsulated form appears to serve as a medium for consumption to produce its known effects rather than when combined within the food matrix of mango sorbet.

#### *Limitations and Future Directions*

The lack of significant interactions observed in this study can potentially be attributed to a sample size that was reflective of the nature of this study being a pilot trial. Furthermore, due to the exclusion of data (participant exclusion), future studies should account for including a higher number of participants as well as participants with pre existing health conditions such as anxiety and high blood pressure. Furthermore, it is equally important to compare the differences between L-THE in its relatively pure form against L-THE as an integrated component of the food matrix. Although the population was male-only, we can only assume that the results are reflective of the current population included.

Further, L-THE consumed orally as a component of mango sorbet (200 mg) is a lower dose compared to the pharmacodynamic rat studies by Yokogoshi et al. [45] as well as Yokogoshi and Kobayashi [46], which were both similar in design (1500 mg/day and 2000 mg/kg of L-THE administered intraperitoneally). In future studies, administering a standardised dose adjusted for the kilograms of body weight (kg/bw) is one potential way to maximise the effects of L-THE and determine the 'ideal' and personalised dosage for healthy humans as well as potentially increasing the bioavailability of L-THE in the body. However, it is equally important to acknowledge that further research must be conducted where L-THE as a component of a functional food product is compared alongside L-THE in its pure encapsulated form to warrant any clinically relevant claims.

#### **5. Conclusions**

Our results suggest that consumption of L-THE within the functional food product mango sorbet did not cause significant changes in the physiological responses such as blood pressure and heart rate, as well as parasympathetic nervous system activation (as determined by HF and LF/HF band HRV markers). It is also important to note that reduction of physiological responses via pharmacological interventions does not necessarily mean that the capacity to cope and manage stress is increased. Further food consumption studies with longer duration of consumption should be conducted to evaluate the use of L-THE as a supplement when consumed alongside other foods. It is anticipated that the results of this study will provide baseline information in understanding the activities of the human physiology and its effect on the stress response as well as the important notion to consider the effect food of matrixes have when constructing functional food ingredients.

**Author Contributions:** Conceptualization, J.W., A.J.M. and N.N.; Data curation, N.M.D.; Formal analysis, J.W., A.J.M. and E.N.G.; Investigation, J.W.; Methodology, D.S. and N.N.; Project administration, N.N.; Supervision, A.J.M. and N.N.; Writing—original draft, J.W.; Writing—review & editing, A.J.M., E.N.G., J.K., N.M.D., D.S., D.M. and N.N. All authors have read and agreed to the published version of the manuscript.

**Funding:** This project was funded by the University of Canberra Faculty of Health seed grant funding. J.W., is supported by an Australian Government Research Training Program scholarship. N.M.D. is supported by a Dementia Australia Research Foundation PhD scholarship. All other authors declare no funding sources.

**Acknowledgments:** This research was supported by the Faculty of Health Research Support Funding, University of Canberra, Canberra, ACT, Australia.

**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/).

*Article*
