**Chemical Composition and Biological Activities of the Nord-West Romanian Wild Bilberry (***Vaccinium myrtillus* **L.) and Lingonberry (***Vaccinium vitis-idaea* **L.) Leaves**

**Bianca-Eugenia S, tefănescu 1,2, Lavinia Florina Călinoiu 2,\*, Floricut,a Ranga 3, Florinela Fetea 3, Andrei Mocan 1,4, Dan Cristian Vodnar 3,\* and Gianina Cris, an <sup>1</sup>**


Received: 4 April 2020; Accepted: 3 June 2020; Published: 5 June 2020

**Abstract:** This study was performed to evaluate and compare the in vitro antioxidant, antimicrobial, and antimutagenic activities, and the polyphenolic content of the Nord-West Romanian wild bilberry (*Vaccinium myrtillus* L.) and lingonberry (*Vaccinium vitis-idaea* L.) leaves from three different natural habitats (Smida, Turda, Borsa). In the case of both species, the flavanols level was higher in Smida habitat (altitude 1100 m), whereas quercetin derivates were more abundant in Borsa habitat (altitude 850 m). The bilberry leaf extracts contained in the highest amounts the feruloylquinic acid (59.65 ± 0.44 mg/g for Borsa habitat) and rutin (49.83 ± 0.63 mg/g for Borsa habitat), and showed relevant 2,2-diphenyl-1-picrylhydrazyl (DPPH) antioxidant activity (271.65 mM Trolox/100 g plant material for Borsa habitat, 262.77 mM Trolox/100 g plant material for Smida habitat, and 320.83 mM Trolox/100 g plant material for Turda habitat), for all the three extracts. Gallocatechin was the dominant flavanol in lingonberry species, with the highest amount being registered for Smida habitat (46.81 ± 0.3 mg/g), revealing a DPPH antioxidant activity of 251.49 mM Trolox/100 g plant material. The results obtained in the antimicrobial tests showed that the best inhibitory effect among bilberry species was attributed to the Turda (altitude 436 m) and Smida locations, against both Gram-positive and Gram-negative bacterial strains. For lingonberry, the differences in habitat did not influence the antibacterial effect, but the antifungal effect, only in the case of *Candida zeylanoides*. A strong antimutagenic effect was registered by the bilberry leaves toward *Salmonella typhimurium* TA100. Our study may be able to provide a better understanding of the correlation between natural habitat conditions and the accumulation of secondary metabolites and their related bioactivities in studied leaves.

**Keywords:** bilberry; lingonberry; polyphenols; antioxidant compounds; antimicrobial activity; antimutagenicity; altitude variations

#### **1. Introduction**

Most recent epidemiological studies have reported that certain medicinal plants can be responsible for preventing the development or evolution of several diseases [1–4]. The naturally-derived antioxidants are a topic of major interest considering their proven health effects on humans [2,5], but also to gradually replace the synthetic antioxidants that have been reported as endocrine disrupters or even carcinogenic compounds [6,7]. Dietary polyphenols have diverse therapeutic uses and several proven biological properties [1,8–11], being of important consideration to study their varieties in medicinal plants and natural foods [12].

The development of newly plant-derived functional products and nutraceuticals, known as edible sources with high antioxidant content, have been the intensively studied research topics in recent years [13]. Among them, *Vaccinium* species are constantly reported for their diversity in phenolic compounds [14–16], whereas cranberry (*Vaccinium macrocarpon* Ait.) and bilberry (*Vaccinium myrtillus* L.), being more debated than lingonberry (*Vaccinium vitis-idaea* L.), contributed to their high consumption rate under several forms: as fresh fruits, processed products, and dietary supplements. Recent literature reported that lingonberry occupies a significant position in the antioxidant and antimicrobial capacity ranking of *Vaccinium*-derived species [17,18].

Bilberry (*Vaccinium myrtillus* L.), also known as the European blueberry, and lingonberry (*Vaccinium vitis-idaea* L.), commonly known as cowberry or partridgeberry, are two small, spontaneous growing shrubs belonging to the genus *Vaccinium*, *Ericaceae* family. Their berries mature from July to September, while the ripeness time is highly affected by the site conditions, precisely altitude, and habitat type. Usually, higher altitudes generate later plant ripening when compared with lower elevations.

The bilberry and lingonberry leaves are the main by-products of berry harvesting and recent investigations [14,19] have reported a significantly higher content of phenolic compounds in the leaves and stems of *Vaccinium* species in contrast to the berries, in line with the strongest antioxidant activities registered by these aerial parts than fruits [20], indicating that they may be utilized as an alternative source of bioactive natural products for the development of food supplement, nutraceuticals, or functional food. Literature studies have shown that the leaves of bilberry and lingonberry contain fewer anthocyanins than fruits, but the content of phenolic compounds is higher in leaves than in fruits [16,21,22]. Several studies have reported the presence of hydroxycinnamic acids, flavonols, proanthocyanidins, cinchonains, and iridoids in the bilberry leaves [19,23–25]. Traditionally, bilberry leaves extracts are used for treating urinary tract affection and diabetes. Owing to the presence of various phenolic compounds, bilberry leaves also have antibacterial, anti-inflammatory, and antioxidant activities [26–28]. Chemical composition and biological properties of lingonberry leaves are similar to those of bilberry. Phenolic compounds found in lingonberry leaves are hydroxycinnamic acids, proanthocyanidins. flavonols, cinchonains, iridoids, and arbutin derivatives [14,23,27]. Extracts of lingonberry leaves have shown multiple beneficial diuretics and antiseptic properties for the urinary tract, anti-cough, phlegm removing, anti-inflammatory, neuroprotective, and antioxidant activity [21,29,30].

The genetic factor must be considered when referring to polyphenol biosynthesis in the different parts of the plant, including leaves. Moreover, the biotic and abiotic conditions may be responsible for certain variations (increases or decreases) in phenolic concentration, as reported in the recent literature for bilberry leaf and stem [19,24] and lingonberry leaf [25,31]. A multitude of environmental factors change with the altitude of the growing site, precisely precipitation, mean temperature, soil, wind speed, low- and high-temperature extremes, duration of snow cover, length of vegetation period, and intensity of radiation under clear sky conditions. Enhanced UV-B radiation and lower temperatures at high altitudes have been constantly debated as having an impact on plant secondary metabolism [32,33]. As a protective mechanism towards damage induced by excessive UV-B radiation, plants support and stimulate the biosynthesis of UV-B-absorbing phenolic compounds with an

antioxidant capacity [32,34]. The stimulation of enzymes responsible for flavonoid biosynthesis in UV-enhanced radiation experiments was highly underlined [35,36].

The latitude-related factor was discussed in particular for *V. myrtillus* L., being reported for the high influence on the quality and quantity of phenolic compounds [24,37–40], suggesting that higher phenolic amounts may be supported by northern latitudes, altitude, and sunny weather. However, most studies have aimed to investigate an individual morphological part of the bilberry plant, with fruits as most debated, and leaves in a small percentage. In this context, this study aims to provide a better understanding of the correlation between natural habitats and the accumulation of phenolic compounds in the leaves of *Vaccinium myrtillus* L. and *Vaccinium vitis-idaea* L. and their related bioactivities: antioxidant, antimicrobial, and antimutagenic. Thus, the investigation on the differences, derived from natural habitats within the same region (Nord-West), on polyphenolic content of the Romanian wild bilberry and lingonberry leaf extracts could be useful to broaden the knowledge on this field.

Considering that the chemical composition of the Nord-West Romanian wild bilberry and lingonberry leaves has never been the subject of a scientific paper to best of our knowledge, this study aimed to determine the phenolic composition of bilberry and lingonberry leaves and to measure their antioxidant, antibacterial, antifungal, and antimutagenic activities, whereas the antimutagenic and antimicrobial activities of the leaves are of significant novelty. Furthermore, the differences between the three different natural habitats of Romanian bilberry and lingonberry leaves were also investigated.

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

#### *2.1. Plant Samples and Growing Conditions*

The leaves of bilberry (*Vaccinium myrtillus* L.) and lingonberry (*Vaccinium vitis-idaea* L.) were collected in the autumn (September) of 2017 from spontaneous species of three different locations in Romania, differing in altitude and habitat type: (1) Turda (46◦32- 00" N, 23◦52- 00" E), Cluj County; (2) Smida (46◦38- 33" N, 22◦52- 49" E), Cluj County; and (3) Borsa (47◦39- 19" N, 24◦39- 47" E), Maramures County. Leaves of both species were randomly sampled from ca. 10 shrubs in the same 20 m × 20 m area for each habitat. The plant material was dried at room temperature for 7–10 days and grounded to a fine powder and kept in the dark prior to analyses. The results were calculated based on the dried and grounded plant material/powder. The numbers of Plant Voucher Specimens are VM103 and VVI105.

#### *2.2. Description of Habitats*

Turda is a municipality in the county of Cluj, Transylvania, Romania, and it is located about 30 km southeast of Cluj-Napoca. Turda developed mainly on the left side of the Aries, river. The minimum altitude is 310 m in the eastern extremity, on the Aries, valley, and the maximum is in the northeast of the city, on Slăninii Hill (436 m), from where the leaves were collected. The karst relief is present and develops into soluble rocks (limestone, salt, gypsum), being characterized by mineral soils. The climate in September is quite dry with 44 mm of rainfall, and involves a maximum temperature of 23 ◦C and a minimum of 15 ◦C. The collection place had a moderate solar exposition considering the slope exposure [41].

Smida is located in the heart of the Apuseni Natural Park, a protected area that is among the last large areas of large, forested karst (spreading its wild beauty on approximately 76,000 hectares) throughout Europe. Smida village is at an altitude of 1100 m and benefits from a moderate continental climate, whereas in September, there is a maximum temperature of 23 ◦C and a minimum of 1 ◦C, with 6.5 mm of rainfall. The soil is characterized by acid brown soils with medium texture, and good aquatic drainage considering the winters rich in snow. It possesses large areas of natural forests and meadows, with a variety of fauna and flora. The relief is a karstic one, well developed, and made

up of caves [42]. The collection place had partial sun exposure considering the open-spaced areas surrounded by forest.

The Borsa town is located in the south of Maramures, county, Transylvania, Romania at an altitude of 850 m in the Rodnei Mountains, on the Vis, eu river valley. The relief of the area is mountainous, very rugged, and with steep slopes and high-level differences, being characterized by the moderate continental climate sector, with a maximum temperature of 13 ◦C and a minimum of 8 ◦C, and with the average annual rainfall of 1100 mm and permanent exposure to the advection of the western air masses of oceanic nature, whose characteristics are reflected in the evolution of all climatic elements. The collection place is characterized by acid brown soils and good solar exposition [43].

#### *2.3. Chemicals and Reagents*

Catechin, chlorogenic acid, quercetin, cyanidin chloride, and gallic acid used as standards for the HPLC-DAD-ESI-MS analysis were purchased from Sigma-Aldrich (Steinheim, Germany). Folin–Ciocalteu's phenol reagent, sodium carbonate (Na2CO3), sodium nitrate (NaNO2), hydrochloric acid (HCl), aluminum chloride (AlCl3), sodium hydroxide (NaOH), acetic acid, acetonitrile, methanol, ethanol, DPPH (2,2-diphenyl-1-picrylhydrazyl), and Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid)were purchased from Sigma-Aldrich (Steinheim, Germany). For antimicrobial assays,Mueller–Hinton agar, thioglycollate broth with resazurin, and Mueller–Hinton broth were purchased from BioMerieux (France), and Tween 80 and Broth Malt medium were purchased from Sigma-Aldrich (Steinheim, Germany).

#### *2.4. Ultrasound-Assisted Extraction Procedure*

The fine powder obtained from the leaves (0.25 g) was extracted with 7 mL 40% *v*/*v* ethanol in water for 30 min in an ultrasonic bath, at 20 ◦C. After centrifugation (5000 rpm for 10 min at 24 ◦C), the supernatant was filtered and stored (−18 ◦C) until analysis (total phenolic content, total flavonoid content, total anthocyanin content, antioxidant, antimutagenic and antimicrobial activities, and HPLC-DAD-MS analysis).

#### *2.5. Analysis of Phenolic Compounds*

#### 2.5.1. HPLC-DAD-ESI-MS Analysis

Identification and quantification of phenolic compounds in the leave extract were performed on an HPLC-DAD-ESI-MS system consisting of an Agilent 1200 HPLC with DAD detector, coupled to an MS-detector single-quadrupole Agilent 6110. For phenolic compounds' separation, the Eclipse column, XDB C18 (4.6 × 150 mm, particle size 5 μm) (Agilent Technologies, USA), was used at 25 ◦C. The binary gradient was prepared from 0.1% acetic acid/acetonitrile (99:1) in distilled water (*v*/*v*) (solvent A) and 0.1% acetic acid in acetonitrile (*v*/*v*) (solvent B) with a flow rate of 0.5 mL/min, according to the elution program described by Dulf et al. [44]. For MS fragmentation, the ESI (+) module was used, with a scanning range between 100 and 1200 m/z, capillary voltage 3000 V, at 350 ◦C, and with a nitrogen flow of 8 l/min. The eluent was monitored by DAD, and the absorbance spectra (200–600 nm) were measured and collected during each run. For analyzing the data, Agilent ChemStation Software (Rev B.04.02 SP1, Palo Alto, CA, USA) was performed. The phenolic compounds from the extracts were identified by comparing the retention times, UV visible, and mass spectra of the peaks with four reference standards, as follows: the compounds of the flavanol subclass were quantified using the calibration curve performed with catechin standard on the concentration ranges of 10–200 μg/mL and expressed as equivalents of catechin (mg catechin/g plant material) (*r*<sup>2</sup> = 0.9985); for the hydroxycinnamic acid subclass, the compounds were quantified using the calibration curve performed with chlorogenic acid on the concentration range of 10–50 μg/mL, expressed as chlorogenic equivalents (mg chlorogenic acid/g plant material) (*r*<sup>2</sup> = 0.9937); flavonols were quantified using the calibration curve performed with quercetin on the concentration ranges of 10–200 μg/mL, expressed as quercetin equivalents (mg

quercetin/g plant material) (*r*<sup>2</sup> = 0.9951); and anthocyanins were quantified using the calibration curve made with cyanidin on the concentration ranges of 10–100 μg/mL, expressed as cyaniding equivalents (mg cyanidin/g plant material) (*r*<sup>2</sup> = 0.9951).

#### 2.5.2. Total Phenolic Content

The determination of total phenolic content (TPC) was performed by the Folin–Ciocalteu method [44,45]. Briefly, 25 μL of sample extract was combined with 125 μL of Folin–Ciocalteu reagent (0.2 N) and 100 μL of sodium carbonate solution (Na2CO3, 7.5% *w*/*v*). Afterward, the mixture was incubated for 2 h in the dark at room temperature (25 ◦C). The absorbance was recorded at 760 nm, using ethanol as blank. A standard curve was prepared using gallic acid (0.01–1 mg/mL), and the TPC in the extract was expressed as gallic acid equivalents (GAE) (mg GAE/100 g plant material).

#### 2.5.3. Total Flavonoid Content

Total flavonoid content (TFC) was determined by a spectrophotometric method [46] based on the formation of a complex flavonoid—aluminum. Shortly, 1 mL of sample extract was mixed with 0.3 mL NaNO2 (5%); after 5 min, 0.3 mL AlCl3 (10%) was added; afterward, 2 mL NaOH (1M) and water to a total volume of 10 mL. The absorbance was measured immediately, at 510 nm. A standard curve was prepared using quercetin (0.117–1 mg/mL) and the TFC was expressed as quercetin equivalents (QE) (mg QE/ 100 g plant material).

#### 2.5.4. Total Anthocyanin Content

The total anthocyanin content (TAC) was determined by UV/visible spectrophotometric method [47]. The extracts were diluted with 40% ethanol, and the absorption was measured at 530 nm using a Jasco UV-VIS Spectrophotometer (V-530 double beam, Tokyo, Japan). The anthocyanin content was estimated as cyanidin-3-glucoside at 530 nm using a molar absorptivity coefficient of 26,900 and was expressed as milligrams per 100 g of plant material [47].

#### *2.6. DPPH Free-Radical-Scavenging-Assay*

DPPH free-radical–scavenging activity was performed using the method described by Ebrahimabadi et al. [48] with slight modifications. First, 250 μL of each leaf hydroethanolic extract was mixed with 1750 μL of freshly prepared DPPH solution (0.1 mM in 40% ethanol). The absorbance was measured after 30 min of rest under dark conditions, at 517 nm, using the spectrophotometer Biotek and 40% ethanol as blank. In the DPPH assay, the antioxidant activity of the extracts was evaluated using the calibration curve performed with Trolox, and then the absorbance was recorded for all the tested extracts, to calculate the percentage inhibition (expressed as percentage inhibition of the DPPH radicals). The percentage inhibition (I%) was calculated as I% =[ (AB − AA)/ AB] × 100, where AB = absorbance of blank and AA- = absorbance of hydroethanolic extract.

#### *2.7. Antimicrobial and Antifungal Capacity*

#### 2.7.1. Stains and Cultivation Conditions

To determine the antimicrobial activity for all extracts, six bacterial strains were used: three Gram-positive bacteria: *Staphylococcus aureus* (ATCC 49444), *Enterococcus faecalis* (ATCC 29212), *Rhodococcus equi* (ATCC 6939), and three Gram-negative bacteria: *Pseudomonas aeruginosa* (ATCC 27853), *Klebsiella pneumonia* (DSMZ 2026)*, Escherichia coli enterotoxigen* (ATCC 25922). All tested microorganisms were obtained from the Food Biotechnology Laboratory, UASVM CN, Romania.

#### 2.7.2. Microdilution Technique

Evaluation of the antimicrobial activity was done according to the guidelines of the Clinical Laboratory Standards Institute (CLSI) [49], using the standard broth microdilution technique for bacteria that grow aerobically, with slight modifications. Briefly, all the bacteria were cultured on Mueller–Hinton agar, followed by their storage at 4 ◦C and subculture once a month. Before antibacterial susceptibility testing, each strain was inoculated on Mueller–Hinton agar plates and incubated at 37 ◦C for 24 h. The medium used for susceptibility testing was Mueller–Hinton broth. Inoculums (density of 0.5 in McFarland scale) were prepared in a 0.9% NaCl sterile solution. Then, tested strains were suspended in Mueller–Hinton broth medium, to give a final density of 2 <sup>×</sup> 10<sup>5</sup> colony-forming units (CFU)/mL. The inoculum was stored at 4 ◦C for further use. Determinations of minimum inhibitory concentrations (MICs) were performed by a serial dilution technique using 96-well plates. The 100 μL Mueller–Hinton broth was placed into each of the 96 wells of the microplates. Aliquots of 100 μL of each extract (concentration of 0.1 g/mL) were added into the first rows of the microplates and twofold dilutions of the extracts were made by dispensing the solutions into the remaining wells. Then, 10 μL of the culture suspensions was inoculated into the wells. We used ethanol (40%) in water as a control. The microplates were incubated for 24–48 h at 37 ◦C. The MIC of the plant extracts was detected after the addition of 20 μL (0.2 mg/mL) of resazurin solution to each well, and the plates were incubated for 2 h at 37 ◦C. A change from blue to pink indicates the reduction of resazurin and, therefore, bacterial growth. The MIC was defined as the lowest concentration of the extract that inhibited the growth of the bacterial strain [50], which respectively prevented this color change. The minimum bactericidal concentrations (MBCs) were determined by serial subcultivation of 2 μL into 96-well plates containing 100 μL of Mueller–Hinton broth per well and further incubation for 48 h at 37 ◦C. The MBC was defined as the lowest concentration of the tested extract/compound/antibiotic killing the majority (99.9%) of bacterial inoculum, thus with no visible growth [50]. Streptomycin (Sigma P 7794, Santa Clara, CA, USA) (0.05–3 mg/mL) was used as a positive control for bacterial growth. Water was used as a negative control.

#### 2.7.3. Antifungal Assay

Evaluation of the antifungal activity was done according to the guidelines of the CLSI [51], using the reference method for broth dilution antifungal susceptibility testing of yeasts, with slight modifications. To determine the minimum inhibitory concentration (MIC) and minimum fungicidal concentration (MFC) of the tested extracts, three fungi were used: *Candida albicans (ATCC 10231), Candida zeylanoides (ATCC 20367*), and *Candida parapsilosis (ATCC 22019)*. All the tested fungal strains were obtained from the above-mentioned source. The cultures were stored on malt agar at 4 ◦C and subcultured monthly. Before antifungal susceptibility testing, each strain was inoculated on malt agar plates to ensure optical growth characteristics and purity. The medium used for susceptibility testing was broth malt. The initial density of *Candida* spp. was approximately 2 <sup>×</sup> 106 colony-forming units/mL (CFU/mL). Inoculums (density of 0.5 in McFarland scale) were prepared in a 0.9% NaCl sterile solution. Then, tested strains were suspended in broth malt medium, to give a final density of 1.5 <sup>×</sup> 105 CFU/mL. For the minimum inhibitory concentration test, the broth microdilution method was applied by preparing a serial of dilutions in 96-well plates. The 100 μL medium was placed into each of the 96 wells of the microplates. Aliquots of 100 μL of each extract diluted in 0.85% saline (concentration of 0.1 g/mL) were added into the first rows of the microplates, and twofold serial dilutions were made by dispensing the solutions into the remaining wells. Then, 10 μL of the inoculum was added to the wells. Plates were incubated at 28 ◦C for 72 h on a rotary shaker. Minimum inhibitory concentration (MIC) values were determined by adding resazurin (20 μL, 0.02%) followed by incubation for 2 h. The MIC was defined as the lowest concentration required to inhibit the growth of the fungal strain (observed through a binocular microscope). The MFCs were determined by serial subcultivation of 2 μL of tested extracts dissolved in medium and inoculated for 72 h into microtiter plates containing 100 μL of broth per well, followed by further incubation 72 h at 28 ◦C. The lowest concentration with no visible growth was defined as the MFC, indicating the death of 99.9% of the original inoculum. The positive control used was fungicide fluconazole (1–3500 μg/mL) (Sigma F 8929, Santa Clara, CA, USA), while the negative control used was water. All the tests were done in duplicate and repeated thrice.

#### *2.8. Mutagenic and Antimutagenic Assay*

According to the plate incorporation method [52], described in more detail by Sarac and Sen [53], the plant extracts were tested for mutagenicity and antimutagenicity towards *S. typhimurium* TA98 and *S. typhimurium* TA100, whereas the positive controls used were 4-nitro-ophenylenediamine (4-NPD, 3 mg/plate) for TA98 and sodium azide (NaN3, 8 mg/plate) for TA100. The negative control was ethanol/water (1:1, *v*/*v*), and the concentration of plant extracts was established to 5 mg/plate. According to the equation described by Ong et al. [54], the antimutagenicity was calculated as follows: %Inhibition = [1 − T/M] × 100, where T is the number of revertants per plate in the presence of mutagen and the plant extract, and M is the number of revertants per plate without plant extract (positive control). The antimutagenicity of the reference mutagens in the absence of the plant extract was defined as 0% inhibition. For each of the two species, the testing was done in duplicate with three subsamples each, and in accordance, the data are reported as the mean ± standard deviation (SD). The following percentage ranges were used to express the antimutagenicity: strong: 40% or more inhibition; moderate: 25–40% inhibition; low/none: 25% or less inhibition [55].

#### *2.9. Statistical Analysis*

All of the analyses were done in triplicate, and the data were reported as the means ± standard deviation (SD). The statistical differences among the leave extracts of the three different locations for each type of species were performed using one-way analysis of variance (ANOVA) (Tukey multiple comparison tests) via GraphPad Prism Version 8.0.1 (Graph Pad Software Inc., San Diego, CA, USA). Differences between means at the 5% level were reported to be statistically significant.

#### **3. Results and Discussion**

#### *3.1. Phenolic Profile of Wild Bilberry and Lingonberry Leaves*

In this study, in the leaves of bilberry and lingonberry, 21 phenolic compounds were identified, originating from four phenolic groups: hydroxycinnamic acids, flavonols, flavanols, and anthocyanins, whereas 19 were found in bilberry leave extracts and 18 in lingonberry leave extracts (Table 1). In the case of bilberry leaves, the most abundant compounds for all three locations were represented by the flavonols class comprising only quercetin derivates. The second most abundant class was flavanols. For lingonberry leaves, the most abundant class of compounds was flavanols, as reported in the literature [56], for all the three different altitude habitats, followed by hydroxycinnamic acids. In the case of lingonberry leaves, the flavonols class registered small levels for each compound, except rutin. Moreover, the anthocyanins group was not detected.


**Table 1.** The phenolic compounds content in the leaves of bilberry and lingonberry using HPLC-DAD-ESI-MS and expressed as mg/g. individual for each type of species (one-way analysis of variance (ANOVA); multiple comparison test; Tukey multiple range test (*<sup>p</sup>* = 0.05); GraphPad Prism Version 8.0.1, Graph Software, Inc., San Diego, CA, USA). VMT, *V. myrtillus* leaves from Turda; VMS, *V. myrtillus* leaves from Smida; VMB, *V. myrtillus* leaves from Borsa; VVIT, *V. vitis-idaea* leaves from Turda;VVIS,*V.vitis-idaea*leavesfromSmida;VVIB,*V.vitis-idaea*leavesfromBorsa;n.d,notdetected.

#### *Antioxidants* **2020**, *9*, 495

The flavanols identified among the two studied species were catechin, epicatechin, gallocatechin, epigallocatechin, two procyanidin dimers, and procyanidin trimer. The procyanidin dimers and trimers are known as proanthocyanidins as well.

In the case of bilberry leaves, for almost all flavanols, the *V. myrtillus* leaves from Smida (VMS) reported the highest amounts. Exceptions were catechin and procyanidin dimers II, in which case *V. myrtillus* leaves from Turda (VMT) presented significantly higher values. The major flavanol identified was procyanidin trimer in all three natural habitats, whereas the VMS had the highest value (24.30 ± 0.72 mg/g plant material), closely followed by VMT, while *V. myrtillus* leaves from Borsa (VMB) registered a 2.5-fold lower value. In particular, gallocatechin was twofold more in VMS than VMT and threefold more when compared with VMB. Epigallocatechin presence only in VMS contributes to the range of differences found among the three different locations. Procyanidin dimer I was not detected in the bilberry leaves of any of the three habitats, whereas epicatechin was the minor compound identified. The VMB leaves had the lowest values among all the flavanols identified, whereas epigallocatechin, epicatechin, and procyanidin dimer I were not present. Compounds present in lower proportions were catechins, in line with the previous results on bilberry leaves from Northern Europe [24]. Significant differences in gallocatechin, epigallocatechin, epicatechin, and procyanidin trimer were detected among locations, and up to threefold higher levels (in the case of gallocatechin) were measured, which can be linked to specific growth conditions of the sites (soil, solar exposure, microclimatic conditions). The habitat can specifically influence the amounts of phenolics as follows: either by the influence of pedological or climatic factors and their interactions [57]. Likewise, Martz et al. [24] reported that high-light-intensity location, higher altitudes, and/or latitudes contributed to more than twofold higher levels of phenolics in the leaves in contrast to lower altitudes or low-light-intensity sites.

Exceptionally high levels of flavanols were quantified in lingonberry leaves, in agreement with the results of previous studies [14,16,56,58], whereas the most recent study of Tian et al. [59] found, as the two most common flavanols, (+)-catechin and (-)- epicatechin, at the highest level in lingonberry (*V. vitis-idaea*) leaf extract (118 mg/100 mL). In our study, gallocatechin was quantified in high amounts, whereas epigallocatechin was detected in significant levels ranging from 23.35 ± 0.61 to 35.97 ± 0.23 depending on the habitat. Their occurrence has never been reported in lingonberry so far, only in bilberry leaves [23] and bilberry stems [19]. However, Bujor et al. [14] reported only a trace amount in lingonberry leaves and quantified the gallocatechin in lingonberry stems. In our study, *V. vitis-idaea* leaves from Smida (VVIS) registered the highest values among almost all the flavanols identified, except for procyanidin trimer and procyanidin dimer II. Gallocatechin was the major flavanol identified in all the three locations, whereas VVIS had the highest amount (46.81 ± 0.38 mg/g plant material), while epigallocatechin and catechin were close behind. Epicatechin was the minor compound identified, and only in *V. vitis-idaea* leaves from Borsa (VVIB); moreover, VVIB registered the highest values for procyanidin dimer II and procyanidin trimer. Furthermore, recent literature reported rich contents of procyanidin dimers and trimers in the extracts of lingonberry leaf (85 mg/100 mL) [59]. All these results underline that rising concentrations of flavanols and especially gallocatechin, epigallocatechin, and catechin in lingonberry leaves, as well as procyanidin trimer in bilberry leaves, were observed in the habitats with higher altitude. This fact was explained in the previously reported results, whereas an increase of flavonoids level with elevation in herbal plants [60] and bilberry leaves [24] was registered. According to the literature, catechin dominated in the red berries, lingonberry, and cranberry, while epicatechin dominated in blue and blackberries [56], in agreement with our findings.

*V. vitis-idaea* L. yielded greater amounts of gallocatechin, epigallocatechin, catechin, and procyanidin dimer I with increasing altitude and its related climatic and soil conditions, except lower levels were found at the altitudes of 1100 when compared with 850 m in the case of procyanidin dimer II and procyanidin trimer (Table 1). The variation of the flavonoid fraction turned out to be closely related to the altitude-derived conditions, because we found the percentages of four out of six flavonoid compounds rising significantly at the highest altitude. It can be concluded that environmental

factors at higher altitudes lead to elevated levels of flavanols, with gallocatechin and epigallocatechin above all, in dried and grounded lingonberry leaves.

Hydroxycinnamic acids are the most widespread phenolic acids in plants, which are described as cinnamic acid-derived compounds. Four derivates of hydroxycinnamic acid were identified: chlorogenic acid, feruloylquinic acid, dicaffeoylquinic acid, and caffeoylarbutin. In the case of bilberry species, VMB together with its environmental-derived conditions (good solar exposure, acid brown soil, low-temperature range) presented the highest levels among the four compounds, except for caffeoylarbutin and dicaffeoylquinic acid, which were not detected. The major compound reported was feruloylquinic acid (59.65 ± 0.44 mg/g plant material in VMB), followed by chlorogenic acid (5.94 ± 0.05 mg/g plant material in VMB) as 10-fold less than feruloylquinic acid. The study of Martz et al. [24] indicated that bilberry leaves from higher latitudes and higher altitudes (boreal forests in Finland, thus low solar exposure) had lower levels of chlorogenic acid derivatives. In the recent paper investigating the Finnish bilberry (*V. myrtillus* L.) leaf extract, hydroxycinnamic acid derivatives represented 82% of the total content of phenolics, mostly as 3-O-caffeoylquinic acid, whereas other hydroxycinnamic acids (coumaric acid, caffeic acid, and ferulic acid) were identified as esters of acids or hexoses [59].

Concerning the lingonberry species, a specific hydroxycinnamic acid was found, precisely the caffeoylarbutin (not detected in bilberry leaves), with an increase of three- to fourfold when compared with dicaffeoylquinic acid, depending on the location. Similarly, Liu et al. [25], Tian et al. [16], and Hokkanen et al. [23] found 2-caffeoylarbutin as the major caffeic acid derivative. The lowest levels of all hydroxycinnamic acids were reported for VVIS, whereas the highest levels were reported for VVIB (caffeoylarbutin). The major compound was again feruloylquinic acid (33.42 ± 0.37 mg/g plant material), as almost half the amount when compared with bilberry species.

Bidel et al. [61] found that the amount of hydroxycinnamic acid highly increased with a higher photosynthetic active radiation (PAR) level, while Li et al. [62] also reported a comparable pattern in apple peel. Hydroxycinnamic acids protect the fundamental tissues from adverse UV radiation; therefore, their expanded accumulation in intense light exposure is anticipated [36]. The high UV-B exposure at higher altitudes is the key determinant for the increased synthesis of phenolic acids in plants [63]. Moreover, lower temperatures at higher altitudes also sustain secondary metabolism [33,37], particularly the accumulation of hydroxycinnamic acids. Following all the above, the Borsa habitat, characterized by a good solar exposure, low-temperature range, and brown acid soils, may explain the highest levels of hydroxycinnamic acids when compared with a moderate/partial solar irradiation and higher temperature ranges (characterizing the other two habitats).

From the flavonols group, seven phenols were identified, all were quercetin derivatives: quercetin, quercetin-rutinoside (rutin), quercetin-glucoside, quercetin-acetyl-rhamnoside, quercetin-arabinoside, quercetin-xyloside, and quercetin-diglucoside. The major flavonol identified within both species was quercetin-rutinoside (rutin), being approximately 40 times higher than most of the flavonols compounds identified, and about 2 times higher than quercetin-acetyl-rhamnoside. Among studied bilberry leaves, VMB presented the highest level (49.83 ± 0.63 mg/g plant material), where all the extracts showed a level above 40 mg/g plant material. Concerning the lingonberry leaves, again, VVIB has shown the highest amount (21.88 ± 0.19 mg/g plant material), where VVIS was half of this level. The second major flavonol compound identified in both species was quercetin-acetyl-rhamnoside with VMT presenting the highest amount (18.60 ± 0.16 mg/g plant material) in the bilberry species, and VVIB in the lingonberry species (8.01 ± 0.01 mg/g plant material). In the case of almost all flavonol compounds, the Smida location (1100 m altitude) reported the lowest values. These results suggest that a Borsa habitat-type (altitude 850 m, good solar exposure, low-temperature range, brown acid soils) may be more beneficial for the biosynthesis of major flavonols compounds, whereas a good solar exposure (low forest environment) may positively contribute to flavonols level. The existing literature [19] on bilberry leaves reported, as major flavonols compound, the quercetin glycosides, namely, quercetin-3-O-galactoside, quercetin-3-O-glucoside, quercetin hexuronides, quercetin pentosides, and a quercetin rhamnoside. Moreover, the specific quercetin-3-O-(400-(3-hydroxy-3-methyl glutaryl))-a-rhamnoside was identified in all of the morphological parts studied, being previously reported in leaves by Hokkanen et al. [23] and Ieri et al. [27]. In the study of Bujor et al. [14], investigating the lingonberry leaves, a range of 12–19% flavonols were found. There were 18 quercetin glycosides identified, whereas the quercetin-3-O-galactoside, quercetin-3-O-glucoside, quercetin rutinoside, quercetin pentosides, and quercetin-3-O-rhamnoside were in line with previous findings [16,23,25] and our study.

Concerning the anthocyanins class, the three anthocyanins identified were cyanidin-glucoside, cyanidin-arabinoside, and cyanidin-acetyl-glucoside. The anthocyanins group was present only in VMT and VMS leave extracts, but in a small amount (<0.35 mg/g). Our findings are in agreement with the study of Jaakola et al. [36], where elevated gene expression and, therefore, flavonoid biosynthesis including cyanidin glycosides owing to an increased UV exposure in bilberry leaves was described. Therefore, an assumption that anthocyanins from both bilberry leaves may have occurred in higher amounts with increased solar radiation can be made, if we consider the presence of cyanidin-arabinoside only in VMS. When compared with their related berries [64], anthocyanin synthesis is highly affected by light exposure and, as a consequence, bilberries from shaded sites [40] contained lower amounts of red pigments, as low light conditions limit photosynthetic activity. The mechanism involves firstly a decrease in carbohydrate synthesis, followed by a low level of substrate generated for secondary metabolism. Certain phenolic classes are then downregulated, with anthocyanin synthesis negatively influenced. Li et al. [62] also found that anthocyanin level as well as the flavonol content and activity of phenylalanine ammonia-lyase (PAL), and certain enzymes, were increased in the sun-exposed apple peel compared with the shaded peel, underlying the upregulation of the phenylpropanoid pathway generated by a favorable light condition. Rieger et al. [60] reported that the anthocyanins level in bilberries decreased with increasing altitude, while Roslon et al. [65] did not found a relationship between the content of anthocyanins in bilberry fruits and the position of habitats at different altitudes. In the same study, the leaves anthocyanins were not investigated as they were considered not specific for leaves.

#### *3.2. Total Phenolics and Total Flavonoids*

According to Figure 1A, the TPC among both species was very similar, whereas there were no statistical differences between the three types of lingonberry leaves. In the case of bilberry leaves, the highest level of TPC was registered by VMT with 13,588.95 ± 9.25 mg GAE/100 g plant material (135.8 ± 9.25 mg GAE/g plant material), being significantly different only from VMS, but not from VMB. In the study of Bujor et al. [19], on the same period of vegetation (September), for the bilberry leaf extracts, a TPC of 142.9 ± 19.2 (mg GAE/g dry extract) was reported, while in the most recent study on lingonberry of the same author [14], the lingonberry leaves extract shown a TPC of 158.9 ± 6.0 (mg GAE/g dry extract). These results are in the same range as our findings. Nevertheless, the *Vaccinium* plants have the same Romanian origin, but different habitats and environmental factors. In the study of Tian et al. [59], the lingonberry leaf ethanolic extracts showed a TPC of 859.5 ± 9.9 (GAE mg/100 mL), while for bilberry leaf ethanolic extracts, a TPC of 201.7 ± 18.2 (GAE mg/100 mL) was found. A possible explanation for the evident differences in contrast to our results may lie in the geographical location, Finnish versus Romanian. According to the same study [59], a higher value of Folin–Ciocalteau was found in the extracts from leaves than in the extract from berries and branches, and the leaf extracts showed higher antioxidative activities (3–20-fold in ORAC assay, 10–20 fold in TRAP) than the berry extracts, in association with the higher contents of phenolic compounds in the leaf extracts [59]. However, regarding our findings, it may be stated that the different habitats did not statistically influence the TPC of both bilberry and lingonberry leaves, considering the high fluctuation in amounts in the different phenolic sub-classes among the three different locations.

**Figure 1.** Total phenolic content (Folin–Ciocalteu method) (**A**), total flavonoids content (**B**), total anthocyanin content (**C**), and 2,2-diphenyl-1-picrylhydrazyl (DPPH) antioxidant activity (**D**) of the two species leave extracts, from all three locations. The total phenolic content of the extract is expressed as gallic acid equivalents (GAE) in mg/100 g plant material. Total flavonoid content is expressed as quercetin equivalents (QE) in mg/100 g plant material. The DPPH activity was expressed as percentage inhibition (I%). Values are reported as mean ± SD of triplicate determinations and different symbols (\*, \*\*, \*\*\*) indicate significant differences (*p* < 0.05) between the three different locations for each of the two species leave extracts, separately (one-way analysis of variance (ANOVA), multiple comparison tests, Tukey multiple range tests), while symbol (ns) indicate no significant difference. VMT, *V. myrtillus* leaves from Turda; VMS, *V. myrtillus* leaves from Smida; VMB, *V. myrtillus* leaves from Borsa; VVIT, *V. vitis-idaea* leaves from Turda; VVIS, *V. vitis-idaea* leaves from Smida; VVIB, *V. vitis-idaea* leaves from Borsa.

Figure 1B presents the TFC of the leaves extracts from both examined species. Among the studied bilberry leaves, there was no statistical difference with the habitat variation, with all three extracts having similar levels of approximately 7300 mg QE/100 plant material. The bilberry species had twofold higher TFC when compared with lingonberry. This may be explained by the fact that chromatographic profiles of flavonoids, but not only them, are different among bilberry, blueberry, lingonberry, and cranberry [66,67]. In the case of lingonberry extracts, the VVIB registered the highest value (4994.18 ± 8.03 mg QE/100 plant material), being statistically different from the other two habitats. Uleberg et al. [37] found that the amount of flavan-3-ols was higher in bilberries growing at lower

temperatures, a fact that might explain why Borsa habitat, with lower temperatures (ranging between 13 ◦C and 8 ◦C), registered higher amounts. In the study of Mikulic-Petkovsek et al. [40], low levels of flavanols were found in fruits collected in shaded forests characterized by a low photosynthetic active radiation (PAR), and high flavanol amounts in bilberries from sun-exposed locations with high PAR. Jaakola et al. [36] reported that the levels of flavan-3-ols were significantly higher in bilberry leaves exposed to direct sunlight. Considering that Borsa collection place had a good solar exposure when compared with partial or moderate exposure in the other two habitats, these previous findings might explain our results.

Other articles [60,63] reported that increased solar exposure by higher altitudes contributes directly to increased flavonoids content in plants, which is partially true in our study considering that Borsa location (850 m altitude) had the highest level, and Smida location (1100 m) did not. Moreover, another study investigating the bilberry leaves concluded that leaves collected from open and forest areas showed that synthesis and accumulation of flavonoids were delayed in the forest compared with the high light open sites [24]. Several flavonoids and hydroxycinnamic acids act as characteristic UV shields and contribute to the plants' protective mechanism determined by high irradiation sites [60].

#### *3.3. Total Anthocyanin Content*

To our knowledge, this is the first study investigating the variation of the amounts of anthocyanins in wild-grown bilberries and lingonberries leaves in correlation to the geographical habitat. According to Figure 1C, only bilberry leaves via VMT and VMS extracts were found to have anthocyanin content. The anthocyanin profile contained small amounts of cyanidin glycosides, whereas VMS registered the highest level, precisely 13.29 ± 0.13 mg/100 g plant material. Our results matched well with the results of a previous comparison of eastern and southern European plants with those from Scandinavia [68], suggesting that a higher altitude may provide an increased sunlight exposure, and thus a higher anthocyanins content. The low levels of total anthocyanin content might be owing to a non-specificity in the leaves of berries when compared with the fruits.

#### *3.4. DPPH Antioxidant Activity*

The percentage inhibition of the DPPH radicals for tested leaf extracts is shown in Figure 1D. For bilberry leaves, VMT showed the best radical scavenging capacity based on the DPPH assay expressed as percentage inhibition of DPPH radicals (also as Trolox equivalents, precisely 310.74 mMT/100 g), while in the case of lingonberry leaves, the VVIB had the highest percentage inhibition of the DPPH radicals (and as Trolox equivalents, precisely 320.83 mMT/100 g). These results could be explained by the highest value registered for TPC, in the case of bilberry leaves, and considering the well-known correlation between increased phenolic content and strong antioxidant capacity, whereas for VVIB, the highest TFC value reported may be responsible for the scavenging capacity, as previously reported in the literature [69]. Previous articles have reported that more solar exposure at increased altitude contributed to elevated biosynthesis of ortho-dihydroxylated flavonoids [34,63], as well as a better radical scavenger capacity [70]. In the case of lingonberry leaves, flavanols, flavonols, and caffeic acid derivatives bring highly antioxidant 1,2-dihydroxyphenyl moieties, whereas coumaric acid derivatives display the less antioxidant monohydroxyphenyl moiety. For similar phenolic contents in the case of lingonberry leaves, the *V. vitis-idaea* leaves from Turda (VVIT) present significantly lower antioxidant capacity than both VVIS (−26%) and VVIB (−39%) (Figure 1D). Concerning the higher levels of flavanols, and precisely of procyanidin dimers and procyanidin trimer, compared with VVIT, as well as for VVIB, a significant contribution via feruloylquinic acid might explain the differences. Extension and terminal epicatechin units in flavanols were already proven to be similarly reactive in the quenching of the nitrogen-centered DPPH radical [71]. In the study of Tian et al. [59], the DPPH radical scavenging capacity of the berry leaf extracts varied among species and cultivars, whereas within 10 min, all the leaf extracts succeeded to capture around 80% of DPPH radicals.

The difference in reactivity of leaf extracts, from the three different habitats, in the DPPH test can be attributed to their varying contents in polyphenols containing dihydroxyphenyl moieties, molecular sizes, like for flavanols [14], or the presence of unidentified antioxidant substances. Soobrattee et al. [72] classified the antioxidant activity in the following order: procyanidin dimer > flavan-3-ols > flavonols > hydroxycinnamic acids > simple phenolic acids. Heim et al. [73] explained why the proanthocyanidins (procyanidin dimers and procyanidin trimer) as oligomers and polymers of flavan-3-ols exhibit stronger DPPH capacity, namely owing to more catechol groups, coupled with C3-OH and C4-C8 linkage.

#### *3.5. Assessment of Antimicrobial Capacity*

The studied leaves extract registered antibacterial and antifungal capacity towards bacteria and fungal strains. The results of MIC are described in Table 2, and those of MBC are provided in Table 3, for both bacteria and fungi strains. An important range of bacteriostatic effects of the bilberry and lingonberry leaves extracts was reported, depending on the tested strain.


**Table 2.** Minimum inhibitory concentration (MIC) of bilberry and lingonberry leaves expressed as mg/mL.

**Table 3.** Minimum bactericidal/fungicidal concentration (MBC/MFC) of bilberry and lingonberry leaves expressed as mg/mL.


In the case of bilberry leaves, towards *S. aureus*, the best antibacterial activity was registered for both VMT and VMB (MIC = 0.06 and MBC = 0.12 mg/mL). This result may be because of the increased TPC, considering that several studies underlined the fact that polyphenols may attack an important number of bacteria, and the antimicrobial capacity depends on interactions between polyphenols and bacterial cell surface [74,75]. *R. equi* was the most sensitive strain towards all the bilberry extracts, whereas *E. faecalis* Gram-positive strain was the most resistant one.

Towards the lingonberry extracts, *S. aureus* and *E. faecalis* exhibited a higher resistance in comparison with *R. equi.* Considering the Gram-negative strains, *E. coli enterotoxigen* was not as sensitive as *K. pneumonia* and *P. aeruginosa* towards extracts' antibacterial effects. In the case of lingonberry species, the same antibacterial pattern was registered as for Gram-positive ones.

The highest inhibitory activity among bilberry species against all the strains is attributed to the VMT and VMB, while for lingonberry, it seems that natural habitat conditions did not influence the antibacterial effect, but the antifungal effect, only in the case of *Candida zeylanoides*. It can be concluded that the Gram-positive strains were much more sensitive to all the tested extracts when compared with the Gram-negative ones.

There are only a few studies reporting the antibacterial capacity of bilberry and lingonberry species. In vitro antimicrobial effect of flavonol glycosides, anthocyanins, procyanidins, and flavan-3-ols groups derived from lingonberry juice were demonstrated towards *S. mutans* and *F. nucleatum* [76]. Moreover, the antibacterial capacity of fruits and leaves of bilberry in different types of solvents, like water, ethanol, and ethyl acetate, was tested on *E. coli*, *E. faecalis*, and *P. vulgaris*, and it was reported that all extracts had a higher effect towards *E. faecalis* and *P. vulgaris* [77]. The study of Tian et al. [59] demonstrated that the extracts of lingonberry leaves, hawthorn leaves, sea buckthorn leaves, Saskatoon leaves, and raspberry leaves registered high inhibitory effects towards *S. aureus*, *L. monocytogenes*, and *B. cereus*. The findings also suggested increased sensitivity of Gram-positive in contrast to Gram-negative bacteria to the phenolic extracts. Moreover, the same study found that the TPC and the content of non-flavonoid phenolics presented a stronger correlation with the inhibitory effects on *S. aureus* and *Bacillus cereus* when compared with TFC. Another study [78] established that lingonberry fruit extracts containing mainly type-A proanthocyanidins may be bactericidal against *S. aureus* or inhibit the hemagglutination of *E. coli.* Considering that both species had important amounts of proanthocyanidins (procyanidin dimer II, procyanidin trimer), this hypothesis could explain the antibacterial effects on *S. aureus* and *E. coli*. Several mechanisms of action in the growth inhibition of bacteria are involved, such as destabilization of the cytoplasmic membrane, permeabilization of the plasma membrane, inhibition of extracellular microbial enzymes, direct actions on microbial metabolism, and deprivation of the substrates required for microbial growth [79].

Regarding the antifungal capacity, both species, in the case of all three types of habitats, had the same effect towards *Candida albicans*, precisely none (MIC = 125 and MFC = 250 mg/mL) when compared with control Fluconazole (MIC = 15.62 and MFC = 31.25 mg/mL), with the highest antifungal effect towards *Candida parapsilosis* (MIC = 31.25 and MFC = 62.5). Against *Candida zeylanoides,* the highest inhibitory potential was registered by VMB and VMT (MIC = 31.25 and MFC = 62.5 mg/mL), as well as VVIS, respectively. A possible explanation of why *Candida* species were not sensitive to our berry leaf extracts could be explained by the lack of ellagitannins, reported previously as the main antimicrobial compounds against these microorganisms [80].

#### *3.6. Assessment of Antimutagenic E*ff*ects of Bilberry and Lingonberry Leaves*

The investigation and discovery of antimutagenic properties of plants are of great practical and therapeutic importance in pharmacology and medicine. Research over the past few years has revealed that mutation has a key role in carcinogenesis [81]. The wild bilberry and lingonberry leaf extracts were tested for their antimutagenic activity, considering our recent review study on their significant antioxidant capacity [1]. The influence of the habitat conditions, as well as the type of *Vaccinium* species, had an important influence on the number of revertants in *S. typhimurium* TA98 and TA100. The antimutagenicity potential of both species towards *S. typhimurium* TA98 and TA100 is reported in Table 4, whereas the tested direct-acting mutagens were 4-NPD for TA98 and sodium azide (NaN3) for TA 100.


**Table 4.** Antimutagenicity capacity towards *Salmonella typhimurium* TA98 and TA100 strains.

<sup>a</sup> Values expressed are means ± S.D of three replications. <sup>b</sup> 4-nitro-ophenylenediamine (4-NPD) and NaN3 were used as positive controls for *Salmonella thyphimurium* TA98 and TA100 strains, respectively.

Concerning *S. typhimurium* TA98, the leaf extracts proved to significantly inhibit the number of revertants of strain TA98 induced by 4-NPD. Therefore, moderate antimutagenic activity was reported for both types of species, whereas the higher inhibition was registered by the bilberry leaves, precisely VMT (31.95%), closely followed by VMS (31.44%). A possible explanation for this could be the fact that the VMT sample had an increased scavenging capacity, as the literature constantly links the antioxidant potential with the antimutagenic capacity of different types of plant extracts [82,83]. Moderate inhibition of around 25% was registered by all the lingonberry types of leaves, a fact that might be explained by the lack of statistically significant levels of TPC among the three extracts.

Towards TA100, both types of species registered a higher antimutagenic activity, suggesting that *S. typhimurium* TA100 was much more sensitive to *Vaccinium*-type of leaf extracts. Following Table 4, all three types of bilberry leaves showed a strong inhibition capacity (>40%), whereas the best inhibition was exhibited by the VMT (43.26%) extract, closely followed by VMS and VMB. Regarding the lingonberry leaves, all of the extracts proved a moderate antimutagenic effect, whereas the VVIT extract inhibited the mutagenic effect of sodium azide of more than 36%, while the VVIS had the lowest inhibition percentage (29.79%). The best antimutagenic capacity was registered toward the strain *S. typhimurium* TA100 by all the leaf extracts.

In the case of both species, all the above-mentioned favorable effects are more likely to be associated with a high content of flavanols and flavonols, which significantly decrease the mutagenic activity of the standard mutagens examined. To the best of our knowledge, this is the first study evaluating the antimutagenic activity of Nord-West Romanian wild bilberry and lingonberry leaves, thus with significant novelty for the present paper. Moreover, the literature lacks in studies investigating this specific features of bilberry and lingonberry fruits or leaves. In the study of Smith et al. [84], the antimutagenic activity of different berry extracts was investigated. Among the tested berries were strawberry, raspberry, and blueberry of different fresh cultivars, and in several kinds of solvents (H2O, ethanol, methanol). The antimutagenic inhibition range was between 23% and 53%. Another study [85] investigating the capacity to prevent mutation induced by two promutagenic dietary quinolines, namely MeIQ and 4-NQO of the *V. floribundum* and *V. myrtillus* berries extracts, reported being inactive at concentrations up to 1000 g/plate. In the recent review study on flavonoids' bioactivity [86], it was reported that plant flavonoids exhibit an important antimutagenic activity. Moreover, cranberry (*V. macrocarpon*), as a significant source of polyphenols, has been reported within vitro antimutagenic properties [2,87].

Therefore, regarding the existing data, we can state that the berries contain anthocyanins and procyanidins, constantly reported for their strong antioxidative activity, leading to both in vitro and in vivo antibacterial and antimutagenic activities [88]. In this study, the hydroethanolic extracts from leaves of *V. myrtillus* L. and *V. vitis-idaea* L. did inhibit the mutations on the Ames *Salmonella* test. Besides, the present study has shown for the first time that hydroethanolic extracts from leaves of wild bilberry and lingonberry are a promising source for its antimutagenic compounds. These results indicate that it may be considered to be a safe and useful agent for the prevention of mutations.

#### **4. Conclusions**

The influence of natural habitats on the level of individual phenolic compounds and biological activities were examined, and considerable variations in phenolic profile and significant differences of bioactivities between bilberry and lingonberry leaves, collected from three different locations, were observed.

This study reports a qualitative analysis of bilberry and lingonberry leaves with structures proposed for 21 phenolic compounds. Quantitative analysis revealed that flavonoids class contribute more than half of the phenolic pool in leaves; whereas for bilberry species, rutin represents 50% of this subclass; and for lingonberry species, the flavanols comprise the majority via gallocatechin, epigallocatechin, catechin, and procyanidin trimer. Of significant novelty was the antimutagenic testing among these species, at different habitats, concluding that bilberry leaves have a stronger antimutagenic capacity, whereas better sun exposure may contribute to an increased flavonols synthesis, leading to better antioxidant and antimutagenic activities. Regarding the antimicrobial effects of the studies species, the Gram-positive bacteria were more susceptible to the activity of the extract, presenting high antibacterial effects, whereas the antifungal capacity was low.

Thus, the hypothesis that plants from higher altitudes contain higher amounts of radical scavenging compounds as a result of their exposure to more severe climatic conditions including enhanced solar radiation cannot be affirmed in general. On the basis of our results, distinct differences between the amounts of phenolic compounds due to habitat-derived conditions (altitude, solar exposure, temperature range, and so on) can be expected at least in the case of the flavanols, flavonols, hydroxycinnamic acids, and anthocyanins investigated in these species.

**Author Contributions:** Conceptualization, B.-E.S, ., and G.C.; methodology, B.-E.S, .; software, L.F.C.; validation, G.C., A.M., and D.C.V.; investigation, B.-E.S, ., F.F., and F.R.; resources, D.C.V.; writing—original draft preparation, B.-E.S, . and L.F.C.; writing—review and editing, G.C., A.M., and D.C.V.; supervision, G.C.; project administration, D.C.V.; funding acquisition, D.C.V. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by MCI-UEFISCDI, grant number 2 PCCDI, Proiect PN-III-P1-1.2- PCCDI-2017-0056, and the publication was funded by 37 PFE.

**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* **(**−**)-Loliolide Isolated from** *Sargassum horneri* **Protects against Fine Dust-Induced Oxidative Stress in Human Keratinocytes**

**Mawalle Kankanamge Hasitha Madhawa Dias 1, Dissanayaka Mudiyanselage Dinesh Madusanka 1, Eui Jeong Han 1, Min Ju Kim 1, You-Jin Jeon 2,3, Hyun-Soo Kim 2,4, Ilekuttige Priyan Shanura Fernando 5,\* and Ginnae Ahn 1,5,\***


Received: 11 April 2020; Accepted: 31 May 2020; Published: 2 June 2020

**Abstract:** The emergence of fine dust (FD) among air pollutants has taken a toll during the past few decades, and it has provided both controversy and a platform for open conversation amongst world powers for finding sustainable solutions and effective treatments for health issues. The present study emphasizes the protective effects of (–)-loliolide (HTT) isolated from *Sargassum horneri* against FD-induced oxidative stress in human HaCaT keratinocytes. The purification of (–)-loliolide was carried out by centrifugal partition chromatography. HTT did not show any cytotoxicity, and it further illustrated the potential to increase cell viability by reducing the reactive oxygen species (ROS) production in FD-stimulated keratinocytes. Furthermore, HTT suppressed FD-stimulated DNA damage and the formation of apoptotic bodies, and it reduced the population of cells in the sub-G1 apoptosis phase. FD-induced apoptosis was advancing through the mitochondria-mediated apoptosis pathway. The cytoprotective effects of the HTT against FD-stimulated oxidative damage is mediated through squaring the nuclear factor E2-related factor 2 (Nrf2)-mediated heme oxygenase-1 (HO-1) pathway, dose-dependently increasing HO-1 and NAD(P)H dehydrogenase (quinone) 1 (NQO1) levels in the cytosol while concomitantly improving the nuclear translocation of Nrf2. Future studies could implement the protective functionality of HTT in producing pharmaceuticals that utilize natural products and benefit the diseased.

**Keywords:** *Sargassum horneri*; (–)-loliolide; fine dust; oxidative stress; HaCaT; apoptosis

#### **1. Introduction**

Recently, air pollution has become one of the most debated environmental issues throughout the globe. A real-time air quality index has been created to alert the public regarding levels of various air quality parameters, namely temperature, humidity, pressure, and pollutants such as nitric oxide (NO2(g)), sulfur dioxide (SO2(g)), and carbon monoxide (CO(g)) with immediate updates. This system is operational in many cities around the globe. Fine dust (FD) is one of the principal contributors to air pollution that is prominently observed in highly industrialized East Asian countries, including China, Japan, and Korea [1,2]. Recent studies have predicted increasing global warming in

the upcoming decades, fueled with various deleterious anthropogenic activities that can drastically catalyze the desertification process, which escalates the overall dust emissions from the desert areas and consequently affects environmental dynamics and ecosystems [3]. Strong winds from the northern and northwestern parts of the arid and semi-arid regions of Mongolia and China carry particulate matter as far as the North American continent that can be identified from the mineral signature of the aerosol composition [3,4]. The overexploited use of coal and petroleum for energy has become a major contributor to the emerging FD pollution with unburnt particles constantly being released into the atmosphere [2]. According to Raloff (2001), FD particles originating from industries may contain unburnt hydrocarbons, soot, CO(g), carcinogens (asbestos, pesticides, and silica), and heavy metals such as cadmium, mercury, led, chromium, arsenic, and copper [5]. The above environmental hazards can cause various detrimental effects on human health, triggering allergic reactions and oxidative stress. Chronic exposure would cause the development of conjunctivitis, asthma, rhinitis, and dermatitis [6,7]. In some instances, intercontinental FD clouds can also harbor pathogenic bacteria and viruses [5]. Though numerous studies have been carried out to understand the health impacts of FD, the available literature regarding the effects of FD-induced oxidative stress in human keratinocytes is insufficient.

The consumption of marine algae in East Asian countries has increased over the last few decades with the identification of a large number of bioactive natural products that benefit human health [8,9]. *Sargassum horneri* is an edible brown alga that is abundant along the coasts of China, South Korea, Japan, and the North American continent [10,11]. Due to its high biomass and nutritional value (as it is packed with vitamins, polysaccharides, amino acids, and dietary fibers), it is regarded as one of the delicacy dishes in Korea. Moreover, for centuries, *S. horneri* has been used as an ingredient in indigenous medicine [10]. Various studies have addressed the beneficial effects of bioactive compounds such as sargachromenol, phenolics, fucoxanthin, phlorotannins, proteoglycan, and sulfated polysaccharides isolated from *S. horneri* [7–10,12].

(–)-Loliolide ((6S,7aR)-6-hydroxy-4,4,7a-trimethyl-5,6,7,7a-tetrahydro-1-benzofuran-2(4H)-one) is a frequently available monoterpenoid lactone. First discovered in English Ryegrass (*Lolium perenne*) in 1964, (–)-loliolide (HTT) has since been found in many plants and animals in both terrestrial and marine ecosystems [13]. Various biological functions of HTT have been reported, including antioxidant, anti-fungal, antibacterial, and anti-cancer activities; in some instances, it has been used as an alternative medicine for depression and diabetes [13,14]. The major emphasis of the current study was to demonstrate that FD has the ability to induce keratinocytes to produce reactive oxygen species (ROS), inevitably resulting in oxidative stress that causes cell damage and apoptosis; another emphasis was the potential protective effects of HTT in developing future pharmaceuticals and cosmeceuticals.

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

#### *2.1. Raw Materials, Chemicals, and Reagents*

Urban aerosols (NIES CRM No. 28) were acquired from the National Institute for Environmental Studies (Tsukuba, Ibaraki, Japan). TrytonX-100, 3-(4-50dimethyl-2yl)-2-5-diphynyltetrasolium bromide (MTT), 2- ,7- -dichlorofluorescin diacetate (DCFH-DA), dimethyl sulfoxide (DMSO) ethidium bromide, and low melting agarose were purchased from Sigma-Aldrich (ST. Louis, MO, USA). Human HaCaT keratinocytes were donated by the American Type Culture Collection (Manassas, VA, USA). Dulbecco's Modified Eagle Medium (DMEM) and antibiotics (streptomycin and penicillin) were purchased from GibcoBRL (Grand Island, NY, USA), and the fetal bovine serum (FBS) was obtained from Welgene (Gyeongsangbuk-do, South Korea). Relevant antibodies were purchased from Cell Signaling Technology Inc. (Beverly, MA, USA) and Santa Cruz Biotechnology Inc. (Dallas, TX, USA). All additional chemicals and reagents used were purchased from commercial sources with the highest quality.

#### *2.2. Sample Preparation and Isolation of HTT*

The collection of *S. horneri* samples was done during the spring season in 2015 along Jeju Island coasts in South Korea. Samples were washed with tap water to remove excess salts and other contaminants, and this process was followed by storage under refrigerated conditions. A detailed extraction and purification method of HTT was described in our previous publication [15]. Briefly, *S. horneri* dry powder was extracted with 80% methanol at 37 ◦C and concentrated using a rotary evaporator. The crude extract was sequentially fractioned into n-hexane, chloroform (CMSH), and ethyl acetate. CMSH was further purified by a high-performance centrifugal partition chromatography system (Sanki Engineering, Kyoto, Japan) using upper and lower phases of an equilibrated solvent system composed of n-hexane/ethyl acetate/methanol/water (5:5:5:5, *v*/*v*). The separation was monitored at 240 nm by an L-4000 UV detector (Hitachi, Japan). The eluant was collected into test tubes using a fraction collector (FC 203B, Gilson, South Korea). The further purification of the active fraction was carried out by a Prep HPLC system (Waters, Milford, MA, USA) using a semi-preparative C18 (YMC-Pack ODS-A, 5 μm, 10 × 250 mm) column. The gradient elution was carried out using acetonitrile: distilled water as; 0–60 min 5:95–100: 0 *v*/*v*; 60–70 min 100:0–100: 0 *v*/*v* at a flow rate of 3 mL min−1, while the UV absorbance was observed at 230 and 254 nm by a Waters 2998 photodiode array detector (Waters, Milford, MA, USA).

#### *2.3. In Vitro Analysis*

#### 2.3.1. Cell Culture and Maintenance

HaCaT (human keratinocyte) cells were cultured using DMEM media containing 10% (*v*/*v*) heat-inactivated FBS and 1% (*v*/*v*) streptomycin and penicillin (100 μg mL<sup>−</sup>1). Cells were maintained inside a controlled environment with 5% CO2 at 37 ◦C. Periodic subculturing was practiced, and cells with exponential growth were used for experiments that achieved a 95% confluency. HTT was initially dissolved in DMSO and diluted using DMEM for cell culture experiments. The final concentration of DMSO in treated samples was kept below 0.1%. DMEM was added to control cell groups to maintain a constant volume.

#### 2.3.2. Cell Viability and ROS Production

The concentration of FD used, exposure time, and the method were based on previous observations and preliminary experiments [7]. Initially, FD was suspended in DMEM, followed by sonication for one minute to separate any clumps. FD suspended solvent was vortexed prior to each treatment. HaCaT cells were seeded at a concentration of 1 <sup>×</sup> 10<sup>5</sup> cells mL−<sup>1</sup> in 96 well plates, with 2 <sup>×</sup> 10<sup>4</sup> cells in each well, which were incubated for 24 h. Different concentrations of HTT were used to treat the cells, and after one hour, cells were stimulated with FD (150 μg mL<sup>−</sup>1) for one-hour. The FD-containing media were carefully removed and washed three times, followed by adding new media. MTT and DCFH-DA assays were conducted to measure the cell viability and intracellular ROS generation, respectively [16].

Simultaneously, the same concentration of cells was seeded in a 24 well plate and 6 cm culture dishes with 5 <sup>×</sup> 104 cells per well and 3 <sup>×</sup> 105 cells per dish for the respective detection of ROS levels by using fluorescence microscopy and flow cytometry under DCFH-DA staining. The side scatter vs. forward scatter (SSC vs. FSC) gating strategy was used in the flow cytometry to eliminate cell debris. Images of the fluorescence microscopy were obtained by using an Invitrogen™ EVOS™ Auto 2 fluorescence microscope (Thermo Fisher Scientific, Bothell, WA, USA), and flow cytometry analysis was conducted using a Beckman Coulter CytoFLEX system (Beckman Coulter, Brea, CA, USA).

#### 2.3.3. Nuclear Morphological Analysis

Nuclear staining was carried out by using Hochest 33342 following the method described by Yang et al. (2011) with slight modifications [14]. HTT treatment and FD stimulation followed the same

method mentioned in Section 2.3.2. After washing with phosphate-buffered saline (PBS), the Hoechst 33342 reagent was added, achieving a final concentration of 2 μg mL−<sup>1</sup> per well, each of which were incubated for 20 min at 37 ◦C in the dark. The visualization of nuclear morphology was carried out using a fluorescence microscope.

#### 2.3.4. Annexin V Assay

Early apoptosis detection was carried out by flow cytometry using an eBioscience™ Annexin V Apoptosis Detection Kit (Thermo Fisher Scientific, Carlsbad, CA, USA) according to the manufacture's guidelines. HTT-treated, FD-induced cells were incubated for six hours prior to harvesting for the assay.

#### 2.3.5. Cell Cycle Analysis

Cell cycle analysis was conducted according to a pre-described method by Fernando et al. (2017) with slight alterations [17]. Briefly, HTT-treated and FD-induced cells, as described in Section 2.3.2, were used. Cell harvesting was carried out 24 h after the incubation, which was followed by fixing in 70% ethanol at 4 ◦C. Cells were washed using PBS, treated with 2 mM of ethylenediaminetetraacetic acid (EDTA), and resuspended in 300 μL of PBS–EDTA-containing RNase (0.2 μg mL<sup>−</sup>1) and propidium iodide (50 μg mL<sup>−</sup>1) for 30 min. The analysis was conducted using a flow cytometer.

#### 2.3.6. Alkaline Comet Assay

The alkaline comet assay was carried out following the procedure by Fernando et al. (2017) with minor modifications [17]. The sample treated cells were incubated for 24 h and dispersed in 1% low melting agarose at 40 ◦C. The cell-agarose mixture was pipetted onto agarose pre-coated microscopic slides and rested for solidification. Then, the slides were lysed overnight by submerging in a 1% TrytonX-100-containing lysis buffer at 4 ◦C. Electrophoresis was conducted for 30 min at 30 V/300 mA, while the temperature was maintained at 4 ◦C. A neutralizing buffer was used to wash the slides three times, prior to staining with ethidium bromide (2 μg mL<sup>−</sup>1) for five minutes. Pre-chilled water was used to wash away any excess dye, and the visualization was conducted using a fluorescence microscope.

#### 2.3.7. Western Blot Analysis

As elucidated in Section 2.3.2, HTT-treated FD-induced cells were harvested and lysed using a NE-PER® nuclear and cytoplasmic extraction kit and radioimmunoprecipitation assay (RIPA) buffer (Thermo Scientific, Rockford, IL, USA). Isolated cytoplasmic and nuclear protein levels were determined by using Bio-Rad Protein Assay Dye (Bio-Rad Laboratories, Inc., Hercules, CA, USA), and a concentration gradient of bovine serum albumin was used as the reference standard. A normalized amount of protein was loaded into 10% polyacrylamide gels and electrophoresed. After transferring the protein bands to nitrocellulose membranes, blocking was carried out with 5% skim milk in tris buffered saline (TBS) containing Tween-20 for two hours. Primary antibodies were added to the membranes and incubated overnight at 4 ◦C with continuous agitation. Then, the membranes were incubated with their respective enzyme-linked secondary antibodies for two hours at room temperature and visualized using SuperSignal™ West Femto Maximum Sensitivity Substrate (Thermo Scientific Inc., Rockford, IL, USA) on a Core Bio DavinchChemi imager (Seoul, Korea).

#### *2.4. Statistical Analysis*

PASW Statistics 18 was used to conduct the statistical analysis, and data are expressed as mean ± standard error of the mean (SEM). The significant differences amongst data sets were determined by using the ANOVA with the Duncan's multiple range test. The significance of data was determined at *p* < 0.05.

#### **3. Results**

#### *3.1. E*ff*ect of HTT on Cytotoxicity, Cell Viability and Intracellular ROS Production of FD-Induced HaCaT Cells*

The extraction of *S. horneri* and further purification that led to the isolation of HTT is depicted in Figure 1A. Details of this isolation procedure were described in our previous publication [15]. According to Figure 1B, a non-significant increment of cell viability and ROS was observed, suggesting that the used concentrations of HTT had no cytotoxicity towards HaCaT cells. FD had induced the production of intracellular ROS compared to the control cell group while simultaneously reducing cell viability. However, treating cells with different concentrations of HTT significantly increased the cell viability compared to the cells that were exposed only to FD, concomitantly minimizing the production of intracellular ROS in a dose-dependent manner (Figure 1C). The highest concentration of HTT (200 μM) showed a similar cell viability activity to those of the positive control, indomethacin (IM). The protective effects of HTT against FD-induced ROS generation were confirmed by flow cytometry and fluorescence microscopic analysis (Figure 1D,E). FD-stimulated cells indicated a rightward shift of the cell population with a higher intensity of fluorescence compared to the control cells. The green fluorescence in Figure 1E was at its peak for the FD-stimulated cell group compared to the control. In both instances, treating with HTT dose-dependently reduced the fluorescence of DCFH-DA, confirming the protective effects of HTT against FD-induced oxidative stress.

#### *3.2. HTT Inhibited Early Apoptosis and Apoptotic Body Formation in FD-Induced HaCaT Cells*

Phosphatidylserine (PT) is an inner plasma membrane phospholipid that becomes translocated to the outer plasma membrane during the early stages of apoptosis. Annexin V is a phospholipid-binding protein that has a high affinity towards PT, and that can be utilized to detect early apoptotic cells [18]. The Hoechst 33342 nuclear staining is a unique fluorescent staining technique used to visualize apoptotic bodies and thereby determine the level of nuclear condensation and DNA damage [17]. According to the results of Figure 2A, HaCaT cells stimulated by FD indicated an apparent reduction of total cell density, with a notable increment of apoptotic bodies compared to the control; this indicated nuclear condensation. Cells that were pre-treated with HTT (200 μM) had the least amount of apoptotic bodies and the highest cell population compared to the rest of the treatment groups. Compared to the control, FD-induction increased the cell population in early apoptosis from 0.63% to 16.47%, as depicted in the lower right quadrants of the histograms in Figure 2B. The dose-dependent reduction of early apoptotic cell population further confirmed the protective effects of HTT against FD-induced apoptosis.

#### *3.3. HTT Attenuates DNA Damage and Apoptotic Cells in the Sub-G1 Phase*

The alkaline comet assay—also known as single-cell gel electrophoresis—is a widely used method to detect DNA damage in cells. The comet tail length and tail DNA content are considered to be proportional to DNA damage [17]. As shown in Figure 3A, the tail DNA content drastically increased in the FD-treated cells compared to the control cell group, while the HTT pre-treatment reduced the tail DNA content in a dose-dependent manner suggesting the potential ability of HTT to reduce DNA damage induced by FD. Furthermore, as illustrated in Figure 3B, a significant proportion of FD-induced untreated cells were in the sub-G1 phase (21.46%) compared to the control cell group (1.06%). However, the sub-G1 cell population was dose-dependently reduced with HTT treatment. These results coincided with the results obtained for the Hoechst 33342 nuclear staining, further establishing the protective effect of HTT against FD-induced DNA damage in HaCaT cells.

**Figure 1.** (**A**) Purification process and the chemical structure of (–)-loliolide (HTT) with 1H NMR chemical shifts. The effects of HTT on cell viability and intracellular reactive oxygen species (ROS) production (**B**) in the absence of fine dust (FD) and (**C**) with FD-stimulated HaCaT cells. Analysis of ROS levels in 2- ,7- -dichlorofluorescin diacetate (DCFH-DA)-stained HaCaT cells by (**D**) flow cytometry and (**E**) fluorescence microscopy. Seeded cells were pre-treated with different concentrations of HTT (50–200 μM) at one hour prior to stimulation with 150 μg mL−<sup>1</sup> of FD. Indomethacin (IM) was used as the positive control. The experiments were conducted in three independent determinations (*n* = 3), and the values are given as means ± SEM. Error bars with different letters are significantly different (*p* < 0.05).

**Figure 2.** The effect of (–)-loliolide (HTT) on apoptotic body formation and early apoptosis in fine dust (FD)-induced HaCaT cells. Fluorescence microscopy of nuclear morphological analysis using (**A**) Hoechst 33342 dye and early apoptosis detection by (**B**) annexin V with flow cytometry. HTT pre-treated (50–200 μM) exposed to FD for one hour before replenishing with new media. The experiments were conducted in triplicates with three independent trials to confirm repeatability.

**Figure 3.** (–)-loliolide (HTT) inhibited DNA breakage and apoptosis in fine dust (FD)-stimulated HaCaT cells. (**A**) Comet tails and (**B**) sub-G1 cell populations of FD-stimulated HaCaT cells treated with HTT. Cells were treated with different concentrations of HTT (50–200 μM) prior to stimulation with FD. Comet tail DNA percentages were analyzed using the OpenComet plugin of the ImageJ software. Results were analyzed as three independent determinations (*n* = 3) to confirm repeatability. Values are given as means ± SEM, and error bars representing different letters are significantly different (*p* < 0.05).

#### *3.4. HTT Attenuates Apoptosis via the Mitochondrial Pathway*

The effects of HTT on apoptosis-related proteins were analyzed usingWestern blotting.According to the results shown in Figure 4A, stimulating cells with FD increased the levels of pro-apoptotic proteins p53, B-cell lymphoma 2-associated X protein (Bax), cleaved caspase-9, cytochrome c, cleaved poly (ADP-ribose) polymerase (PARP), and caspase-3 while reducing the anti-apoptotic proteins B-cell lymphoma 2 (Bcl-2), PARP, and B-cell lymphoma extra-large (Bcl-xL) compared to the control cell group. Nonetheless, the HTT pre-treated cells showed increased levels of Bcl-2, Bcl-xL, and PARP while simultaneously suppressing the levels of Bax, cleaved PARP, cleaved caspase-9, caspase-3, cytochrome c, and p53 in a concentration-dependent manner that suggested the protective effect of HTT against FD-induced apoptosis in HaCaT cells. Furthermore, according to Figure 4B, treating cells with HTT (without FD stimulation) had no effect on apoptotic related protein levels (Bax, Bcl-2, and Bcl-xL) in HaCaT cells.

**Figure 4.** Effect of (–)-loliolide (HTT) on the levels of key apoptosis and antioxidant molecular mediators. Effect of HTT on key apoptotic mediators in (**A**) fine dust (FD)-induced and (**B**) non-induced cells. Effect of HTT on nuclear factor E2-related factor 2 (Nrf2)/heme oxygenase-1 (HO-1) signaling molecules in (**C**) FD-induced and (**D**) non-induced HaCaT cells. Seeded cells were treated with different concentrations of HTT one hour before stimulation. Indomethacin (IM) was used as the positive control. The experiment was done in three independent trials to confirm repeatability.

#### *3.5. E*ff*ect of HTT on Upregulation of Nrf2*/*HO-1 Pathway Proteins*

Figure 4C illustrates nuclear factor E2-related factor 2 (Nrf2) pathway-related protein level expressions. FD slightly increased the nuclear translocation of Nrf2. This was consistent with the increment of NAD(P)H dehydrogenase (quinone) 1 (NQO1) and heme oxygenase 1 (HO-1) protein levels in the cytosol compared to the control cells. However, the HTT pre-treated cells had gradually increased levels of NQO1, HO-1, and Nrf2 dose-dependently. Moreover, the levels of NQO1, HO-1, and Nrf2 had a slight dose-dependent increment without FD-stimulation, thus suggesting the potential of HTT to activate the Nrf2/HO-1 antioxidative pathway (Figure 4D). Indomethacin was used as a positive control in the experiment for the comparison of its effects with HTT.

#### **4. Discussion**

"Oxidative stress" and "ROS" are two frequently used terms that complement each other and play major roles in human health [19,20]. Weakened antioxidant defense mechanisms, coupled with excess production of ROS such as hydrogen peroxide (H2O2), superoxide anion radicals (O2 •−), hydroxyl radicals (•OH), hypochlorite (ClO−), and nitric oxide radicals (NOx •), is considered as primary causes of oxidative stress that damages major macromolecules of cells such as carbohydrates, proteins, lipids, and DNA, ultimately leading to cellular death. Moreover, the above incidents may trigger diseases such as diabetes, atherosclerosis, various neurodegenerative and cardiovascular disorders, and even cancer [19,21].

FD is one of the highly discussed issues in the modern world, as it affects countries all around the globe. Prolonged exposure to FD-polluted air containing various toxic substances, namely heavy metals and organic pollutants such as polycyclic aromatic hydrocarbons (PAHs) can synthesize ROS (e.g., the generation of HO• via Fenton's reaction) inside the human body, subsequently leading to pulmonary and systemic oxidative stress [22–24]. According to Mori et al. (2008), the FD used in the current study mainly consisted of Si (14.9%), C (12%), Ca (6.69%), Al (5.04%), S (3.91%), Fe (2.92%), Mg (1.40%), K (1.37%), Cl (0.807%), Na (0.796%), N (0.79%), Ti (0.292%), P (0.145%), and Zn (0.114%), as well as minor amounts of other elements and substances including PAHs such as fluoranthene, benzo (b) fluoranthene, pyrene, indeno (1,3,3,-cd) pyrene, benzo (ghi) perylene, benz (a) anthracene, benzo (k) fluoranthene, and benzo (a) pyrene—all of which have the potential to induce oxidative stress in human keratinocytes [25]. Furthermore, FD reported from other heavily industrialized areas has shown similar general constituents (elements and PAHs) [26,27].

HTT is a crucial bio-active compound available in *S. horneri* that possesses potent medicinal properties and has been used as folk medicine in countries such as Japan, Mexico, Egypt, and the Philippines [13]. Therefore, the potential therapeutic use of HTT in attenuating FD-induced oxidative stress was investigated in the current study.

The HaCaT cell model is commonly used in dermatological studies to represent typical human keratinocytes. Located on the epithelium of the outer skin, these cells are prime candidates that have direct contact with FD and the potential to induce oxidative damage. This was further confirmed from the results shown in Figure 1C–E, where the concentration of the used FD increased the generation of ROS in HaCaT cells, thus increasing oxidative damage and inevitably reducing cell viability. However, the used concentrations of HTT increased the cell viability while minimizing the overall ROS production without any cytotoxicity, thus further illustrating the potential of HTT in attenuating the oxidative stress generated from exposure to FD. Further evidence was acquired from flow cytometry and fluorescence microscopy using DCFH-DA staining, which indicated the reduction of ROS levels upon HTT treatment. The flowcytometric analysis allows to omit the florescence incidents from dead cells and cell debris. The above results, together with fluorescence microscopy, confirmed the ability of HTT to attenuate oxidative stress caused by FD.

Apoptosis is considered to be the major outcome in cells and tissues that are regularly exposed to high levels of oxidative damage [19,21]. DNA fragmentation and nuclear condensation, along with the formation of apoptotic bodies, is a crucial identification criterion of controlled cell death [28]. Hoechst 33342 staining is a widely used staining technique that is implemented to visualize the formation of apoptotic bodies. The annexin V assay has the capability to identify different stages of apoptosis in particular cell lines and to provide the foundation to detect and differentiate between viable, early, and late apoptotic cells [18]. Based on the obtained results, HTT minimized the apoptotic body formation, an effect that was further confirmed by the results of the annexin V and alkaline comet assays. The cell cycle analysis histograms gave clear evidence of the ability of HTT on attenuating apoptosis from the reduction of sub-G1 hypodiploid subpopulations in the cell cycle [29].

The mitochondrial-mediated apoptosis pathway is a highly complex cascade of reactions that regulates cell death. Bcl-2 family proteins, including pro-apoptotic proteins such as BH3 interacting-domain death agonist (Bid), Bcl-2-associated death promoter (Bad), Bcl-2 related ovarian killer (Bok), Bcl-2 homologous antagonist killer (Bak), Bcl-2-interacting killer (Bik), Bcl-2-modifying factor (Bmf), Bim, Nova, p53 upregulated modulator of apoptosis (Puma) and Bax; anti-apoptotic proteins (Bcl-xL, Bcl-2, and Mcl-1) coupled with caspases regulate the complex procedure [17,21]. The activation of the mitochondrial-mediated apoptosis pathway takes place with the activation of p53 and pro-apoptotic Bcl-2 family proteins in the cytosol, which leads to the inhibition of anti-apoptotic Bcl-2 family proteins localized on the mitochondria's outer membrane. This phenomenon increases the mitochondrial outer membrane permeabilization, which leads to the release of apoptosis-promoting proteins cytochrome c, endonuclease G, and the apoptosis-inducing factor to the cytosol. The combination of procaspase-9, apoptosis activating factor-1, and cytochrome c are known as the apoptosome. This complex is the prime candidate for activating caspase-9, which enables a cascade of reactions to take place activating caspases-3, -6, and -7 that further continue the apoptosis process [21]. Moreover, initiator caspases, namely caspase-2, -8, -9, and -10 are responsible for the apoptosis initiation and execution, while the effector caspases such as caspase-3, -6, and -7 are key intermediates that connect the cascade. This leads to the rise of morphological and biochemical modifications in key regulatory molecules such as PARP, which is a protein responsible for the repairing ability, transcription, and stability of DNA [17]. Results observed in the present study showed that HaCaT cells exposed to FD demonstrated apoptosis via the mitochondrial-mediated apoptosis pathway, and HTT dose-dependently minimized the pro-apoptotic molecule levels in cells. Furthermore, an in-depth analysis of mRNA expressions of key mediators related to the mitochondrial-mediated apoptosis pathway could be implemented in future studies and could be utilized to provide a thorough insight into the mechanism.

The activation of the Nrf2 pathway is considered to be one of the key regulatory processes that enable cells to exhibit resistance against redox imbalances and oxidative stress [30]. Inactive Nrf2 is localized in the cytosol bound with Kelch-like ECH-associated protein 1 (Keap1) that inhibits the activation of Nrf2 [31]. With the ongoing oxidative stress conditions in the cell, the Keap1 protein is inactivated by an upstream molecule known as p62 and promotes the nuclear translocation of activated Nrf2. This phenomenon is the key inducer of activating NQO1—a drug-metabolizing enzyme—and HO-1, which triggers protective effects by regulating redox imbalance. Nevertheless, the oxidative stress caused by the stimulant overwhelms the protective effects of antioxidative gene HO-1, limiting its potential ability to maintain redox imbalance [31]. However, based on the results, HTT potentially increased the Nrf2 levels in the nucleus, simultaneously increasing the HO-1 and NQO1 proteins in a concentration-dependent manner in the cytosol, suggesting that the cytoprotective effects of HTT against FD-induced oxidative damage was carried via the Nrf2/HO-1 signaling pathway.

#### **5. Conclusions**

Considering the results of the current study, HTT holds potent protective competence in attenuating FD-induced oxidative stress in human keratinocytes. Furthermore, new techniques and methods could be materialized to identify and increase the efficiency of extracting HTT like bio-active compounds from *S. horneri*. Future studies could be implemented on applications of HTT-based pharmaceutical and cosmetic products as a means of maintaining healthy skin.

**Author Contributions:** Conceptualization, M.K.H.M.D., I.P.S.F., and G.A.; methodology, H.-S.K., M.K.H.M.D., and D.M.D.M.; software, M.K.H.M.D. and D.M.D.M.; validation, E.J.H. and M.J.K.; formal analysis, M.K.H.M.D. and D.M.D.M.; investigation, M.K.H.M.D. and D.M.D.M.; resources, Y.-J.J. and G.A.; data curation, E.J.H. and M.J.K.; writing—original draft preparation, M.K.H.M.D.; writing—review and editing, I.P.S.F. and G.A.; visualization, M.K.H.M.D.; supervision, I.P.S.F. and G.A.; project administration, Y.-J.J. and G.A.; funding acquisition, G.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2017R1D1A1B04035921).

**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*
