**Contents**


#### **Yanan Xu and Patricia J. Harvey**


#### **Jose´ Joaqu´ın Merino, Jose´ Mar´ıa Parmigiani-Izquierdo, Adolfo Toledano Gasca and Mar´ıa Eugen´ıa Cabana-Mu ˜ noz ˜**

The Long-Term Algae Extract (*Chlorella and Fucus sp*) and Aminosulphurate Supplementation Modulate SOD-1 Activity and Decrease Heavy Metals (Hg++ , Sn) Levels in Patients with Long-Term Dental Titanium Implants and Amalgam Fillings Restorations Reprinted from: *Antioxidants* **2019**, *8*, 101, doi:10.3390/antiox8040101 ................ **163**

## **About the Editors**

**Christophe Brunet** senior researcher, Stazione Zoologica Anton Dohrn, Naples, Italy) obtained his Ph.D. in 1994 in biological oceanography at the University of Paris VI (Pierre et Marie Curie). He became researcher at the Stazione Zoologica di Napoli (SZN, Italy) in 2000. Since 2017, he is senior scientist at the SZN. His research interests deal with and microalgal ecophysiology and biotechnology. C. Brunet is interested in algal growth, production of bioactive compounds from microalgae and technological development for carrying out ecological and biotechnological experiments. He is involved in projects in which he is carrying out fundamental and applied research as well as in technological development for improving the role and importance of microalgae in biotechnological applications. He is author of more than 70 scientific publications and three European or Italian patents. He is involved in the Editorial Board of Antioxidants, Scientific reports and Advances in Oceanography and Limnology journals.

**Clementina Sansone** is a Researcher since 2012 at Stazione Zoologica Anton Dohrn. Her research activity is focused on Marine Biotechnology for human and ocean health. Her interest lies in drug discovery from marine micro and macro-organisms, for applications as pharmaceuticals, nutraceuticals and cosmetics. She was involved in testing the bioactivity of chemical compounds from marine microalgae and invertebrates on several human cell lines.

### *Editorial* **Marine Algal Antioxidants**

#### **Clementina Sansone \* and Christophe Brunet**

Stazione Zoologica Anton Dohrn, Villa comunale, 80121 Napoli, Italy; christophe.brunet@szn.it **\*** Correspondence: clementina.sansone@szn.it

Received: 19 February 2020; Accepted: 28 February 2020; Published: 2 March 2020

Sea and marine biodiversity exploration represents a new frontier for the discovery of new natural products with human health benefits ("the exploitable biology", [1]).

New compounds suitable for nutraceuticals, cosmeceuticals, or pharmaceuticals require (i) eco-friendly production and (ii) bioactivity against illness, thus making microorganisms potential interesting targets. Among them, microalgae provide advantages as they are photosynthetic, high growth rate organisms that are easy to cultivate and require less space than higher plants, together with displaying high chemodiversity—though this has barely been explored—coupled with high biodiversity [2]. Although very attractive compared to higher plants, microalgae biotechnology still requires further research and development to lower its cost and enhance practical and industrial interest [3,4]. One of the main branches of the biotechnological exploration of marine algae concerns bioactive and, especially, antioxidant compounds.

This Special Issue, concerning marine algal antioxidants, contains eleven contributions detailing recent advances in this field; experimental results and technical improvements are presented and discussed.

Antioxidant bioactivity concerns different families of compounds, but this issue is focused on the microalgae richness of such compounds (e.g., [5,6]). Among the huge variety of antioxidant compounds, algae derived carotenoids are the most well known, together with other bioactive compounds, such as polyphenols, sterols, carbohydrates, and vitamins [6]. The synergistic effect of all of these families in unicellular organisms induces the high antioxidant power of microalgae that is comparable, or even higher than, the antioxidant activity of higher plants or fruits [5]. The potentiality of a single microalgae cell compared with that of a multicellular plant presents a biotechnological challenge for developing microalgae as an efficient and ecosustainable "bio factory" of bioactive molecules with antioxidant activity. For this reason, it is very important to invest in research programs aiming to investigate the diversity of bioactive molecules along the microalgal biodiversity scale and its intracellular modulation [2].

In a recent study [7], the coastal diatom *Skeletonema marinoi* was used to investigate the modulation of lipophilic antioxidant compounds and the hydrophilic vitamin c by light manipulation. The results revealed a significant effect of light (intensity and/or distribution) on the production of antioxidants as well as a strong link between carotenoid operating photoprotection and the antioxidant molecules and activity modulation. This study confirms the role of light manipulation as a powerful tool for modulating the synthesis of antioxidant compounds in microalgae.

The most frequently investigated algae compounds are carotenoids due to their well-known bioactivity and human wellness benefits as well as their plasticity which allows them to be enhanced through abiotic factors, for example, light modulation in microalgae [2,8].

*Dunaliella salina*, a chlorophyte that is mostly used for biotechnological investigations and applications, mainly relies on the production of β-carotene [9], has been used as a model to study the modulation of carotenoids and β-carotene concentration with respect to the light spectrum [10,11]. This study demonstrates that monochromatic red light strongly affects the carotenoid pool, enhancing the β-carotene concentration as well as modifying the ratio between the different forms of β-carotene

towards 9-cis β-carotene. These studies confirm the relevant role of light in shaping the carotenoid profile in microalgae, demonstrating that its modulation is of great interest for the biotechnological production of such bioactive compounds.

The enhancement of carotenoid production in algae can use genetic engineering and biomanipulation. In order to reach this goal, it is necessary to increase the knowledge about the biosynthetic pathways of these compounds as well as the modulation factors affecting the gene expression involved. Two brown algae, *Saccharina japonica* and *Cladosiphon okamuranus*, have been investigated thanks to the analysis of Genome–Scale Metabolic Networks (GSMNs, [12]). The authors were able to reconstruct the biosynthetic pathways of the main carotenoids in these two algae, highlighting the interest and scientific richness of such approach for the study of targeted biochemical pathways.

Together with carotenoids, chlorophylls and their derivatives are also of interest for biotechnological applications [13]. Enhancing the production of chlorophylls per biomass unit in microalgae and understanding the biosynthetic and degradation pathways of such molecules is therefore biotechnologically relevant. The study by Maroneze and collaborators [13] reported and discussed the modulation of the chlorophyll and carotenoid contents in the model species *Scenedesmus obliquus* with respect to the growth phase and the presence/absence of light, turning growth from autotrophy to heterotrophy. The authors demonstrated that the content and chemical forms of these compounds are affected by growth conditions, laying the foundation for up-scaling and massive production for industrial application.

On the other hand, i.e., in brown algae, fucoxanthin is now being investigated for its potential activity related to human health protection [14]. This pigment might be extracted from numerous classes of microalgae, including diatoms as well as brown macroalgae [14]. The anti-inflammatory, antioxidant, and antiproliferative effects of fucoxanthin were investigated on blood mononuclear cells and different cell lines [15]. The results clearly displayed the antiproliferative and antioxidant activities of fucoxanthin in vitro, highlighting the great interest in its potential use in nutraceuticals.

Pigments are not the only compounds presenting antioxidant properties; other lipophilic compounds such as phenols and hydrophilic compounds accompany them.

It is therefore of interest to investigate the best solvents for obtaining the best yield from the extraction of bioactive compounds from algal biomass. For this reason, in three brown algae [16], the antioxidant properties and antioxidant compound concentration were compared between seven extraction solvents or mixtures between them. This work defined the best extraction procedure for enhancing the harvesting of phenols, flavonoids, carotenoids, and chlorophylls. In the same framework, a technical approach comparing methodologies for the quantification of polyphenols was undertaken on the macroalga *Ulva intestinalis* [17], highlighting some uncertainties and difficulties in actual methodologies that require further optimization of the extraction, identification, and quantification of polyphenols.

In addition, polysaccharides are also relevant antioxidant compounds, and algae might be a relevant source for their production and use as nutraceuticals [18]. In the contribution by Le et al. [19], the authors compared data from different extraction procedures of the green alga *Ulva pertusa* in terms of antioxidant activity together with polysaccharide and ulvan contents. The differences between the various extracts were compared with regard to operational parameters such as power, time, water-to-raw-material ratio, and pH, in order to optimize the quantity yield of ulvan.

Last, but not least, extracts from *Fucus* spiralis and *Chlorella vulgaris* were tested as enhancers of the removal of heavy metals (Hg++, Ag, Sn, Pb) in patients with long-term dental titanium implants and amalgam filling restoration [20]. The authors demonstrated that long-term effects from nutritional supplementation with these algae result in the enhancement of heavy metal removal.

All of these contributions highlight the great potential of marine algae to provide substances/extracts that are able to protect or increase human wellness, and the need for optimization and technological/technical/scientific improvement to increase the biomass-harvesting efficacy with reduction of the production cost. Marine biotechnology relies on the exploration, discovery, and exploitation of marine algal species and/or products still requires research dealing with biodiversity (searching for new targeted species with peculiar biochemical profiles, for instance), chemodiversity (richness and diversity of bioactive molecule screening), bioactivity (antioxidant ability of the algal extracts), technological cultivation improvement (lowering the costs, co-cultivation, environmental modulation), and optimization of the extraction techniques. These steps are crucial to achieve the challenges offered by the green (blue in case of marine) biotechnological revolution which, in our point of view, cannot exist without deployment of the industrial use of (micro)algae.

**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* **Quantification of Polyphenols in Seaweeds: A Case Study of** *Ulva intestinalis*

**Marie Emilie Wekre 1,2, Karoline Kåsin 1,3, Jarl Underhaug 1, Bjarte Holmelid <sup>1</sup> and Monica Jordheim 1,\***


Received: 14 October 2019; Accepted: 30 November 2019; Published: 3 December 2019

**Abstract:** In this case study, we explored quantitative 1H NMR (qNMR), HPLC-DAD, and the Folin-Ciocalteu assay (TPC) as methods of quantifying the total phenolic content of a green macroalga, *Ulva intestinalis*, after optimized accelerated solvent extraction. Tentative qualitative data was also acquired after multiple steps of purification. The observed polyphenolic profile was complex with low individual concentrations. The qNMR method yielded 5.5% (DW) polyphenols in the crude extract, whereas HPLC-DAD and TPC assay yielded 1.1% (DW) and 0.4% (DW) respectively, using gallic acid as the reference in all methods. Based on the LC-MS observations of extracts and fractions, an average molar mass of 330 g/mol and an average of 4 aromatic hydrogens in each spin system was chosen for optimized qNMR calculations. Compared to the parallel numbers using gallic acid as the standard (170 g/mol, 2 aromatic H), the optimized parameters resulted in a similar qNMR result (5.3%, DW). The different results for the different methods highlight the difficulties with total polyphenolic quantification. All of the methods contain assumptions and uncertainties, and for complex samples with lower concentrations, this will be of special importance. Thus, further optimization of the extraction, identification, and quantification of polyphenols in marine algae must be researched.

**Keywords:** seaweeds; green algae; marine algae; *Ulva intestinalis*; *Enteromorpha intestinalis*; quantification; polyphenols; flavonoids; apigenin; accelerated solvent extraction; ASE; HPLC-LRMS; HPLC-HRMS; HPLC; TPC; Folin–Ciocalteu; TFC; qNMR

#### **1. Introduction**

Marine macroalgae, or seaweed, is a large group of macroscopic organisms that are an important component in aquatic ecosystems. The wide diversity of marine organisms is being recognized as a rich source of functional materials and, in 2015, the global seaweed aquaculture production reached 30 million tons [1]. Although marine algae have gained increasing attention over the last years due to the fact of their bioactive natural substances with potential health benefits, they are still identified as an underexploited resource [2–6].

Natural antioxidants with multifunctional potential are of high interest, and numerous studies have focused on natural antioxidants, including polyphenols and flavonoids, from terrestrial plants [7–9]. However, the application potential of polyphenolic analyses of marine sources suffers from several factors, most importantly, the lack of exactness with respect to quantitative and qualitative data at a molecular level. Marine plant material with analytic matrices at very low concentrations and a high and variable dissolved salt concentration makes polyphenol analyses challenging [4,10]. The diversity

of phenolic compounds also varies from simple to highly polymerized substances which makes qualitative and quantitative procedures, involving sample preparation and extraction, difficult to standardize. Thus, this makes for a further challenge in the analyses and in furthering the research in this field.

Colorimetric assays, such as Folin-Ciocalteu, have been extensively used to quantify phlorotannins and polyphenolic content in seaweeds. However, since the assay is difficult to standardize and not selective, it has been recommended to use the assay for approximate measurements of an extract's antioxidant potential only [11–15]. Since the colorimetric assays neither separate nor give a correct quantitative measurement of the individual compounds, high-performance liquid chromatography (HPLC) has been the method of choice for separation and quantification of polyphenols in plants. The HPLC with multiple diode array UV-Visible detection (DAD) quantifies according to Lambert-Beer's law (*A* = ε*cl*). A compound's ability to absorb UV-Visible light (*A*) is related to the compound's molar absorptivity value (ε) and molar concentration (*c*). The diversity of molar absorptivity values of polyphenols is almost as large as the number of polyphenols existing; even within the same polyphenol class, there will be differences [16]. In the lack of commercially available standards, one standard is often chosen when total amounts of polyphenols or phlorotannins are quantified. Gallic acid (GA) seems to be the most used standard for total polyphenolic quantification and phloroglucinol (PG) for the phlorotannin quantification in brow algae [17–20]. In addition to the limitations with commercially available standards, HPLC will also suffer from a lack of separation of complex extract matrices and loss of compound amounts due to the irreversible retention on the HPLC column during elution.

In recent years, quantitative 1H NMR (qNMR) have gained increasing attention as a method for quantitative determination of metabolites in complex biological matrices [21–23]. According to the review by Pauli et al. (2012) [22] and references therein, qNMR methods have proven successful when standard chromatographic methods have been ineffective [22]. In general, qNMR can be considered a primary ratio method of measurement in which the analytes can be correlated directly to a calibration standard, and since the reference compound differs from the analytes, generating a calibration curve becomes unnecessary. However, the quantification needs to be validated with reference compounds. Some work on quantification of phlorotannins in brown algae (*Ascophyllum nodosum*, *Fucus vesiculosus*, and *Cystoseira tamariscifolia*) with qNMR has been done using internal standards [14,23].

In this case study, we examined the polyphenolic content of the green algae *Ulva intestinalis* (syn. *Enteromorpha intestinalis*) collected on the west coast of Norway. An optimized extraction of the polyphenolic content was performed. The extract and semi-purified fractions were further analysed utilizing qNMR with an external reference for quantification of the total phenolic content. For comparison, HPLC-DAD and TPC assay analyses were also performed. To further explore the diverse group of polyphenols in *Ulva intestinalis*, qualitative analyses were performed with HPLC-DAD, HPLC-LR, and HR-MS. We entered this case study with the overarching goal of examining which analytical methods could lead to a more reliable value of polyphenolic content in seaweed and, thus, obtain a better view of the grand potential of seaweed phenolics.

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

#### *2.1. Plant Materials*

Samples of *Ulva intestinalis* (syn. *Enteromorpha intestinalis*) were collected in June from the western coast of Norway; Rogn, Ormhilleren (60◦29'38.8" N 4◦55'11.9" E). The voucher specimen of *Ulva intestinalis* was deposited in the Herbarium BG (Voucher no. BG-A-75) at the University Museum of Bergen, Bergen.

#### *2.2. Chemicals*

All chemicals used were of analytical grade. Methanol (≥99.9%), acetonitrile (≥99.8%), trifluoroacetic acid (TFA) and Folin-Ciocalteu reagent were all acquired form Sigma-Aldrich (Sigma-Aldrich, St. Louis, MO, USA). Formic (98–100%) and acetic (99.8%) acids were both acquired from Riedel-de Haën (Honeywell Inc., Charlotte, NC, USA). Luteolin, apigenin, myrcetin, diosmetin, quercetin, caffeic acid, coumaric acid, ferulic acid, sinapic acid, and gallic acid reference standards were all purchased from Sigma–Aldrich (Sigma-Aldrich, St. Louis, MO, USA). The analytical standard of tricin was purchased from PhytoLab (PhytoLab BmbH & Co. KG, Vestenbergsgreuth, Germany), (+)-catechin was purchased from USP (USP, Rockville, MD, USA), and DPPH free radical was purchased from Merck (Merck, Kenilworth, NJ, USA). Deionized water was deionized at the University of Bergen (Bergen, Norway).

#### *2.3. Extraction and Purification*

The collected plant material was washed thoroughly in fresh water and air dried. Dried plant material was stored at −20 ◦C when not used. Dried material was extracted using ASE (Accelerated Solvent Extraction) (Dionex™ ASE™ 350, Thermo Fisher Scientific, Waltham, MA, USA). A dried sample of *Ulva intestinalis* (55.9 g) was mixed with Dionex ASE prep DE sand and added to 66 mL stainless-steel cells with two glass fiber filters placed at the bottom end of the cell, before being extracted using a Dionex ASE 350 Accelerated Solvent Extractor. The extraction procedure consisted of two different methods, one being a pre-soak method, and the other being the primary extraction method. Pre-soaking consisted of extraction at 23 ◦C under 1500 psi. The static extraction period was 1 min with a flush volume of 50% of cell volume, purged with N2 for 70 s, and 100% deionized water was used as the solvent in the pre-soak method. The primary extraction method consisted of preheating for 5 min, and samples were then extracted at 70 ◦C under 1500 psi. Static extraction time was 5 min with a flush volume of 60% of the cell volume, purged with N2 for 100 sec. The solvent used for the primary extraction was a mixture of deionized water and methanol (40:60, *v*/*v*). Primary extraction was repeated two times. The volume of the combined extract was reduced using a rotavapor, and the concentrated aqueous extract was partitioned against ethyl acetate (EtOAc) four times. The contents of both the EtOAc phase and the water phase were examined using HPLC-DAD, HPLC-LRMS, HPLC-HRMS, and colorimetric assays including Total Phenolic Content Assay (TPC) and Total Flavonoid Content Assay (TFC). Before analysis, all phases were carefully reduced to dryness using rotavapor, and, finally, the samples were dried under N2 gas.

The aqueous extract was applied to an Amberlite XAD-7 column and washed with distilled water. Methanol was applied for elution. The pre-eluted washing water was analyzed for polyphenols with HPLC. Collected methanolic fractions (XAD7-A, XAD7-B, XAD7-C) were reduced using a rotavapor and analyzed on analytical HPLC. The XAD-7 fraction A contained the highest number of polyphenols and was chosen to be submitted to preparative HPLC to obtain three purified fractions; prepLC-A1, -A2, and -A3 (Figure 1).

**Figure 1.** Overview of the extraction and purification steps in the *Ulva intestinalis* analysis.

#### *2.4. General Instrumentation*

#### 2.4.1. Preparative HPLC

The preparative HPLC system consisted of a Gilson 321 pump (Gilson Inc., Middleton, WI, USA), an Ultimate 3000 variable wavelength detector (Dionex, Thermo Fisher Scientific, Sunnyvale, CA, USA), and a 25 × 2.12 cm (10 μm) UniverSil C18 column (Fortis Technologies Ltd., Neston, UK). Two solvents were used: (A) super distilled water (0.1% acetic acid) and (B) acetonitrile (0.1% acetic acid) with initial conditions of 90% A and 10% B followed by an isocratic elution for the first 5 minutes, and the subsequent linear gradient conditions, 5–18 min: to 16% B, 18–22 min: to 18% B, 26–31 min: to 28% B, 31–32 min: to 40% B, 32–40 min: isocratic at 40% B, 40–43 min: to 10% B. The flow rate was 15 mL/min, and the aliquots of 750 μL were injected.

#### 2.4.2. Analytical HPLC-DAD

All HPLC-DAD analyses were performed on an Agilent 1260 Infinity HPLC system (Agilent Technologies, Santa Clara, CA, USA) equipped with a 1260 diode array detector (DAD) and a 200× C analysis was performed using two solvents, (A) super distilled water (0.5% TFA) and (B) acetonitrile (0.5% TFA), in a gradient (0–10 min: 95% A + 5% B, 10–20 min: 85% A + 15% B, 20–34 min: 60% A + 40% B. 34–35 min: 95% A + 5% B). The flow rate was 1.0 mL/min, and aliquots of 20 μL were injected with an Agilent 1260 vial sampler. UV-Vis absorption spectra were recorded during the HPLC analysis over the wavelength range of 200–600 nm in steps of 2 nm.

The established HPLC method was validated for linearity, sensitivity, precision, and accuracy. Table 1 presents data for calibration curves, test ranges, limit of detection (LOD), and limit of quantification (LOQ) for gallic acid. The LOD and LOQ were calculated based on the standard deviation of *y*-intercepts of the regression line (Sy) and the slope (S), using the equations LOD = 3.3 × Sy/S and LOQ = 10 × Sy/S.

**Table 1.** Calibration curve, limit of detection (LOD), and limit of quantification (LOQ) for gallic acid (GA) (Sigma-Aldrich) at 280 nm and 330 nm.


#### 2.4.3. HPLC-LRMS and HPLC-HRMS

Liquid chromatography low-resolution mass spectrometry (HPLC-LRMS) (ESI+/ESI−) was performed using an Agilent Technologies 1260 Infinity Series system and an Agilent Technologies 6420A triple quadrupole mass spectrometry detector. The following conditions were applied: ionization mode: positive/negative, capillary voltage = 3000 V, gas temperature = 300 ◦C, gas flow rate = 3.0 L/min, acquisition range = 100–800 *m*/*z*. The elution profile for HPLC consisted of the following gradient: 0–3 min: 90%A + 10%B, 3–11 min: 86%A + 14%B, 11–15.5 min: 60%A + 40%B, 15.5–17 min: 90%A + 10%B, at a flowrate = 0.3 mL/min, where solvent A was super distilled water (0.5% formic acid), and solvent B was acetonitrile (0.5% formic acid). A 50 × 2.1 mm internal diameter, 1.8 μm Agilent Zorbax SB-C18 column was used for separation. Calibration curve of Apigenin ran on HPLC-LRMS and used for quantification is listed in Table 2.


**Table 2.** Calibration curve, limit of detection (LOD), and limit of quantification (LOQ) for apigenin (Sigma-Aldrich) acquired using HPLC-LRMS.

Liquid chromatography high-resolution mass spectrometry (HPLC-HRMS) (ESI+/TOF) was performed using an AccuTOF JMS-T100LC (JEOL, Peabody, USA) mass spectrometer in combination with an Agilent Technologies 1200 Series HPLC system. The following instrumental settings/conditions were used: ionization mode: positive, ion source temperature = 220 ◦C, needle voltage = 2500 V, desolvation gas flow = 4 L/min, nebulizing gas flow = 3 L/min, orifice1 temperature = 125 ◦C, orifice2 voltage = 10 V, ring lens voltage = 20 V, ion guide RF voltage = 1600 V, detector voltage = 2350 V, acquisition range = 15–1000 *m*/*z*, spectral recording interval = 0.50 sec, wait time = 0.033 nsec, and data sampling interval = 2 nsec. The elution profile for HPLC consisted of the same gradient and column as described for HPLC-LRMS, but the flowrate was increased to 0.35 mL/min.

#### 2.4.4. NMR Spectroscopy

Quantification of the extracts of *Ulva intestinalis* was performed using 1H NMR analyses on a Bruker 600 MHz instrument (Bruker BioSpin, Zürich, Switzerland). All spectra were recorded in DMSO-*d*<sup>6</sup> at 25 ◦C. The pulse sequence applied was *zg30* with the following acquisition parameters: sweep width of 19.8 ppm, 64 k data points, 16 scans, and 2 dummy scans. The relaxation delay, d1, was set to 40 sec (equal to 5 × *T*1,max) to ensure complete relaxation between scans. The spectra were processed using a line broadening of 0.3 Hz. The crude extract was used for *T*<sup>1</sup> measurements, utilizing the *t1ir* pulse sequence with a sweep width of 19.8 ppm, 16 k data points, 8 scans, 2 dummy scans, and 9 different inversion recovery delays between 1 ms and 5 s. Measured *T*<sup>1</sup> values ranged from 1.0–8.1 s.

Quantification using the 1H NMR spectra was performed using the ERETIC2 function in TopSpin with DMSO2 (10 mM) as an external reference. The DMSO2 signal (~3.0 ppm) was integrated and defined as the ERETIC reference (No. H = 6, Mm = 94.13 g/mol, V(sample) = 0.75 mL, C = 10 mM).

Reference compounds for validation were gallic acid (GA), *p*-coumaric acid, ferulic acid, (+)-catechin, and luteolin (10 mM, DMSO-*d*6). An average standard deviation of < 10% was observed. The integrations were repeated three times.

Two-dimensional heteronuclear single quantum coherence (1H-13C HSQC), heteronuclear multiple bond correlation (1H-13C HMBC), and double quantum filtered correlation (1H-1H DQF COSY) spectra were also recorded on the Bruker 600 MHz instrument.

#### *2.5. Total Phenolic Content Assay*

For the determination of total phenolic content, the Folin-Ciocalteu total phenolic content assay (TPC) was used. The method used was adapted from Ainsworth and Gillespie (2007) [24]. 200 μL of the sample or standard was added to the cuvettes (10 × 45 mm, 3 mL), followed by 400 μL 10% (*v*/*v*) Folin–Ciocalteu reagent in super distilled water. Further, 1600 μL 700 mM Na2CO3 in super distilled water was added to the cuvettes. The mixture was incubated for 30 minutes, and the absorbance was measured at 765 nm using a Shimadzu UV-1800 UV spectrophotometer and a Shimadzu CPS-100 cell positioner (Shimadzu, Kyoto, Japan). Data was expressed as gallic acid equivalents (GAE). An incubation time of 2 h was also tested.

#### *2.6. Total Flavonoid Content Assay*

For the determination of the total flavonoid content, 2 mL test solution (standard or sample) was added to four cuvettes (10 × 45 mm, 3 mL) and the absorbance measured at 425 nm with solvent in the reference cuvette. An aliquot of AlCl3 solution (0.5 mL, 1%, *w*/*v*) was added to three of the four cuvettes, and the same volume of solvent was added to the fourth (blank sample). The content of the cuvettes was stirred thoroughly, and the absorbance measured at 1 minute intervals at 425 nm for 10 minutes at 22 ◦C. For quantitative analysis apigenin was chosen as the reference compound (concentration range of 1–500 μg/mL). Procedure modified from P ˛ekal and Pyrzynska (2014) [25].

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

#### *3.1. Quantification of Polyphenols in* Ulva Intestinalis

In this work, extraction of polyphenols was performed after optimization of extraction parameters utilizing a Dionex ASE 350 extraction instrument (see Section 2.3). Aliquots (10 mL) of the different phases, ASE (Accelerated Solvent Extractor) Crude, (A) EtOAc and (B) water (see Figure 1) were sampled and dried for weight determination and further quantification with HPLC-DAD, qNMR, TPC, and TFC. The results of the different quantification methods are shown in Tables 3–5.

**Table 3.** Quantification of polyphenols in the crude extract and liquid–liquid extraction phases of crude with HPLC.


PP = polyphenol; (A) EtOAc = ethyl acetate phase; (B) water phase; GAE = gallic acid equivalents; DW = Dry Weight.

**Table 4.** Quantification of polyphenols in the crude extract and liquid–liquid extraction phases of crude with qNMR.


PP = polyphenol; (A) EtOAc = ethyl acetate phase; (B) water phase; GAE = gallic acid equivalents; 330 Mw eq. = equivalents of average mass found from MS; 2H, 4H, and 6H = assumptions made related to the number of aromatic 1H in each polyphenolic spin system; DW = Dry Weight.

**Table 5.** Quantification of polyphenols in the crude extract and liquid–liquid extraction phases of crude with total phenolic content (TPC).


PP = polyphenol; (A) EtOAc = ethyl acetate phase; (B) water phase; GAE = gallic acid equivalents; DW = Dry Weight.

#### *3.2. Quantification Utilizing High-Performance Liquid Chromatography (HPLC) with Wavelength Detector (DAD)*

Quantification of polyphenols in plants and foods has been a topic of discussion and research for years, and among the different methods HPLC-DAD it has been the method of choice due to the possibility of separation of compounds before individual quantification. However, with the use of retention times, absorption spectra, and molar absorptivity, the technique is often limited when it comes to simultaneous determination of polyphenols of different groups [9]. Table 6 illustrates the different area responses observed in HPLC for different standards with the same concentration, reflecting the molar absorptivity differences.

**Table 6.** Illustration of molar absorptivity differences expressed with HPLC integrated peak areas (280 nm and 330 nm) of selected standards (5 mM) used in polyphenolic quantification.


HCA = hydroxycinnamic acid, HBA = hydroxybenzoic acid

When dealing with complex polyphenolic mixtures with unknown identities, which is the case for seaweeds, one standard is often selected for quantification. Traditionally, gallic acid is chosen for total polyphenolic quantification and phloroglucinol (PGE) for total phlorotannin quantification as seen for brown algae [17–20]. In this work, gallic acid (GA) was chosen as the reference standard, since the nature of the polyphenols in the green algae *U. intestinalis* was unknown, and since we wanted to compare different quantification methods. However, there is no doubt that the estimation of the total polyphenol content will suffer from this.

The HPLC peaks with maximum intensity in the 280 nm (*R*t: 1–15 min) were quantified according to the 280 nm GA standard curve (Table 1), while peaks with maximum intensity in the 330 nm (*R*t: 15–35 min) window were quantified according to the 330 nm GA standard curve. This resulted in an HPLC-DAD quantification of 1.1% polyphenols in the algae, based on quantification on the ASE crude extract (11.3 ± 1.4 mg GAE/g DW) (Table 3). The recovery of the polyphenols after the liquid-liquid ethyl acetate partition was quantified to be 1.2% (12.1 ± 0.5 mg GAE/g DW), almost evenly distributed into the (A) EtOAc phase (0.7%) and the (B) water phase (0.6%). Thus, the total recovery for A + B was relatively close to the initial amounts found in the crude.

#### *3.3. Quantitative NMR (qNMR)*

In order to get closer to a "true" estimation of polyphenol content in seaweeds, quantifications using 1H NMR (qNMR) were performed (Table 4). One of the advantages of qNMR is that there is no need to consider the large variation observed regarding the molar absorptivity of different phenolic compounds (Table 6) nor the loss of sample during chromatography as with HPLC analyses. When quantifying polyphenols from NMR, one can consider two regions for quantification: the –OH spectral region, as shown by Nerantzakie et al. [23], or the aromatic 1H region [14,26]. Nerantzaki et al. presented a method for total phenolic content determination of crude plant extracts based on phenol type –OH resonances in the region between 14–8 ppm. Signals were selected after observation of elimination, or reduction, of the signal intensities after irradiation of the residual water resonance. In our marine *U. intestinalis* samples, the phenol –OH type resonances were observed at low intensities and were too broad to perform reliable integration. The broad signals may be attributed to the nature of the marine extract, containing many different types of phenol –OH resonances. Additionally, the ASE crude and the water phase contained some water, even after careful drying, which increases the phenol –OH exchange with the water peak. The 10–8.5 ppm region of the EtOAc phase (Figure 2) showed several sharp signals; however, these signals were found to not represent phenol –OH resonances due to the fact of their observed <sup>1</sup>*J*CH correlations in the HSQC spectrum.

**Figure 2.** 1H-NMR spectrum (600 MHz) for ASE crude (blue), (**A**) EtOAc phase (red), and (**B**) water phase (green) recorded in DMSO-*d*<sup>6</sup> at 25 ◦C. 2D spectra were used to deselect peaks in the 8.5–6 ppm region belonging to the same spin system, avoiding multiple quantification.

For qNMR calculations, characteristic aromatic signals in the 8.5–6 ppm region of the 1H NMR spectra were integrated individually, and quantifications were added together to yield the total phenolic content (Section 2.4.4, Figure 2) [21,25]. Additionally, two-dimensional NMR spectra, such as COSY, HSQC and HMBC, were recorded to deselect signals belonging to the same molecule as far as possible in order to avoid multiple quantifications. The qNMR calculations were validated with quantification of standards (Section 2.4.4). Quantifications were calculated using the ERETIC2 function in TopSpin (Bruker) with DMSO2 as an external reference (C = 10 mM). However, to quantify the signals, a molar mass is needed. The molar mass of gallic acid was chosen in order to obtain comparable results. Quantifications were also calculated using an average molar mass of 330 g/mol based on observed masses from the MS analyses (Table 4). Additionally, an average value of aromatic protons found in each polyphenolic spin system must be chosen. This assumption will also introduce uncertainty. Nerantzakie et al. [23] made their quantification on phenol –OH and used an average of 2 OH for each spin system related to their standard, caffeic acid. In Table 4, the polyphenolic content calculation utilizing different average aromatic protons are shown, resulting in a 33% difference between the maximum (2 aromatic H) and minimum (6 aromatic H) values calculated. Based on our tentatively

identified compounds in Table 7 it seemed like 4 aromatic protons (H) was a reasonable assumption. The qNMR method thus yielded a polyphenolic content of 5.3% in the crude (52.9 ± 5.2 mg 330 Mw eq./g DW). Due to the parallel numbers, using gallic acid (170 g/mol) and 2 aromatic protons yielded similar results (Table 4).


**Table 7.** Overview of tentatively identified low-mass polyphenols/simple phenolics at different stages of purification with HPLC-LRMS.

HCA = hydroxycinnamic acid, HBA = hydroxybenzoic acid, HT = hydrolysable tannins, PAC = proanthocyanidin. \* Several possible isomers; HR HR-LC-MS mass; + = identity confirmed with standard on LR-LC-MS, - = identity not confirmed with standard on LR-LC-MS.

#### *3.4. Colorimetric Assays: Total Phenolic Content (TPC) and Total Flavonoid Content (TFC)*

The Folin–Ciocalteu assay is the most common assay used to quantify phenolic content (TPC) in both terrestrial plants and seaweeds. However, the assay is debatable due to the lack of standardization and lack of specificity in the reaction mechanism resulting in the colorimetric quantification [11–15,27]. This is of importance for all colorimetric assays, including the total flavonoid content (TFC) assay [25,28]. With increasing purity of the samples, direct quantitative measurements seem to be more reliable. However, the difficulty of standardizing this assay does not seem to be without importance.

The TPC assay (Table 5) resulted in a total of 0.4% in the ASE crude (5 ± 1 mg GAE/g DW), with a recovery of 0.04% in the (A) EtOAc phase (0.035 ± 0.001 mg GAE/g DW) and 0.4% in the (B) water phase (0.4 ± 0.1 mg GAE/g DW). Relatively high standard deviations were observed for the aqueous phases, potentially reflecting the lack of reliability of the method and difficulties with standardization.

The relative partition of polyphenols found between the two phases (A:B) in the TPC assay seem to follow the pattern observed from the qNMR quantification (10:90) (Table 4), rather than the partition ratio found in the HPLC-DAD analyses (50:50) (Table 3). The different ratio observed from the HPLC analyses is most likely due to the impact of molar absorptivity difference between the standard used and the compounds present.

The occurrence of flavonoids in algae is a central topic [29–32], and we chose to run a TFC assay in parallel with our attempts to identify flavonoids in our extracts (Table 8). The TFC assay gave a total of 0.03% flavonoids in the ASE Crude (0.2 ± 0.4 mg apigenin eq./g DW) and 0.13% in the (A) EtOAc phase (0.2 ± 0.4 mg apigenin eq./g DW). No flavonoids were detected in the (B) water phase with the TFC method.


**Table 8.** Quantification of flavonoids in the crude extract and liquid–liquid extraction phases of crude with total flavonoid content (TFC).

<sup>a</sup> Three parallels measured from (0–34 mg); n.d.= not detected; PP = polyphenol; FL = flavonoid; (A) EtOAc = ethyl acetate phase; (B) water phase; TFC = total flavonoid content; DW = Dry Weight.

#### *3.5. Qualitative Analysis of Polyphenols in Ulva intestinalis*

After ASE extraction of the polyphenols (Figure 3; HPLC profile and selected UV-Vis spectra) and partition of the aqueous crude extract against ethyl acetate, the concentrated water phase (B) was applied to a XAD-7 column, washed with distilled water, and then eluted with methanol (Figure 1). The pre-eluted washing water was analyzed for polyphenols with HPLC-DAD. Collected methanolic fractions (XAD7 A–C) were reduced using a rotavapor and analyzed using analytical HPLC. The XAD-7 fraction A showed the highest polyphenol content and was chosen to be submitted to preparative HPLC to obtain three major fractions (prepLC A1–A3, Figure 1). The EtOAc phase was also submitted to preparative HPLC. The liquid–liquid partition with ethyl acetate gave some selectivity with respect to separation of compounds as seen in Figure 4. The compounds found in the EtOAc phase were most likely less polar and seemed to have a shorter chromophore compared to compounds observed in the water phase. The compounds in the water phase also showed an additional absorption band around 412–414 nm.

The preparative HPLC gave some separation of compounds; however, the samples were still complex. All the phases and fractions underwent extensive analyses with HPLC-DAD, HPLC-LRMS, HPLC-HRMS, and NMR. The results of the HPLC-LRMS analyses are shown in Table 7, giving an overview of the tentatively identified compounds.

Fragmentation patterns were difficult to obtain due to low concentrations. The ESI-MS spectra were recorded in both positive and negative modes. The masses of a luteolin-isomer ((M+H)+, calculated: 287.05556, exact: 287.05599, C15H10O6, Δppm 1.5) and a rhamnazin-isomer ((M+H)+, calculated: 331.08178, exact: 331.08178, C17H14O7, Δppm 1.24) were confirmed with HPLC-HRMS. The rhamnazin-isomer (*m*/*z* 331.08178) did not overlap with the commercial standard tricin (330 Mw) in the HPLC-LRMS SIM scan.

The most conclusive evidence of the presence of flavonoids in the green algae *U. intestinalis* was found in the late preparative fraction: prepLC-A3 (Figure 5). This fraction contained many of the peaks observed between 15 and 35 min in the HPLC profile of the crude (330 nm) (Figure 3). Several of the flavonoid masses found were tentatively identified from this fraction (Table 7) which has its origin from the water phase (B). The TFC assay did not detect any flavonoids in the water phase (Table 8) which illustrates the problem with relaying on these colorimetric assays. One flavonoid in the prepLC-A3 fraction was identified to be apigenin, using overlaid an HPLC-LRMS SIM scan at *m*/*z* 271 (M+H)<sup>+</sup> with an apigenin standard (Figure 5). The amount of the apigenin in the algae was found to be 2.617 ng/g (DW) using an apigenin calibration curve (Table 2).

**Figure 5.** Overlaid HPLC-LRMS (+ESI) SIM Scan at *m*/*z* 271 of prepLC-A3 fraction (red line, C (Api, HPLC-LRMS) = 2.62 ng/g DW) and apigenin standard (C = 1.00 mM) (black line).

#### **4. Conclusion**

This case study provides an optimized extraction process for polyphenolic extraction of algae. The total polyphenolic content was quantified with qNMR (5.3%), HPLC-DAD (1.1%), and TPC (0.4%). Flavonoids and polyphenolic acids were tentatively identified in *Ulva intestinalis* samples. Apigenin was confirmed in one of the semi-purified fractions.

The same samples yielded different total phenolic contents when utilizing the different analytical methods, highlighting the difficulties related to polyphenolic quantification in extracts. All methods utilized in this study depend on assumptions and, thus, also uncertainty. This will be of special importance when analyzing complex samples at low concentrations as is the case for the polyphenolic content in marine algae. Further standardization and optimization of total phenolic quantifications of marine algae samples should be researched.

**Author Contributions:** M.J. and M.E.W. conceived and designed the experiments; M.J. collected the algae materials; J.U. contributed with support and discussions concerning the qNMR experiments; B.H. contributed with support and discussions concerning the HPLC-LRMS methods and instrumentation; B.H. recorded the HPLC-HRMS data; M.E.W. performed all the laboratorial work, the HPLC-DAD, TPC, qNMR, and HPLC-LR and HRMS experiments; K.K. developed the ASE-350 extraction method and the HPLC-LRMS method, and modified and performed the TFC assay. M.E.W. and M.J. analyzed the data. M.J. and M.E.W. wrote the paper. All authors read and approved the final manuscript.

**Funding:** The authors are grateful to the University of Bergen, Norway, for Open Access funding (710029/884). This work was partly supported by the Bergen Research Foundation (BFS-NMR-1), Sparebankstiftinga Sogn og Fjordane (509-42/16), and the Research Council of Norway through the Norwegian NMR Platform, NNP (226244/F50).

**Acknowledgments:** M.E.W. gratefully acknowledges the Norwegian Research Council, NFR, and Alginor ASA (Haugesund) for her fellowship

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

#### **References**


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

## *Article* **Chlorophyll Oxidative Metabolism During the Phototrophic and Heterotrophic Growth of** *Scenedesmus obliquus*

**Mariana Manzoni Maroneze 1, Leila Queiroz Zepka 1, Eduardo Jacob Lopes 1, Antonio Pérez-Gálvez <sup>2</sup> and María Roca 2,\***


Received: 21 October 2019; Accepted: 27 November 2019; Published: 29 November 2019

**Abstract:** Different cultivation strategies have been developed with the aim of increasing the production rate of microalgal pigments. Specifically, biotechnological approaches are designed to increase antioxidant metabolites as chlorophyll and carotenoids. However, although significant advances have been built up, available information regarding both the chlorophyll metabolism and their oxidative reactions in photobioreactors is scarce. To unravel such processes, the detailed chlorophyll and carotenoid fraction of *Scenedesmus obliquus* has been studied by HPLC-ESI/APCI-hrTOF-MS from phototrophic and heterotrophic cultures. *Scenedesmus* is provided with a controlled strategy of interconversion between chlorophyll *a* and *b* to avoid the formation of reactive oxygen species (ROS) at high irradiances in addition to the photoacclimation of carotenoids. Indeed, precise kinetics of 132-hydroxy- and 151-hydroxy-lactone chlorophyll metabolites shows the existence of a chlorophyll oxidative metabolism as a tool to manage the excess of energy at high light conditions. Unexpectedly, the oxidation under phototrophy favored chlorophyll *b* metabolites over the chlorophyll *a* series, while the heterotrophic conditions exclusively induced the formation of 13*2*-hydroxy-chlorophyll *a*. In parallel, during the first 48 h of growth in the dark, the chlorophyll fraction maintained a promising steady state. Although future studies are required to resolve the biochemical reactions implied in the chlorophyll oxidative metabolism, the present results agree with phytoplankton metabolism.

**Keywords:** phototrophic; heterotrophic; *Scenedesmus*; chlorophylls; carotenoids; hydroxy-chlorophyll; oxidative metabolism; ROS; lactone-chlorophyll; photoacclimation

#### **1. Introduction**

Chlorophyll and carotenoids are challenging compounds in microbial biotechnology that find several applications in the food industry. The present food market trend is towards more natural ingredients, and colorants are not an exception [1]. Hence, artificial food colorants have been associated with health problems and consequently new sources of natural colorants are under investigation. Although natural food colorants have been traditionally extracted from fruits and vegetables sources, microalgae are currently a promising natural resource. Several advantages as the fast growth, the high pigment concentration, and the physiologically plasticity, make microalgae the new objective of biotechnological companies for pigment production. Among them, *Scenedesmus obliquus* stands out as source of pigments in food and cosmetics as well as is considered for human consumption [2,3]. In addition, chlorophyll and carotenoids exert beneficial health properties for human beings that increase their value as functional ingredients [4]. Specifically, both groups of pigments have shown to develop antioxidant activities. The antioxidant behavior of chlorophylls is highly dependent of the type of chlorophyll derivative, with significant antioxidative performances among metabolites [5]. The porphyrin structure, the central magnesium, and the functional group at C7 seem to be determinants for the antioxidant activity [6–9]. In the same line, carotenoids are highly appreciated by their antioxidant properties [10]. Consequently, the production of chlorophyll and carotenoids is one of the most successful applications of microalgal biotechnology [11], although the improvement of the feasibility of their commercial production through better cultivation strategies is still the main goal.

Photo-autotrophy is the classical culture system to grow microalgae, where the energy source comes from the sunlight and the carbon source from the atmospheric CO2. Different light regimes (intensity, photoperiod, and wavelengths) generate different chlorophyll and carotenoid patterns. It is generally assumed that sub saturating light intensities induce higher chlorophyll synthesis, while high light irradiation reduces the chlorophyll content [12]. On the contrary, specific microalgae (*Haematococcus pluviales, C. zofingiensis*) enhance the production of secondary carotenoids when grown with high light intensities [13].

Open and closed photobioreactors present several disadvantages, such as the presence of contaminants, the need of robust species, or the requirement of vigorous mixing [13]. Consequently, growing microalgae in heterotrophic conditions in conventional bioreactors is at present an attractive and economical option. An additional advantage is that green algae can synthesize chlorophylls in dark conditions unlike angiosperms. This is possible thanks to the presence of a light-independent POR (protochlorophyllide *a* oxidoreductase) enzymatic process, one of the key enzymes in the chlorophyll biosynthetic pathway [14]. However, only a few microalgal species have been shown to grow in the dark so far, because the capacity of sugar utilization is not a universal strategy. In this sense, a constitutive glucose transport and utilization system has been reported for *Scenedesmus obliquus* growth [15]. In fact, different new strategies are continuously developing. Cyclic autotrophic/heterotrophic cultivation, where organic carbon is added during light or during the dark phase has been studied to optimize the production of chlorophylls and carotenoids [16]. The strategy of cultivation in two stages has been explored as an alternative to avoid the division between cell growth and the production of secondary metabolites [17]. The first phase is dedicated to obtaining the maximum biomass production, followed by a stressed second phase to increase the accumulation of lipid-derived compounds. In fact, such strategy (two-stage heterotrophy/photoinduction) has been successfully applied in *Scenedesmus* reaching values of lutein productivity 1.6 times higher than in autotrophic conditions [18]. With the same aim, that is, to enhance the lutein productivity, the conditions during mixotrophic cultures of *S. obliquus* have been optimized, determining the best operating parameters for photoperiod, source of light, nutrients and batch system [11]. Different conditions have been established to modulate, improve, or control the chlorophyll content in different *Scenedesmus* sp. [19,20]. Light conditions, stirring, depletion of nutrients, open fields or bioreactors, and carbon source are among the most explored variables [12]. *Scenedesmus* seems to have a lower sensitivity to photoinhibition and a higher capacity to adapt to high irradiance conditions by increasing its photosynthetic capacity, in comparison with other species such as *Chlorella* [21]. However, the following step is to analyze the chlorophyll profile in detail, to decipher the responsible mechanism(s) for the synthesis and degradation of chlorophylls in response to modifying parameters.

The biochemical reactions implied during the chlorophyll degradation pathway have been unraveled in higher plants [22], but in green microalgae only a few steps have been discovered [23,24]. Thus, the initial catabolic steps of chlorophyll *a* are the de-esterification of phytol (Figure 1) and the loss of the central magnesium to finally yield pheophorbide *a*. For these consecutive reactions, two plausible alternative routes have been proposed [22], which are potentially catalyzed by different enzymes. The last proposal has been to postulate that both enzymatic systems can operate simultaneously, although at different functional levels [25]. Next, the macrocycle of the pheophorbide *a* intermediate is

oxygenolytically open, yielding a sequence of linear chlorophyll catabolites denominated phyllobilins (Figure 1). Regardless, it has been established that knowledge level of chlorophyll degradation in microalgae is at present in a preliminary stage. Specifically, as it was regarding that of the higher plants 30 years ago. A very close relationship has been always suggested between the chlorophyll catabolic pathway in chlorophytes and in higher plants [26], which is not unlikely assuming the phylogenetic relationship between both taxonomic groups. In fact, open chlorophyll catabolites with similar structures to those of phyllobilins have been identified in *Chlorella* [26,27] or in *Desmodesmus subpicatus* [28]. In parallel, chlorophylls are subjected to an oxidative metabolism [29,30]. At present, two reactions have been identified. Hence, the chlorophyll skeleton can be oxidized at C132 to form 132-hydroxy-compounds and secondly, while the isocyclic ring can further react to form a lactone group, that is, the 152-hydroxy-lactone chlorophyll derivatives [31] (Figure 2). Their formation could arise from different pathways, enzymatic [30] or by an increase in oxygen reactive species, and even they can also be produced in dark anoxic conditions [32]. Regardless, the presence of hydroxy-chlorophylls has been related with conditions of high environmental presence of peroxide species, as the former are the main products of chlorophyll *a* in presence of the latter [33]. Therefore, hydroxy-chlorophylls are related to the response to oxidative stress. Physiologically, they have been associated with senescence [34], virus infection [35], and even cell death [36]. In any case, hydroxy-chlorophylls are common chlorophyll metabolites found in phytoplankton species in their natural environments [36,37]. For some authors, oxidative chlorophylls are the origin of "petrochlorophylls", as they have been identified in numerous phytoplankton sediments [38]. However, to the best of our knowledge, chlorophyll oxidation has not been analyzed in relation to microalgae cell culture.

**Figure 1.** Chlorophyll degradation pathway. CHL: chlorophyllase, SGR: stay-green, PPH: pheophytinase.

**Figure 2.** Oxidative chlorophyll reactions, following previous proposal [31]. The structures correspond only with the V ring (or isocyclic ring) of the chlorophyll molecule. The wavy line means the rest of the chlorophyll structure (see Figure 1). The oxidative reactions can be performed over diverse chlorophyll compounds (see the tables).

As stated before [39], a deep knowledge of their metabolic pathways is necessary to select the best cultivation conditions to improve the microalgal pigment production. Although significant advances have been developed to maximize the total chlorophyll and carotenoid content in some microalgae species, it is necessary to understand the individual behavior within the heterogenous pigment profile. It is necessary not only to consider the total pigment content, but also to determine which metabolites are producing or degrading. The aim of this study was to analyze in detail the chlorophyll and carotenoid metabolism during the phototrophic and heterotrophic cultivation of *Scenedesmus* with special emphasis in the oxidative reactions occurring at the chlorophyll fraction.

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

#### *2.1. Microorganisms and Culture Media*

The axenic culture of *Scenedesmus obliquus* (CPCC05) was supplied by the Canadian Phycological Culture Centre (Waterloo, Canada). We applied the following incubation conditions, 26 ◦C, photon flux density of 30 μmol m−<sup>2</sup> s−<sup>1</sup> and a photoperiod of 12 h to obtain the stock cultures, which were propagated and maintained in synthetic BG11 medium [40].

#### *2.2. Cultivation Conditions*

The phototrophic experiments were carried out in a 2 L bubble column photobioreactor (Tecnal, Piracicaba-SP, Brazil) operated in batch mode [41]. We applied the following experimental conditions: 100 mg/L for the initial cell concentration, and 26 ◦C for the isothermal reactor, which was fed with 2 L of B11 medium, pH set to 7.6, 150 μmol m−<sup>2</sup> s−<sup>1</sup> for the photon flux density and a light cycle of 24:0 h (light:dark). Continuous aeration of 1 VVM (volume of air per volume of culture per minute) was applied with the injection of air enriched with 15% carbon dioxide. The conditions for the heterotrophic cultivations were set up in a 2 L bubble column bioreactor operating under a batch regime [42]. It was operated at 26 ◦C in the absence of light, with a carbon/nitrogen ratio of 20, pH adjusted to 7.6, aeration of 1 VVM, and initial cell concentration of 100 mg/L. The culture medium consisted of BG11 synthetic medium supplemented with 12.5 g/L of D-glucose.

#### *2.3. Kinetic Parameters*

We used the biomass data to calculate the biomass productivity [PX = (Xi − Xi−1) × (ti − ti−1) <sup>−</sup>1, mg/L h], the maximum specific growth rate [ln(Xi/X0) = <sup>μ</sup>max <sup>×</sup> t, 1/h], and generation time [tg = 0.693/μmax, h]. Hence the Xi is the biomass concentration at time ti (mg/L), while Xi−<sup>1</sup> is the biomass concentration at time ti−<sup>1</sup> (mg/L) and X0 is the biomass concentration at time 0. μmax: maximum specific growth rate (h–1). Residence time (t, in h) is defined as the time required for cells to reach the end of the stationary phase.

#### *2.4. Extraction of Photosynthetic Pigments*

Aliquots of microalgae or cyanobacteria biomass (5 mL) were filtered with a Whatman grade GF/F glass microfiber filter (47-mm diameter, Merck, Darmstadt, Germany), and immediately frozen at −80 ◦C [43]. The filter was grinded with liquid nitrogen into powder and mixed with 10 mL of DMF:water (9:1) under stirring at 4 ◦C for 15 min and spinning (10,000 rpm, 5 min). Subsequently, the solvent phase was collected in a separation funnel whereas the solid residue was re-extracted with 10 mL hexane, ultrasonicated (5 min, 720 W), and vortexed (5 min). Then, 10 mL NaCl solution (10% *w*/*v*) was added to the mixture, centrifuged (10,000 rpm, 5 min) and the supernatant was added to the first extract in the funnel. Finally, the pellet was dissolved with 10 mL diethyl ether in an ultrasonic bath (5 min, 720 W) and finally vortexed for 5 min. Then, the solution was mixed with 10 mL NaCl solution (10% *w*/*v*) and the mixture was centrifuged (10,000 rpm, 5 min) and added to the previous extracts in the funnel. There, the mixed solvent layers were extracted with diethyl ether and NaCl solution (10% *w*/*v*). The water layer was discarded, and the organic phase was concentrated to dryness in a rotary evaporator. The residue was dissolved in acetone. Samples were stored at −20 ◦C until analysis within 1 week.

#### *2.5. Identification of Photosynthetic Pigments by HPLC-ESI*/*APCI-HRTOF-MSn*

The chromatographic separation of the individual chlorophyll derivatives and carotenoids was achieved in a Dionex Ultimate 3000RS U-HPLC equipment (Thermo Fisher Scientific, Waltham, MA, USA). The column applied for chlorophyll pigments was a reversed-phase C18 column (200 × 4.6 mm i.d., Teknokroma, Barcelona, Spain), 3 μm particle size, while the elution gradient was the one described previously [44]. The separation of the carotenoid profile required different chromatographic conditions. A reversed-phase C30 column (250 × 4.6 mm i.d., YMC, Schermbeck, Germany), with 3 μm particle size, was applied with the elution gradient described earlier [45,46]. For chlorophyll and carotenoids, the injection volume was 30 μL and the flow rate utilized was 1 mL/min. The UV-visible spectra of the chromatographic peaks were recorded in the 300–700 nm range with a PDA detector. Subsequently, a split post-column of 0.4 mL/min was introduced directly on the mass spectrometer ion source (micrOTOF-QIITM High Resolution Time-of-Flight mass spectrometer with Qq-TOF geometry, Bruker Daltonics, Bremen, Germany). The analysis was developed with an ESI interface (for chlorophyll compounds) or an APCI source (for carotenoid compounds). The instrument was operated in positive ion mode and scanning the *m*/*z* values in the 50–1200 Da range. We operated the acquisition of the mass spectra in broad-band Collision Induced Dissociation mode (bbCID), so that MS and MS/MS spectra were recorded simultaneously. The instrument control was performed with Bruker Compass HyStar software (Bruker Daltonics version 3.2, Bremen, Germany), whereas the processing of MS data was made with the Bruker Compass DataAnalysis software (Bruker Daltonics version 4.1, Bremen, Germany). For the automated screening of signals corresponding to identified chlorophyll derivatives and carotenoids on the EICs, we applied the TargetAnalysisTM software (Bruker Daltonics version 1.2, Bremen, Germany). The validation of the automated identifications was carried out according to different filtering rules, including mass accuracy (tolerance limit set at 5 ppm) and isotopic pattern comparison calculated with the SigmaFitTM (Bremen, Germany) algorithm (tolerance limit set at 50) [44]. The interpretation of the MS/MS spectra and the consistency of the product ions, which have to fulfil the previous filtering rules for mass accuracy and isotopic pattern, was developed with the SmartFormula3DTM (Bremen, Germany) module [44]. The software MassFrontierTM software (Thermo ScientificTM version 4.0, Waltham, MA, USA) allowed the acquisition of the in silico tandem MS spectra of the filtered analytes to compare the theoretical product ions with the corresponding experimental ones. This software allows the evaluation of different product ions when different isomers show the same bbCID spectrum.

#### *2.6. Quantification of Photosynthetic Pigments by HPLC-UV-Visible Detection*

The identified pigments were quantified by reversed-phase HPLC using a Hewlett-Packard HP 1100 liquid chromatograph with the same columns and eluent gradients as for the MS analyses. The on-line UV-visible spectra were recorded in the 350–800 nm wavelength range. Sequential detection was performed at 410, 430, 450, and 666 nm with a photodiode-array detector. Data were collected and processed with the HP ChemStation (Rev.A.05.04) software (Agilent Technologies, Waldbronn, Germany). Calibration curves (amount versus integrated peak area) were obtained by the least-squares linear regression analysis for quantification of pigments. The concentration range considered to build the calibration equations was ascertained from the observed levels of the pigments in the samples. Triplicate injections were made for five different volumes of each standard solution.

#### *2.7. Statistical Analysis*

Normality of data (mean values of three independent measurements) was checked with the Shapiro-Wilk test, and one-way analysis of the variance was performed using the Statistica software (version 6, StatSoft, Inc., 2001, Palo Alto, Santa Clara, CA, USA). Post-hoc comparison for detecting statistic significant differences was made with the Tukey test, setting the significance value a *p* < 0.05.

#### **3. Results**

#### *3.1. Microalgae Growth*/*Kinetic Parameters*

Knowledge regarding the growth pattern of microalgae and its parameters is not only interesting for the quantitative production of both biomass and metabolites, but also for increasing our comprehension of both the synthesis regulation and degradation dynamics of photosynthetic products. In this sense, Figure 3 depicts representative growth curves for the microalgae *Scenedesmus obliquus* through phototrophic and heterotrophic metabolic pathways, whereas the growth parameters are presented in Table 1. Hence, both culture conditions followed an exponential growth from the beginning without lag phase, at the specific growth rates of 0.023 and 0.024 h<sup>−</sup>1, generation intervals of 30.13 and 28.8 h, and finally reaching a stationary phase at 144 h and at 96 h in photosynthetic and heterotrophic cultures, respectively. The highest biomass accumulation was achieved under the photosynthetic cultivation (2650 mg/L) which was only 2% higher than the high biomass concentration of heterotrophic condition (2600 mg/L). The greatest impact of the type of cultivation was on the biomass productivity, where the highest value was obtained under heterotrophy (19.75 mg/L h), which is a consequence of the low residence time (120 h) reached in this condition, when compared with the phototrophic culture (216 h) that resulted in a productivity of 10.87 mg/L h. Regardless, at very long incubation times during the phototrophic growth, it is impossible to ensure that no nutrient deprivation occurs. However, assuming this possibility, we extended the study to analyze the effects of excess of light on pigment composition.




RT (h) 216 ± 0.00 120 ± 0.00 GT (h) 30.13 ± 0.60 28.8 ± 0.49

**Figure 3.** Growth curves in phototrophic and heterotrophic culture regimes of *Scenedesmus obliquus*.

#### *3.2. Pigment Profile During Phototrophic Growth*

The characteristics of the chromatographic and mass spectrometric data for the different pigments analyzed in the present study are shown in Table S1. *Scenedesmus* exhibits the typical carotenoid profile of the Chlorophyta taxon, which mainly contains lutein, β-carotene, and relative amounts of minor xanthophylls, such as violaxanthin and neoxanthin [47]. The chlorophyll fraction has been generally described as comprised by chlorophyll *a* and chlorophyll *b*, a feature of this taxonomic group of green algae. However, our detailed analysis reveals the presence of intermediary chlorophyll metabolites within the chlorophyll profile. Figure 4 displays the structures of the chlorophyll derivatives present in the profile of *Scenedesmus obliquus*.

**Figure 4.** Chlorophyll structures identified in *Scenedesmus obliquus*: (**a**) chlorophyll (R<sup>3</sup> is phytol, C20H40) and chlorophyllide (R3 is H) structure, R<sup>1</sup> is CH3 for chlorophyll *a*, and CHO for chlorophyll *b*, R2 is H for chlorophyll (*a* and *b*) and OH for 132-hydroxy-chlorophyll (*a* and *b*); (**b**) pheophytin (R3 is phytol, C20H40) and pheophorbide (R<sup>3</sup> is H) structure, R<sup>2</sup> is H for pheophorbide, and OH for 132-hydroxy-pheophorbide; (**c**) 152-hydroxy-lactone chlorophyll *b* structure.

It was observed that the total amount of carotenoids (Table 2) increased with the radiation time until the microalgae reached the stationary phase (144 h), to subsequently present a steady state until the end of the phase. In *Scenedesmus*, this behavior is due to the response of the main carotenoids, lutein and β-carotene, to the continuous illumination. As it has been stated [39], the same carotenoid kind may develop different roles in the cell depending on its location. According to the observed data (Table 2), lutein and β-carotene behave as primary photosynthetic pigments in *Scenedesmus*, although β-carotene could perform secondary activities in other chlorophytes, and even transported into oil droplets where they accumulate under stress conditions [17]. Regardless, lutein and β-carotene are photoprotective pigments, minimizing the photoinhibition through additional roles as quenchers or scavengers [39]. However, the minor xanthophylls display a different behavior under continuous radiation in *Scenedesmus* cells. Neoxanthin, violaxanthin, luteoxanthin, and antheraxanthin increased their concentrations in the microalgae culture even after the stationary growth phase. Specifically, violaxanthin and antheraxanthin are involved in the so-called xanthophyll cycle, intimately related with the ability to dissipate the excess of absorbed light. During high light irradiance conditions, the de-epoxidation reaction of violaxanthin to produce antheraxanthin reduces the light-harvesting efficiency in the antenna [48]. Finally, although neoxanthin could be considered as a light harvesting pigment, it also develops a role as photoprotective compound, reacting towards reactive oxygen species and preventing cell damage [49].

**Table 2.** Evolution of the carotenoid profile during the phototrophic growth of *Scenedesmus obliquus* (mg/kg dw).


Neox: neoxanthin, Violax: violaxanthin, Luteox: luteoxanthin, Anterax: anteraxanthin, Total: total carotenoids. +, means presence but under the LOQ. (coefficient of variance < 10% in all cases).

In relation to the response of the chlorophyll fraction to the continuous irradiance (Table 3), it was observed that light exposure initially induces chlorophyll synthesis. Although this result was anticipated, prolonged irradiance times (which means an excess of light) result in a net degradation of the chlorophyll fraction. The detailed analysis of the chlorophyll profile during the phototrophic growth of *Scenedesmus* shows chlorophyll *a* and *b* as the main pigments, but the accumulation of the intermediary metabolites pheophytin and pheophorbide *a* was also concomitant. Pheophytin *a* (Figure 4b) is produced by the substitution of the central Mg2<sup>+</sup> ion by hydrogens, while pheophorbide *a* (Figure 4a) involves an additional dephytylation step at the C17<sup>3</sup> position. However, the outstanding results are the production of a heterogeneous profile of oxidized chlorophylls. Among them, the 132-hydroxy-compounds stand out, which result from the oxidation at the C132 carbon atom (R2 is OH in Figure 4) in chlorophyll of the *a* and *b* series, and in pheophorbide *a*. Furthermore, the formation of a lactone functional group is considered a further step in the oxidative level of the original chlorophyll structure [31]. In this sense, it was very surprising to find 151-hydroxy-lactone chlorophyll *b* (Figure 4c) in the chlorophyll profile of *Scenedesmus* under radiation conditions.


**Table 3.** Evolution of the chlorophyll profile from series *a* during the phototrophic growth of *Scenedesmus obliquus* (mg/kg dw).

Pheo: pheophorbide, OH-Pheo *a*: 132-hydroxy-chlorophyll *a*, Chl: chlorophyll, OH-Chl *a*: 132-hydroxy-chlorophyll *a*, Phy: pheophytin. (CV < 10% in all cases).

It is noteworthy to highlight the different behavior of chlorophyll derivatives from *a* series (Figure 4a, CH3 at C7) from that observed for the *b* series (Figure 4a, CHO at C7). The chlorophyll compounds from *a* series, except pheophytin *a*, were biosynthesized until the maximum growth stage was reached (144 h), and subsequently a progressive degradation initiated. However, metabolites from chlorophyll *b* series (chlorophyll *b*, 132-hydroxy-chlorophyll *b* and 151-hydroxy-lactone chlorophyll *b*) showed their maximum concentrations between 48 and 72 h of illumination, around half of the period required to reach the residence time. After the apex peak, the metabolites of the chlorophyll *b* initiated a net degradation. The interconversion of chlorophyll *a* and *b*, through the denominated chlorophyll cycle (Figure 1 [50]), is an essential mechanism in photosynthetic organisms, as they can adapt their photosynthetic apparatus to the irradiance level. At high levels of illumination, the organism reduces the antenna complexes to avoid excess of photons, so that the production of reactive oxygen species (ROS) is minimized. As antenna complexes are rich in chlorophyll b compounds, at high irradiances the relative amounts of chlorophyll *b* decreased. On the contrary, at low irradiance (shadow) conditions, the organism rises the antenna complexes to capture as many photons as possible, which results in an increase of the chlorophyll compounds of the *b* series. Consequently, microalgae modify the ratio of chlorophyll *a*/*b* according to the irradiance levels [51]. As it can be observed in Table 4, at the initial 72 h of growth the ratio of *a*/*b* series decreased in *Scenedesmus*, as the biosynthesis rate of chlorophyll *b* was higher than that for the chlorophyll *a*. However, when the quantity of light was excessive for the culture (after 72 h of continuous illumination), the antenna complexes decreased, the concentration of chlorophyll *b* diminished and, consequently, the *a*/*b* ratio increased. Similar changes in the *a*/*b* ratio have been observed for *Chlorella* and *Dunalliela* [52]. At high irradiances, the energy received by chlorophyll *a* molecule is higher than its capacity to transfer it towards the photosynthetic electron transport chain, and chlorophyll *a* switches to the triplet excited stage [39]. Next, overexcited chlorophyll *a* molecule is quenched by molecular oxygen yielding ROS. As we can observe in Tables 3 and 4, the interconversion between chlorophyll *a* and *b* contents is the preferred mechanism of *Scenedesmus* cells to avoid the formation of ROS at high irradiances.

Nevertheless, once the maximum concentrations for chlorophyll *a* (144 h, Table 3) and *b* (72 h in Table 4) were reached, the net degradation of chlorophyll compounds was not exhaustive. Otherwise, *Scenedesmus* cells reached a steady state for the chlorophyll content (around 6200 mg/kg dw. for chlorophyll *a*, 1900 mg/kg dw. for chlorophyll *b*) until the end of the controlled period. It seems that once the top biosynthetic capabilities were accomplished, the microalgae found an 'ideal' chlorophyll content, which allows an equilibrated photosynthetic performance, that is, a productive one but not harmful, at least at the irradiance assayed for *Scenedesmus*. As it has been previously stated, photoacclimation is complete only when a balanced growth condition is achieved [53]. However, this is accurate when the chlorophyll content is determined as a whole value. As we have shown, a detailed

study of the complete chlorophyll profile allows to observe different biosynthetic capabilities with some chlorophyll metabolites reaching steady state earlier, precisely to fit with the photoacclimation at 144 h.


**Table 4.** Evolution of the chlorophyll profile from series *b* during the phototrophic growth of *Scenedesmus obliquus* (mg/kg dw).

OH-Lact.-Chl *b*: 151-hydroxy-chlorophyll *b*; OH-Chl *b*: 132-hydroxy-chlorophyll *b*; Chl *b*: chlorophyll *b*. (CV < 10% in all cases).

Pheophorbide *a* and pheophytin *a* are currently considered the metabolites of the chlorophyll degradation pathway (Figure 1). In fact, pheophorbide, pheophytin, and pyropheophorbide have been associated with the chlorophyll degradation in cyanobacteria in sedimentary surfaces [54] and chlorophyll senescence in marine environments [34]. Recently, the gene responsible of the formation of pheophytin (SGR) a has been identified in *Chlamidomonas reinharditii* [24]. However, while the kinetics of production and degradation of pheophorbide *a* is parallel to the chlorophyll *a*, the profile of metabolism of pheophytin *a* seems to progress in a different fashion and not correlated with the metabolism of chlorophyll *a*. The maximum concentration of pheophytin *a* was observed at 24 h, while its progressive decay through the continuous illumination period made the interpretation of the results in base to its implication in the chlorophyll degradation pathway challenging.

In addition, the HPLC-ESI/APCI-hrTOF-MS analyses of the chlorophyll fraction revealed the existence of a specific chlorophyll oxidative metabolism (Tables 2 and 3) during the *Scenedesmus* phototrophic cultivation. As stated before (Figure 2), hydroxylation at C132 is the first step in the oxidative pathway of chlorophylls. Hence, 132-hydroxy-chlorophyll *a* and *b* increased their concentrations in the cell with the continuous illumination for the initial 72 h period, and afterwards a progressive degradation was observed. In any case, it is important to highlight that the maximum of 132-hydroxy-chlorophyll *a* cellular content was not concurrent with the maximum concentration of chlorophyll *a*, which pointed towards a specific linking reaction between both compounds instead of an unspecific process. Noteworthy, 132-hydroxy-pheophorbide *a* was also produced around 3 days of illumination, once pheophorbide *a* is biosynthesized in the microalgae. A further oxidative process is the generation of the lactone rearrangement at the C151 position (Figure 4). During the phototrophic growth of *Scenedesmus* a progressive accumulation of 151-hydroxy-lactone chlorophyll *b* is observed, reaching the maximum value after 72 h (Table 3). In our experimental conditions it seems 72 h is the timeframe for *Scenedesmus* to reach the 'buffer capacity' (from the point of view of chlorophylls) and manage both the excess of energy and, consequently, the potential accumulation of ROS. Afterwards, profound physiological changes are required to avoid oxidative stress, as the commented restructuration of antenna complexes.

Moreover, no 151-hydroxy-lactone chlorophyll *a* formation was detected in any moment of the phototrophic growth, although chlorophyll *a* is the main chlorophyll pigment in the chlorophyll profile of *Scenedesmus*. In fact, although a chlorophyll metabolite with this functional group is not easy to distinguish [1], it is the 151-hydroxy-lactone chlorophyll *a* catabolite observed (if any) in photosynthetic organisms, but not the 151-hydroxy-lactone chlorophyll *b* catabolic product [30]. Indeed, both proportionally and in absolute concentration, the total biosynthesized chlorophyll oxidative compounds of the *b* series overcame those of the *a* series. To the best of our knowledge, this is the first time to describe such phenomenon. The biochemical origin of the oxidized chlorophyll metabolites is still under discussion. In higher plants, different enzymatic systems have been assumed as responsible for such oxidation (lipoxygenase and/or peroxidase) [29–31]. However, although different oxidative mechanisms have been observed in microalgae (peroxidase, superoxide dismutase, polyphenol oxidase, glutathione peroxidase, etc.) [55,56], none of them have been correlated with the chlorophyll metabolism so far. Two possible hypotheses can explain the higher rate of oxidation of chlorophyll *b* catabolism. Thus, the preferential accumulation of chlorophyll *b* catabolites could be due to an unknown chlorophyll *b* affinity by the pool of oxidative enzymes pool, or this singularity could be caused by the different localization of both chlorophyll series in the photosynthetic apparatus. Further research is required to unravel the exact mechanism.

#### *3.3. Pigment Evolution During Heterotrophic Growth*

As it can be seen in Table 5, heterotrophy means carotenoid degradation for *Scenedesmus obliquus* in our experimental conditions, although at very different rates depending on the carotenoid sort. The initial 24 h in darkness produces a significant carotenoid degradation except for neoxanthin, while the concentration of β-carotene and violaxanthin decreased by half. This decrease was extended in a lower degree for lutein. From 24 to 48 h of growth in the darkness, carotenoids were highly stable, the next 24 h interlude (72 h) only being a significant stage for the stability of neoxanthin and violaxanthin. Extending the heterotrophic culture of *Scenedesmus obliquus* far from 96 h implied a carotenoid degradation of at least 85%. In fact, carotenoid production in heterotrophic cultivation requires additional oxidative stress: high salt concentration, high light, etc. [13]. In any case, it is important to highlight the different stability of carotenoids in heterotrophic conditions, to face the future biotechnological strategies aimed to enhance the production of carotenoids.


**Table 5.** Evolution of the chlorophyll and carotenoid profile during the heterotrophic growth of *Scenedesmus obliquus* (mg/kg dw).

Chld *a*: chlorophyllide *a*; Pheo *a*: pheophorbide *a*; Chl *a*, Chl *b*: chlorophyll *a*, chlorophyll *b*; OH-chl *a*: 132-hydroxy-chlorophyll *a*; Phy *a*: pheophytin *a*; carot: carotenoids. (CV < 10% in all cases).

On the contrary, it was remarkable to observe the behavior of the chlorophyll fraction at heterotrophic culture conditions. During the initial 48 h of growth, the total amount of chlorophylls was constant and after that time interval, the chlorophyll profile initiated a phase of net degradation with increased rate at the end of the controlled period. Such modification in the chlorophyll metabolism is coincident with an increase of biomass. The initial steady state of the chlorophyll content means that the biosynthetic and the degradative reactions are evolving at the same rate. Although the exact quantity is unknown, the half-life of a chlorophyll molecule is estimated around several hours [57]. This fact implies that during the steady state of chlorophylls in the initial 48 h of heterotrophic culture, biosynthetic and degradative reactions are running in *Scenedesmus* cells. Regarding the biosynthetic metabolism, as stated before green algae can synthesize chlorophylls in dark conditions. Consequently, during 48 h of heterotrophic cultivation of *Scenedesmus*, a continuous synthesis of chlorophylls took place, although at the same rate as the degradative reactions. The first assumption to consider is that under heterotrophic conditions, the cell does not invest energy in chlorophyll synthesis but focuses on the cell division and growth process with the available resources. In fact, it has been argued that glucose can inhibit the chlorophyll biosynthesis, by means of an inhibitory activity towards the precursor coprophorphyrin III [58]. On the contrary, some reports have shown a certain degree of chlorophyll retention during heterotrophic growth [59], as we have found for *Scenedesmus*. The exact physiological meaning of such energetic investment is unknown to date, although our results are an important starting point for future biotechnological applications aimed to enhance the chlorophyll production.

In addition, the detailed analysis of the chlorophyll profile during the heterotrophic growth of *Scenedesmus* shows accumulation of chlorophyll metabolites produced during the chlorophyll degradation, that mirror the masked reactions that were under progress. Pheophorbide, chlorophyllide, and pheophytin are intermediary catabolites during the chlorophyll degradation pathway. Table 5 shows a significant increment of pheophorbide and chlorophyllide *a* at the end of the controlled period, concomitant with the main degradation of chlorophylls. However, pheophytin levels continuously decreased through the cycle, showing no parallelism with the chlorophyll breakdown. The results suggest that the operating pathways during the heterotrophic cultivation of *Scenedesmus* are better related with the chlorophyllase (CHL) pathway (Figure 1) than with pheophytinase one (PPH). Homologous PPH proteins have been found through BLASTP (Basic Local Alignment Search Tool for Proteins) searches in green algae but not in cyanobacteria, and it has been proposed that PPHs are also likely to be operative in the green algae [60], although no functional analysis has been developed so far. Although such data are not available, PPH seems to not be responsible for the chlorophyll degradation during heterotrophic conditions, at least during the culture conditions assayed in *Scenedesmus*.

To the best of our knowledge, accumulation of 132-hydroxy-chlorophylls is described for the first time in this study during the heterotrophic culture of green microalga, although no 151-hydroxy-lactone derivatives were detected. Interestingly, the heterotrophic strategy only induced oxidation in chlorophyll *a* molecules and no oxidized chlorophyll *b* compounds were detected in any moment of the cycle. 132-hydroxy-chlorophyll *a* production, observed during the initial 48 h of growth in the darkness could involve a role during the chlorophyll turnover, although the main synthesis is accomplished with the net degradation of chlorophylls at the end of the cultivation period. As stated before, the exact role of oxidized chlorophylls in phytoplankton is unclear, but associated with defense, grazing, senescence, or even death cell [33–38]. Our results show both production and degradation kinetics during the heterotrophic culture of *Scenedesmus*, with more than a plausible role during the chlorophyll degradation. Consequently, the results obtained in Table 5 open a door for future research, with a focus on the biochemical mechanisms involved in the chlorophyll oxidative metabolism during the heterotrophic cultivation of green microalgae.

#### **4. Conclusions**

As stated in the introduction, the improvement of pigment production with biotechnological parameters requires a deep understanding of the reactions that take place during the different culture approaches. In this sense, it is essential to know the physiological strategies that green microalgae develop to become acclimatized to the environmental conditions. In addition to the technological data, our study introduces a specific and different chlorophyll oxidative metabolism during phototrophic and heterotrophic cultivation, which agrees with the measurement of oxidized chlorophyll metabolites in natural phytoplankton environment [35,36]. Future assays in controlled bioreactors are required to unravel the precise implication of such oxidative metabolism.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2076-3921/8/12/600/s1, Table S1: Photosynthetic pigments identified by HPLC-PDA-ESI/APCI(+)-Q-TOF in the study.

**Author Contributions:** Conceptualization, L.Q.Z., E.J.L. and M.R.; methodology, M.M.M. and M.R.; software, A.P.-G.; validation, M.M.M.; formal analysis, M.M.M. and M.R.; investigation, M.M.M.; resources, L.Q.Z. and E.J.L.; data curation, A.P.-G. and M.R.; writing—original draft preparation, M.R.; writing—review and editing, L.Q.Z., E.J.L., A.P.-G. and M.R.; supervision and funding acquisition, M.R.

**Funding:** This work was supported by the Ministerio de Ciencia, Investigación y Universidades, Agencia Estatal de Investigación y Fondo Europeo de Desarrollo Regional (FEDER), grant number RTI2018-095415-B-I00. MM was supported with a fellowship from the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code 001 and the Brazilian Funding Agency FAPERGS (Fundação de Amparo a pesquisa do estado do Rio Grande do Sul). The APC was partially funded by CSIC.

**Acknowledgments:** The authors would like to thank to Sergio Alcañiz for his technical assistance.

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

#### **References**


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

## *Article* **In-Vitro Antioxidant Properties of Lipophilic Antioxidant Compounds from 3 Brown Seaweed**

#### **Gaurav Rajauria**

School of Agriculture and Food Science, University College Dublin, Lyons Research Farm, Celbridge, Co. Kildare W23 ENY2, Ireland; gaurav.rajauria@ucd.ie; Tel.: +353-1-601-2167

Received: 21 October 2019; Accepted: 26 November 2019; Published: 28 November 2019

**Abstract:** Lipophilic compounds of seaweed have been linked to their potential bioactivity. Low polarity solvents such as chloroform, diethyl ether, *n*-hexane and their various combinations were used to extract the lipophilic antioxidants from brown seaweed namely *Himanthalia elongata, Laminaria saccharina* and *Laminaria digitata*. An equal-volume mixture of chloroform, diethyl ether and *n*-hexane (Mix 4) gave the highest total phenol (52.7 ± 1.93 to 180.2 ± 1.84 mg gallic acid equivalents/g), flavonoid (31.9 ± 2.65 to 131.3 ± 4.51 mg quercetin equivalents/g), carotenoid (2.19 ± 1.37 to 3.15 ± 0.91 μg/g) and chlorophyll content (2.88 ± 1.08 to 3.86 ± 1.22 μg/g) in the tested seaweeds. The extracts were screened for their potential antioxidant capacity and the extracts obtained from the selected solvents system exhibited the highest radical scavenging capacity against 2,2 -diphenly-1-picrylhydrazyl radical (EC50 98.3 ± 2.78 to 298.8 ± 5.81 mg/L) and metal ions (EC50 228.6 ± 3.51 to 532.4 ± 6.03 mg/L). Similarly, the same extract showed the highest ferric reducing antioxidant power (8.3 ± 0.23 to 26.3 ± 0.30 mg trolox equivalents/g) in all the seaweeds. Rapid characterization of the active extracts by liquid chromatography coupled with photodiode array detector and electrospray ionization tandem mass spectrometry (LC-PDA–ESI-MS/MS) identified cyanidin-3-*O*-glucoside, fucoxanthin, violaxanthin, β-carotene, chlorophyll *a* derivatives and chlorophyll *b* derivatives in the tested seaweed. The study demonstrated the use of tested brown seaweed as potential species to be considered for future applications in medicine, cosmetics and as nutritional food supplement.

**Keywords:** lipophilic antioxidant; solvent blending; macroalgae; LC-ESI-MS/MS; carotenoid pigment; anthocyanin; chlorophyll derivative

#### **1. Introduction**

The concepts of nutrition are changing rapidly as consumers all over the world have become more cautious regarding nutritionally healthier food and its ingredients. Recently, a great interest in using natural plant-derived bioactive compounds in foods, cosmetics and pharmaceuticals has arisen, due to their nutritional and therapeutic effects [1,2]. The epidemiological and observational literatures suggest that free radicals play an important role in affecting human health by causing cancers or age associated neurodegenerative diseases. However, antioxidant-rich foods have shown their relevance in the prevention of these diseases by mitigating the harmful free radicals or reactive oxygen species (ROS) [3]. Chemical compounds such as butylated hydroxytoluene (BHT; E-321), butylhydroxyanisole (BHA; E-320) and ascorbic acid (E-300) are commonly used as synthetic antioxidants in food products to improve the product quality and shelf life. However, due to possible toxicity of synthetic antioxidants as well as consumer preference towards natural substances, natural antioxidants are considered safe and more acceptable for use as ingredients in functional foods, nutraceuticals and cosmetics [4,5]. Among the most studied classes of natural antioxidants, phenolic compounds and carotenoid pigments are widely distributed in the plant kingdom and have received much attention for their high antioxidant activity [6]. Although these functional ingredients are not restricted to terrestrial resources, plants in

general and seaweeds (marine plants) in particular, are good sources of natural antioxidants. Seaweed grows in extreme environmental conditions thus producing a variety of antioxidant compounds to counteract environmental stresses [7]. The most important naturally occurring seaweed substances showing antioxidant properties are polyphenols, phlorotannins, flavonoids, carotenoids, fatty acids, polysaccharides and amino acids, which in varying proportion and quantities, are reported in different seaweed species [8–10]. A variety of in vitro studies have shown that lipophilic compounds such as carotenoid pigments and some polyphenols and flavonoids exhibit strong antioxidant activity [11–14]. These compounds are capable of acting as primary antioxidants by reacting with free radical species or could act as secondary antioxidants (metal chelator) by blocking the generation of hypervalent metal forms [15]. Such antioxidant activities of carotenoids and polyphenols may protect cells from ROS-induced cellular damage, thereby reducing the risk of diseases associated with oxidative stress [16].

Multiple compounds from hundreds of algal species have been studied up until now and a range of compounds possessing antioxidant properties have been discovered. Among these compounds, some compounds are of polar or hydrophilic nature (e.g., phlorotannins), some are semi-polar (e.g., phenolic acids and simple flavonoids), some and others are non-polar or lipophilic in nature (e.g., carotenoids, fatty acids). They may also exist as complexes with sugar, proteins and other cell membrane components; which make them quite insoluble and a selective solvent system is required to solubilize and extract them [17]. Extractability of bioactive compounds is associated with the polarity of solvents (polar/semi-non-polar) used, as well as their complexity with other constituents. Finding a solvent system suitable for the extraction of all classes or a specific class of antioxidant is restricted by the chemical nature of these bioactive compounds. These bioactives are present in matrices as a complex mixture of compounds that provide a cocktail of many active components present in the free, esterified, glycosylated and bound states as conjugates with other components that lead to the formation of insoluble complexes. The solubility of these compounds is administered by the nature of raw material, degree of polymerization and the polarity of solvent used [1]. Therefore, the extraction of these active ingredients from seaweed matrices is the key step to utilizing them for pharmaceutical, cosmeceutical, and foods as well as nutraceutical preparations. Thus, to obtain extracts enriched in lipophilic compounds, it is of critical importance to select efficient extraction solvent systems to improve their extractability and to maintain stability. Additionally, extraction solvent can have a significant effect on the performance of antioxidant reaction mechanisms which can change the chemical behavior of antioxidant compounds [18]. Therefore, there is no uniform or completely satisfactory procedure that is suitable for extraction of all compounds or a specific class of compounds from plant materials [17]. Thus, the objective of the present study was to select the appropriate solvent system that is capable of extracting lipophilic compounds from Irish brown seaweeds and to evaluate the antioxidant capacity and phytochemical constituents of those extracts. Seaweeds were extracted with semi/non-polar solvents and their mixtures, in order to get the lipophilic antioxidant compounds. The crude extracts were screened for total polyphenol, flavonoid, chlorophyll and carotenoid content along with antioxidant reducing power and potential radical scavenging capacity against 2,2 -diphenyl-1-picrylhydrazyl (DPPH) radicals and metal-ions. The identification and characterization of antioxidant compounds were carried out by using liquid chromatography coupled with electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS) and UV-visible spectroscopy.

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

#### *2.1. Chemicals, Solvents and Standards*

Folin-Ciocalteu's phenol reagent, 2,2 -diphenyl-1-picrylhydrazyl (DPPH), 3-(2-pyridyl)-5,6 diphenyl-1,2,4-triazine-4 ,4-disulphonic acid monosodium salt (ferrozine), 2,4,6-tripyridyl-s-triazine (TPTZ) and 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox) were purchased from Sigma-Aldrich Chemical Co. (Steinheim, Germany). For LC-MS analysis, solvents such as water, methanol and acetonitrile were chromatography grade which was purchased from Fisher Chemicals

(Thermo Fisher Scientific Inc., Dublin, Ireland). Authentic standards including L-ascorbic acid, gallic acid, quercetin, cyanidin 3-glucoside, fucoxanthin and violaxanthin were purchased from Sigma-Aldrich Chemical Co. (Arklow, Co. Wicklow, Ireland). All other chemicals used in the study were analytical grade and purchased from Sigma-Aldrich Chemical Co. (Ireland).

#### *2.2. Seaweed Materials and Extraction Procedure*

Irish brown seaweeds (Figure 1) used in this study namely, *Laminaria digitata*, *Laminaria saccharina* and *Himanthalia elongata* (*Phaeophyta*) were purchased from Quality Sea Veg., Co Donegal, Ireland. Among the tested seaweeds, *L. digitata* and *L. saccharina* (also known as *Saccharina latissima*) are large conspicuous dark brown and yellow brown kelp, commonly found down to a maximum depth of 20 m and 30 m in clear waters respectively. The stipes of *L. digitata* are smooth, flexible and oval in cross section while *L. saccharina* has a long undivided frond with a distinct bullations and a distinctive frilly undulating margin. Both are usually found attached to bedrock or other suitable hard substrata in the low water in Intertidal pools and occasionally in the shallow subtidal zones. *H. elongata* is a light yellow-brown fucoid species which has long, narrow, strap-like branched fronds with basal mushroom-like buttons. It is found attached to rocks or hard substrata on moderately semi-wave-exposed shore. Seaweed species were harvested and collected in winter (January/February), washed thoroughly with fresh water to remove epiphytes, eliminate salt, sand or shells and stored at −20 ◦C until analysis.

**Figure 1.** Images of brown Irish seaweeds studied.

Extraction of lipophilic components from seaweed was carried out by crushing the fresh sample with liquid nitrogen followed by extraction with semi/non-polar organic solvents such as chloroform, diethyl ether, *n*-hexane and thereof mixtures according to the method described earlier [19,20]. The mixtures of solvents used were; Mix 1 (*n*-hexane and diethyl ether), Mix 2 (*n*-hexane and chloroform), Mix 3 (diethyl ether and chloroform) and Mix 4 (*n*-hexane, diethyl ether and chloroform). All the solvents were mixed either 1:1 (*v*/*v*) or 1:1:1 ratio (*v*/*v*/*v*) depending upon the mixtures, and dielectric constant (ε) of individual solvent as well as their mixture were taken into account. The extracted samples were filtered with Whatman #1 filter paper and centrifuged at 9168× *g* (Sigma 2–16PK, SartoriusAG, Gottingen, Germany) for 15 min. The resulting supernatant was evaporated to dryness, and the dried lipophilic extract was dissolved in HPLC (high performance liquid chromatography) grade methanol for further analysis. The whole extraction procedure was carried out under dark conditions to minimize the possibility of oxidation/degradation of antioxidant compounds by light.

#### *2.3. Phytochemical Constituent Analysis*

Crude lipophilic extracts of seaweed were screened for total phenolic content (TPC), total flavonoid content (TFC), total chlorophyll content (TChC) and total carotenoid content (TCC). TPC was determined according to Ganesan et al. [9]. Samples were read at 720 nm and the results were expressed as mg gallic acid equivalents (GAE)/g dry weight (dw) extract. TFC and TTC were determined according to Liu et al. [21]. Samples were read at 510 nm and 500 nm, and the results were expressed as mg quercetin equivalents (QE)/g extract (dw) and mg (+)-catechin equivalents (ChE)/g extract (dw), respectively. TChC and TCC were determined according to Arnon (1949) and Kirk and Allen (1965) respectively. For chlorophyll, the samples were read at 645 nm and 663 nm, and total content was calculated by using Equation (1), while for total carotenoids, the absorbance of the same chlorophyll samples was recorded at 480 nm and content was calculated by using Equation (2).

$$\text{TChC (\mu g/g; dw)} = 20.2 \times (A\_{645}) + 8.02 \times (A\_{663}) \tag{1}$$

where *A* = Absorbance at respective wavelength

$$\text{TCC (\mu g/g; d\nu)} = A\_{480} + (0.114 \times A\_{663}) - (0.638 \times A\_{645}) \tag{2}$$

where, *A* = Absorbance at respective wavelength

#### *2.4. Antioxidant Capacity Analysis*

#### 2.4.1. DPPH Radical Scavenging Capacity Assay

This assay was carried out according to the method reported earlier [22]. Ascorbic acid was used as a standard and the absorbance of the standard or samples was recorded at 517 nm using a 96-well plate reader. The ability of samples to scavenge the DPPH radical was calculated using Equation (3):

$$\text{DPPH radical scavenging capacity (\%)} = \left[1 - (A\_{\text{sample}} - A\_{\text{sample blank}}) / A\_{\text{control}}\right] \times 100 \tag{3}$$

where, *A* = Absorbance of sample/sample blank or control

#### 2.4.2. Ferric Reducing Antioxidant Power (FRAP) Assay

Total antioxidant reducing power of various extracts of seaweed was measured using modified FRAP assay [23]. Trolox was used as a standard and the absorbance of the standard or samples was recorded at 593 nm, and the results were expressed as mg trolox equivalents (TE)/g extract (dw).

#### 2.4.3. Metal Ion-Chelating Ability Assay

The chelating ability of metal ion (ferrous ion) by seaweed extracts was estimated using the original method of Decker and Welch [24] with minor modifications. This assay is based upon the formation of blue colored ferrous ion-ferrozine complex which has a maximum absorbance at 562 nm. EDTA (ethylenediaminetetraacetic acid) was used as a standard compound. The percentage of inhibition of ferrozine-Fe2<sup>+</sup> complex formation was calculated using Equation (3).

#### *2.5. Characterization of Lipophilic Compounds using Liquid Chromatography Mass Spectrometry (LC–MS)*

Antioxidant compounds in the lipophilic extracts were analyzed on 6410 Triple Quadrupole LC/MS, fitted with Agilent 1200 series LC, G1315B variable-wavelength photodiode array (PDA) detector and MassHunter Workstation software (version B.04.00, Agilent Technologies, Santa Clara, CA, USA). The separation was performed at 25 ◦C using an Atlantis C-18 (250 × 4.6 mm, 5 μm particle size) column fitted with a suitable C-18 (4.0 × 3.0 mm) guard cartridge (Waters, Dublin, Ireland). The mobile phase consisting of ternary solvents of acetonitrile/methanol/water (75:15:10, *v*/*v*/*v*) containing 1.0 g/L ammonium acetate, eluted at 1.0 mL/min for 25 min, was adopted from Sugawara [25]. The injection volume of 10 μL was kept constant for samples and standard compounds. UV-vis absorption of the selected extracts was recorded from 190 to 600 nm using LC-PDA detector and the λmax (absorption maxima) of each peak was noted. Peaks assignments were made by comparing the UV/visible spectra of analytes to standard compounds, and available literature. Mass spectral data were recorded on

positive ionization mode using electrospray ionization (ESI) interface with 3.5 kV capillary voltage, 120 V fragmentor voltage and 10 eV collision energy in the mass range of *m*/*z* 100–1000. Nitrogen gas was used as the nebulizer and drying gas with 50 psi pressure, 10 L/min flow rate, 350 ◦C drying temperature and 35 nA capillary current. The identification of the peaks was carried out using mass spectral data of standard compounds where possible. Identification of remaining peaks was based on UV-visible spectral (λmax) characteristics and the results were compared with the literature when no standards were available.

#### *2.6. Statistical Analysis*

Statistical analyses were carried out using STATGRAPHICS Centurion XV software (version XV, Statgraphics Technologies, Inc., The Plains, VA, USA). All the experiments were carried out in triplicate and repeated twice. Results are expressed as mean ± standard deviation. Statistical differences between antioxidant activities or phytochemical content of extracts were determined using Analysis of Variance (ANOVA) followed by Least Significant Difference (LSD) testing. Differences were considered statistically significant when *p* < 0.05.

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

#### *3.1. Phytochemical Content in Lipophilic Extracts*

It is widely accepted that bioactive compounds can be classified by their solubility into hydrophilic and lipophilic compounds. Similar to hydrophilic compounds, lipophilic compounds also play an important role in a wide spectrum of biochemical and physiological processes [11]. These lipophilic compounds can be extracted with semi/non-polar solvents in plants wherein polarity of the solvents play a significant role in the resulting yield, extractability and biological activity of bioactive compounds. In this study, various organic solvents and their combinations with varying dielectric constant were used to extract lipophilic compounds from 3 brown seaweed.

Results from Table 1 have shown a considerable variation in the extraction yield among the extracts recovered from various low polarity solvents and their mixtures. The extraction yield varied from 0.05% to 0.20% among all the tested seaweeds. The extracts recovered from *n*-hexane and Mix 1 solvents exhibited significantly (*p* > 0.05) the highest and the lowest extraction yield respectively. It is reported that low polarity (semi/non-polar) solvents generally give less extraction yield as compared to polar solvents [26] which is in agreement with the results obtained in this study.


**Table 1.** Extraction yield and phytochemical content of lipophilic extracts of brown seaweed obtained from various organic solvents and their mixtures (semi/non-polar solvents).


**Table 1.** *Cont.*

Values are expressed as mean ± standard deviation (SD). Values within a species with different letters (a–g) in columns are significantly different (*p* < 0.05), n = 6. \* Values among the three species with different letters (p–r) in columns, are significantly different (*p* < 0.05). Yield (%) is calculated in terms of g of dry extracts/100 g of fresh weight. TPC (total phenolic content) and TFC (total flavonoid content) are expressed as mg gallic acid equivalents/g (dw) and mg quercetin equivalents/g (dw), respectively. TCC (total carotenoid content) and TChC (total chlorophyll content) are reported in μg/g (dw). Mix 1: *n*-hexane and diethyl ether; Mix 2: *n*-hexane and chloroform; Mix 3: diethyl ether and chloroform; Mix 4: *n*-hexane, diethyl ether and chloroform. All the solvents were mixed in 1:1 or 1:1:1 (*v*/*v*/*v*) ratio.

Phytochemical content was majorly affected by the polarity of the extraction solvents as depicted in Table 1. In each of the tested seaweed, TPC from all the extracts was significantly different (*p* > 0.05) among the tested solvent systems. The extracts obtained from *n*-hexane exhibited the lowest TPC (varied from 7.7 ± 0.64 to 14.1 ± 0.79 mg GAE/g) while the extracts recovered from Mix 4 (*n*-hexane, diethyl ether and chloroform) solvents showed the highest TPC (ranging from 52.7 ± 1.93 to 180.2 ± 1.84 mg GAE/g), in all the species studied. The highest and significantly different (*p* < 0.05) amount of TPC was obtained in *H. elongata* followed by *L. saccharina* and *L. digitata* with the Mix 4 solvent system.

In the case of total flavonoid, the results showed that TFC in seaweeds varied considerably with the solvent polarity. The TFC of extracts obtained from different low polarity solvents and their mixtures ranged from 11.3 ± 2.5 to 131.3 ± 4.51 mg QE/g in *H. elongata*, 6.9 ± 0.88 to 56.3 ± 1.77 mg QE/g in *L. saccharina* and 4.4 ± 0.88 to 31.9 ± 2.65 mg QE/g in *L. digitata*. The extract from Mix 4 solvents exhibited the highest and significantly different (*p* < 0.05) TFC in *H. elongata* followed by *L. saccharina* and *L. digitata*. However, the extract obtained from *n*-hexane showed the lowest TFC in all the seaweed species (Table 1). There was no significant difference observed in TFC between the extract of Mix 1 (*n*-hexane: diethyl ether) and Mix 2 (*n*-hexane: chloroform) solvents within an individual seaweed species.

Pigments such as carotenoids play an important role in seaweed reproduction and are responsible for different colors. Fucoxanthin, a major pigment of brown seaweeds, is one of the most abundant carotenoids in nature and constitute 10% to total carotenoid production [27]. It is an orange-colored pigment, found along with chlorophyll pigment (*a* and *c*) and β-carotene, to give a brown or olive-green color to brown seaweed [28–30]. Numerous studies have shown that brown seaweed pigments such as fucoxanthin, violaxanthin and β-carotene have substantial applications in human health. These pigments have been explored for its potential bioactivities including antioxidant, anti-inflammatory, anticancer, anti-obese and antidiabetic property [14,31,32]. Table 1 shows that the spectrophotometric measurement of total carotenoids and total chlorophyll content in various extracts of 3 brown seaweed studied. The results revealed that the Mix 4 solvent system produced significantly higher TCC in *H. elongata* followed by *L. digitata* and *L. saccharina* whereas *n*-hexane extracts presented the lowest values among the tested seaweed. In the case of chlorophyll content, extracts from chloroform (instead of Mix 4 solvents) exhibited the highest TChC while extracts recovered from Mix 2 solvents showed the lowest TChC (*p* < 0.05), among the tested seaweeds and their extracts (Table 1). It was observed that the chloroform extract of *L. digitata* exhibited the highest TChC (*p* < 0.05) whereas *H. elongata* extract presented the lowest TChC.

Furthermore, upon analyzing TPC, TFC, TChC and TCC results against the polarity or dielectric constant of extraction solvents and their mixtures, an interesting relationship was observed. The results interpreted that the phytochemical content was primarily affected by the semi/non-polar extraction solvents. The dielectric constant of solvents and their mixtures was in the range of 5.0 to 2.0 with the following decreasing order: chloroform (5.0) > Mix 2 (4.7) > diethyl ether (4.3) > Mix 4 (3.8) > Mix 3 (3.5) > Mix 1 (3.2) > *n*-hexane (2.0). The dielectric constant of the mixed solvents is calculated on the basis of percentage (*v*/*v*) of each solvent used for the combinations. The dielectric constant of a solvent is an index of its polarity, and an increase in polarity shows a similar increase in the dielectric constant [10]. Mixing of solvents with different polarities is an approach to form a solvent system of optimum polarity to extract the various bioactive compounds. This approach is referred to as "solvent blending" or "co-solvency" and uses the dielectric constant as a guide to develop the co-solvent system [33]. The results indicated that the polarity/dielectric constant of Mix 4 solvent system (*n*-hexane, diethyl ether and chloroform) was more selective to the lipophilic phenolic compounds present in selected seaweeds than the other tested solvents and their mixtures. These findings are also in agreement with the report of Sahreen [34] wherein a range of polarity solvents gave different values of TPC, TFC and extraction yield. Furthermore, these findings also suggest that yield may not be a good indicator of phytochemical content of extracts based on the fact that phytochemical content was the lowest in the *n*-hexane extract, but had the highest extraction yield in all the studied seaweeds, which agrees with the previous reported results [35].

This study, as well as other previously reported publications [1,17,36], clearly illustrates that it is essential to systematically evaluate and optimize the extraction solvent composition for accurate and reproducible estimation of structurally diverse antioxidant compounds from different plants. In the present study, the highest recoveries of lipophilic antioxidants from seaweeds samples were obtained from Mix 4 solvents mixture using a solvent extraction technique.

#### *3.2. Antioxidant Capacity of Lipophilic Extracts*

The lipophilic extracts of all the three tested seaweed, obtained from various solvents and their mixtures, were screened for their potential antioxidant capacity using the stable DPPH radicals, FRAP reagent and by metal ion-chelating ability assay. The results of antioxidant capacity are illustrated in Figures 2 and 3. It was observed that all the seaweed exhibited a treatment effect and the scavenging of DPPH radicals by the seaweed extracts was dose-dependent. Results interpreted that EC50 values of all the extracts obtained from different solvents were significantly different (*p* < 0.05) in each seaweed species. The extracts from Mix 4 solvent exhibited the highest scavenging (lowest EC50 values) while the extracts from *n*-hexane depicted the lowest scavenging capacity (highest EC50 values) against DPPH radicals (Figure 2a). Among the tested seaweed, *H. elongata* showed the highest scavenging capacity (EC50 98.3 ± 2.78 μg/mL) followed by *L. saccharina* (EC50 222.4 ± 0.84 μg/mL) and *L. digitata* (EC50 298.8 ± 5.81 μg/mL). The scavenging capacity of the standard ascorbic acid (EC50 50.6 ± 0.79 μg/mL) was recorded higher than the seaweed extracts.

**Figure 2.** 2,2 -diphenyl-1-picrylhydrazyl (DPPH) radical scavenging capacity (**a**) and ferric reducing antioxidant power (**b**) of the Irish brown seaweeds extracts obtained from semi/non-polar organic solvents and thereof mixtures (1:1 or 1:1:1, *v*/*v*/*v*). [Mix 1: *n*-hexane and diethyl ether; Mix 2: *n*-hexane and chloroform; Mix 3: diethyl ether and chloroform; Mix 4: *n*-hexane, diethyl ether and chloroform]. [ : *H. elongata*; : *L. saccharina*; : *L. digitata*]. Data are expressed as mean ± SD (n = 6). Ferric Reducing Antioxidant Power (FRAP) values are expressed as mg Trolox equivalent (TE)/g extract (dry weight). Letters (a–g) on each bar are significantly different (*p* < 0.05) for various solvents, for each individual species. Letters (p–r) on bars at a specific solvent (Mix 4) are significantly different (*p* < 0.05) among the three species. Mix 1: *n*-hexane and diethyl ether; Mix 2: *n*-hexane and chloroform; Mix 3: diethyl ether and chloroform; Mix 4; *n*-hexane, diethyl ether and chloroform. All the solvents were mixed either 1:1 or 1:1:1 ratio (*v*/*v*).

Reducing power appears to be related to the degree of hydroxylation and the extent of conjugation in polyphenols. The ferric reducing antioxidant power in the various extracts of brown seaweeds was studied and the results are presented in Figure 2b. The reducing ability of all the extracts were significantly different within each species and ranged from 5.5 ± 0.20 to 26.3 ± 0.30 mg TE/g dw extract in *H. elongata*, 1.6 ± 0.06 to 10.9 ± 0.29 mg TE/g dw extract in *L. saccharina* and 1.7 ± 0.06 to 8.3 ± 0.23 mg TE/g dw extract in *L. digitata*. Of the tested extracts, Mix 4 solvent extracts (*n*-hexane, diethyl ether and chloroform) exhibited the highest and statistically different (*p* < 0.05) FRAP value in *H. elongata* followed by *L. saccharina* and *L. digitata*, while the extract obtained from the *n*-hexane showed the lowest reducing power in the tested seaweed. Jiménez-Escrig [37] reported that *Fucoid* species contained more reducing power than *Laminaria* species which is in agreement with the present results.

Ferrous ions are the most powerful pro-oxidants among various species of transition metals present in food systems. Dietary antioxidants (nutrients) having the metal chelating ability, form σ-bonds with metal ions and reduce the redox potential thereby stabilizing the oxidized form of the metal ions [38]. As seen in Figure 3, the formation of Fe2+-ferrozine complex is disrupted in the presence of various extracts from brown seaweeds. The absorption of this complex decreased linearly in a dose-dependent manner. All the extracts had a high level of metal ion chelating ability but were significantly lower as compared to the EDTA. Among all the tested solvents, the extract from Mix 4 solvents showed the highest chelating ability (*p* < 0.05) while extracts from *n*-hexane showed the lowest metal chelating ability at any tested concentration. In contrast to FRAP and DPPH scavenging activity, the metal chelating ability was recorded higher in *Laminaria* species compared to *H. elongata.* The percentage of the metal chelating ability of all the extracts at 1000 μg/mL concentration was found to be 22.7 to 57.8% in *H. elongata*, 48.9 to 81.9% in *L. digitata* and 52.8 to 82.3% in *L. saccharina,* while standard EDTA exhibited almost 100% chelating ability even at very low (125 μg/mL) concentration (Figure 3).

**Figure 3.** Metal-ion chelating ability of ethylenediaminetetraacetic acid (EDTA) standard and the Irish brown seaweeds extracts obtained from semi/non-polar solvents and mixtures (1:1 or 1:1:1, *v*/*v*/*v*) thereof (**a**) *H. elongata*; (**b**) *L. saccharina*; (**c**) *L. digitata*. [Mix 1: *n*-hexane and diethyl ether; Mix 2: *n*-hexane and chloroform; Mix 3: diethyl ether and chloroform; Mix 4: *n*-hexane, diethyl ether and chloroform].

Results also concluded that dielectric constant of extraction solvent has a significant role in antioxidant properties of extracted compounds. In this study, the results interpreted that the *n*-hexane (ε = 2.0) extracts exhibited the lowest while Mix 4 (ε = 3.8) extracts demonstrated the highest DPPH scavenging capacity, reducing power and metal ion chelating ability among the tested seaweed. The pattern of DPPH radical scavenging capacity shown by different solvent extracts were in the order of Mix 4 > diethyl ether > Mix 3 > Mix 1 > Mix 2 > chloroform > *n*-hexane (Figure 2a). While the arrangement of reducing power (FRAP) and metal chelating ability shown by different solvent extracts was as follows: Mix 4 > diethyl ether > Mix 3 > Mix 2 > Mix 1 > chloroform > *n*-hexane (Figures 2b and 3). The recovery of lowest antioxidant activity by *n*-hexane extracts is in agreement with the previous findings wherein, *Carissa opaca* fruit extract obtained from *n*-hexane showed the lowest antioxidant capacity as compared to other higher polarity solvents indicating that solvents polarity significantly affects the antioxidant capacity [34].

#### *3.3. Characterization of Lipophilic Antioxidant Compounds by LC-ESI-MS*/*MS*

The most active extract recovered from Mix 4 solvent system was used for the identification of lipophilic antioxidant compounds from all the seaweed studied. The selected extracts were characterized by liquid chromatography coupled with mass spectrometry using positive electrospray ionization mode (LC-ESI-MS). The identification of bioactive compounds in the extracts was carried out by comparing retention time, characteristic UV/visible (UV/vis) spectra and ESI-MS fragmentation data of each separated peak with that of the authentic standard. The UV/vis spectra provide characteristic chromophore information in pigments which cannot be obtained from MS data [39]. Therefore, chlorophyll and carotenoid pigments which could not be differentiated by only MS, were characterized by a combination of UV/vis spectral data with ESI-MS.

The selected extracts exhibited good separation by reverse phase (RP) HPLC and showed 12 distinct peaks in *H. elongata*, 13 peaks in *L. saccharina* and 12 peaks in *L. digitata.* The UV-vis absorption maxima (λmax) recorded by online HPLC-PDA analyses of each peak are shown in Table 2. The absorption maxima (λmax) of peaks recorded at 280 and 532 nm corresponds to anthocyanin pigments (flavonoid derivatives) in all the extracts [40]. Compounds with typical absorptions between 400 and 500 nm with λmax at around 425 nm corresponding to carotenoids. Absorption bands between 400–500 nm and 500–600 nm with λmax at 430 nm and 660 nm (chlorophyll *a* derivatives) and at 450 nm and 640 nm (chlorophyll *b* derivatives) representing chlorophylls [39,41]. The characteristic UV spectra revealed the presence of 8 carotenoid derivatives, 2 chlorophyll derivatives and 1 anthocyanin pigment while 1 peak was unidentified in *H. elongata*. Similarly, *L. saccharina* extract showed the presence of 9 carotenoid derivatives, 2 chlorophyll derivatives while 2 peaks were unidentified. UV spectral data from *L. digitata* extract exhibited the occurrence of 7 carotenoid derivatives, 2 chlorophyll derivatives and 1 anthocyanin pigment while 2 peaks were unidentified. The pattern of the absorption spectrum, as well as corresponding λmax, was similar for numerous compounds extracted among 3 seaweed species studied which indicates that the tested seaweed may have a few similar compound compositions. Due to the presence of the long chromophore of conjugated double bonds, carotenoid pigments can absorb UV and visible light and provide precious information about their structure [42]. Hence, characteristic UV-visible maxima (λmax) of each individual peak of HPLC-PDA profile of selected extracts were recorded. On the basis of UV-visible spectra, these pigments can be summarized under three categories i.e., tetrapyrroles (chlorophyll derivatives), carotenoids (carotene and xanthophyll derivatives) and flavonoids (anthocyanin derivatives). Generally, both chlorophylls and carotenoids show absorption maxima within the region of 400–500 nm but only chlorophyll derivatives show an additional band within the region of 550–700 nm, which differentiate them from carotenoid derivatives [43].


**Table 2.** UV/visible (λmax) and characteristic mass spectra (MS/MS) of the compounds isolated from lipophilic extracts of brown seaweeds *H. elongata*, *L. saccharina* and *L. digitata.*

Isolated lipophilic compounds from each extract were submitted for LC-ESI-MS/MS analysis. HPLC coupled to mass spectrometry with ESI proved extremely useful for peak assignment and gives a great deal of structural information and characterization of individual substances. Table 2 shows the typical ions resulting from mass spectra of lipophilic compounds obtained by LC-ESI-MS and MS/MS fragmentation. The ESI-MS spectra produced 5 protonated ([M + H]+) molecules at *m*/*z* 449 (peak 2), 536.9 (peak 4), 891.2 (peak 6), 601.5 (peak 7) and 659.6 (peak 11) in *H. elongata*; 3 protonated molecules at *m*/*z* 891.2 (peak 5), 601.5 (peak 7) and 659.6 (peak 12) in *L. saccharina* and 4 protonated molecules at *m*/*z* 449 (peak 1), 601.5 (peak 3), 891.2 (peak 6) and 905.5 (peak 7) in *L. digitata* (Table 2). However, MS spectra did not show any other protonated molecules from the remaining peaks of tested seaweed extracts.

Furthermore, all protonated ions were submitted for MS/MS fragmentation and their major fragmented ions are presented in Table 2. Results indicated that MS-MS fragmentation of peak 2 (*t*<sup>R</sup> 2.5 min) in *H. elongata* and peak 1 (*t*<sup>R</sup> 1.8 min) in *L. digitata* produced a major fragmented ion at *m*/*z* 287 [M + H <sup>−</sup> 162]<sup>+</sup> due to loss of a glucose molecule from the base peak ion *m*/*z* 449 (Table 2), suggesting the presence of cyanidin-3-*O*-glucoside, which corresponds to aglycone cyaniding [44]. Anthocyanin derivatives exhibit the characteristic UV-visible maxima at a range of 515–550 nm (band I) and 275–285 nm (band II) whereas, these compounds do not show any absorption at around 400 nm [23]. Characteristic UV spectra recorded for peak 2 (*H. elongata*) and peak 1 (*L. digitata*) which showed a λmax at 282 nm and 532 nm are in agreement with reported literature [23].

β-carotene, a carotenoid pigment, was identified only in *H. elongata* extract. A characteristic UV spectrum of peak 4 (*t*<sup>R</sup> 3.8 min) showed a λmax at 453 nm and 480 nm while MS data exhibited a molecular ion *m*/*z* 536.9 and upon MS/MS fragmentation, the major fragments were produced at *m*/*z* 444.2 and 430.3 corresponding to the elimination of toluene (92 u, atomic mass unit) and xylene (106 u), part of the central acyclic chain of the β-carotene skeleton, respectively (Table 2). This fragmentation pattern indicates the presence of extensive conjugation within the molecule or the cyclization of fragments of the polyene chain of the β-carotene skeleton [45]. The β-carotene is identified in accordance with the published results [41,46,47].

Peak 5 (*t*<sup>R</sup> 5.3 min) in *H. elongata,* Peak 7 (*t*<sup>R</sup> 5.0 min) in *L. saccharina* and Peak 3 (*t*<sup>R</sup> 5.6 min) in *L. digitata* extracts were identified as violaxanthin according to the λmax and its molecular ions. From MS analysis of these peaks, protonated molecular ion [M + H]<sup>+</sup> was detected at *m*/*z* 601.5 and fragment ions at *m*/*z* 583.5 and 565.5 corresponding to the loss of one (18 u) and two water molecules (36 u) from the protonated ion respectively (Table 2). These assignments are consistent with the ESI-MS/MS fragmentation pattern of violaxanthin standard and are also in agreement with the mass fragmentation data described by Rivera et al. [45] wherein similar MS/MS fragmented ions (*m*/*z* 583 and 565) were recorded for violaxanthin pigment.

Peaks 11 in *H. elongata* and Peak 12 in *L. saccharina* extracts showed the same absorption spectra with the λmax at 266 nm, 332 nm and 448 nm, but have the different retention times (*t*<sup>R</sup> 15.4 and 12.6 min respectively). The MS data showed a molecular ion *m*/*z* 659.6, suggesting the presence of fucoxanthin pigment, which was confirmed by the fragment ions at *m*/*z* 641.6 and 581.5 due to the loss of water (18 u) and acetic acid along with water (78 u) from the base precursor ion respectively (Table 2). A similar ESI-MS/MS fragmentation pattern was recorded with fucoxanthin standard which confirmed the presence of fucoxanthin pigment in both seaweed extracts [14].

Identification of chlorophyll in all 3 tested extracts was confirmed by characteristic λmax, MS and MS/MS fragmentation data. Peak 6 (*t*<sup>R</sup> 6.2 min) in *H. elongata,* peak 5 (*t*<sup>R</sup> 3.8 min) in *L. saccharina* and peak 7 (*t*<sup>R</sup> 9.4 min) in *L. digitata* extracts exhibited the same absorption spectra with the λmax at 430 nm, 620 nm and 662 nm which corresponds to chlorophyll *a* derivatives. Different retention time of chlorophyll *a* derivative in different tested seaweeds are probably due to the presence of different epimers of chlorophyll *a* molecule. Chlorophyll epimers exhibit identical absorption spectra to the chlorophyll molecule but show different chromatographic abilities [48]. For instance, chlorophyll *a*', an epimer of chlorophyll *a*, is less polar and appears on the longer retention time than chlorophyll *a*, because the –CHOOCH3 group at the C-132 position in the chlorophyll a molecule is on a different plane of the C-173 phytyl chain and is therefore less hindered, thus more polar than chlorophyll *a*' [49]. Epimers of chlorophyll and its derivatives are mostly naturally present but sometimes chlorophylls can be converted into its epimers during the extraction process. Therefore, it is anticipated that different seaweed extracts had different epimers of chlorophyll *a* derivative thus eluted at different time intervals. The most abundant product ions in ESI positive ion mass spectra of chlorophyll and its derivatives, usually relate to the dissociation of a quite weak esterifying phytyl linkage at the C-17 position of chlorophyll skeleton resulting in a fragmentation with the loss of the phytyl chain (as the phytadiene, C20H38) which appeared in the mass spectrum at the *m*/*z* value corresponding to [M <sup>+</sup> <sup>H</sup> <sup>−</sup> 278]<sup>+</sup> [48,50]. The MS spectra showed the precursor ion [M + H]<sup>+</sup> at *m*/*z* 893.5 and the fragment ions detected at *m*/*z* 615.2 correlating to the loss of the phytyl chain [M − 278]+ (Table 2). On the contrary, chlorophyll *b* derivative was detected only in *L. digitata* extract which was identified by absorption spectra of peak 6 (*t*<sup>R</sup> 10.4 min) with the λmax at 411 nm, 484 nm and 507 nm, and protonated

molecular ion [M + H]<sup>+</sup> at *m*/*z* 905.5. The presence of chlorophyll *b* derivative was confirmed by the fragment ions *m*/*z* 629.2 correlating to the removal of the phytyl chain from the chlorophyll skeleton (Table 2). Chlorophyll *b* has 14 u higher molecular weight than chlorophyll *a* because of the presence of formyl group (–CHO) instead of a methyl (–CH3) group at the C-7 position in chlorophyll skeleton. The presence of aldehyde group increases the polarity of chlorophyll *b* thus elutes prior to chlorophyll a on a non-polar C-18 column [49]. The identification of chlorophyll compounds was carried out as reported by Zvezdanovi´c et al. [50] who described a similar fragmentation pattern of chlorophyll derivatives using ESI-MS/MS.

Brown seaweeds are a valuable source of lipophilic antioxidants and these compounds have a tendency to dissolve in low polarity solvents and are considered to be lipophilic in nature [43]. Our results revealed that the extraction yield in the extracts from *n*-hexane (least polar) was significantly higher in all the seaweed, however, the same extracts exhibited the lowest antioxidant capacity and phytochemical constituents. Furthermore, extracts from Mix 4 showed lower extraction yield but displayed the highest antioxidant capacity and phytochemical constituents. This indicated that polarity of an extraction solvent has no direct relation with the extraction yield and antioxidant activity, and a selective solvent system (with optimum polarity) is required to extract lipophilic antioxidant compounds from seaweed. On a contrary, Matanjun [26] reported that more polar compounds were found in seaweed extracts and increasing solvent polarity increased the extraction yield.

It was observed that *H. elongata* was better seaweed than *L. saccharina* and *L. digitata* as a source of antioxidants. Results interpret that all extracts from *H. elongata* exhibited highest antioxidant capacity (DPPH and FRAP), total phenol and flavonoid content compared to *L. saccharina* and *L. digitata*. Previous studies also reported that *Fucoid* species (*H. elongata*) contained higher phytochemical constituents and antioxidant activity than kelps (*L. saccharina* and *L. digitata*) which is in agreement with the present results [22,37,46]. The high antioxidant activities of HE may be due to the high phenolic, flavonoid and carotenoid content. The results also indicated a strong correlation between the antioxidant activity (DPPH, FRAP) and total phenolic content, which agree with study of Duan et al. [36]. On a contrary, the metal-ion chelating ability was detected higher in *L. saccharina* and *L. digitata* as compared to *H. elongata* which are in agreement with our previous findings wherein methanolic extracts from *Laminaria* species exhibited higher chelating ability than *H. elongata* [22]. Metal chelating ability in terms of ferrous ion chelating capacity is claimed as one of the important mechanisms of antioxidant activity. The ferrous ions are the most powerful pro-oxidants among various species of transition metals present in food systems [51]. Antioxidants from seaweed could either act as free radical scavengers and mitigate the ROS/free radicals [52] or could prevent the formation of hydroxyl radicals by either deactivating free metal ions through chelation or converting H2O2 to other harmless compounds (such as water and oxygen) [11].

Previous studies have reported many compounds in seaweed, for example zeaxanthin, fucoxanthin, violaxanthin, β-carotene, phlorotannins, anthocyanin, gallic acid, kaempferol, gallic acid 4-*O*-glucoside, cirsimaritin, carnosic acid, epigallocatechin gallate, epicatechin and fatty acids, which are strong antioxidant components [11,14,39–41,45,53–55]. In this study, the antioxidant capacity in lipophilic extract were the result of pigments and phenolic compounds. Compounds such as cyanidin-3-*O*-glucoside, β-carotene, violaxanthin and fucoxanthin were identified in the Mix 4 extract of *H. elongata* which could be the reason that the selected seaweed exhibited the highest antioxidant capacity. Carotenoids compounds such as violaxanthin and fucoxanthin were identified in the *L. saccharina* extract while violaxanthin and cyanidin-3-*O*-glucoside were identified in the *L. digitata* extract. The extract from *H. elongata* exhibited more antioxidant compounds than *L. saccharina* and *L. digitata*, and the antioxidant capacity in 3 species follow the following order: *H. elongata* > *L. saccharina* > *L. digitata*. It is also anticipated that chlorophyll compounds were least responsible for the antioxidant capacity as Mix 4 extracts from tested species showed moderate total chlorophyll content but exhibited the highest antioxidant capacity. Lanfer-Marquez et al. [56] reported that chlorophyll derivatives shows antioxidant capacity at very high concentration by behaving as pro-oxidants. However, they do

not seem to donate hydrogen when exhibiting antioxidant capacity but may be involved in protection of linoleic acid against oxidation or by preventing breakdown of hydroperoxides. The study screened a selective solvent system for extracting lipophilic antioxidants and identified a range of antioxidant compounds. The identification of these lipophilic antioxidant compounds in selected brown seaweeds, can constitute a new move in the understanding of the health benefits of Irish brown seaweed as functional ingredients in food, cosmetics and medicinal preparation.

#### **4. Conclusions**

In conclusion, lipophilic extracts from Irish brown seaweed *H. elongata*, *L. saccharina* and *L. digitata* exhibit strong antioxidant property and metal-ion chelating ability. The phytochemical content and antioxidant capacity were majorly affected by the polarity or dielectric constant of extraction solvents. The highest phytochemical content and antioxidant capacity were achieved by an equal volume mixture of *n*-hexane, diethyl ether and chloroform (Mix 4) in all the seaweed studied. Among all the tested species, *H. elongata* was the most potent species which contained the highest antioxidant capacity followed by *L. saccharina* and *L. digitata*. The antioxidant capacity of *H. elongata* was comparable with that of reference ascorbic acid. A total of 10–11 lipophilic compounds with potential antioxidant capacity across the tested seaweed were identified by comparing retention times and UV spectral data. LC-ESI-MS/MS based characterization of lipophilic extracts confirmed the presence of fucoxanthin, violaxanthin, β-carotene, cyanidin-3-*O*-glucoside and other carotenoid and chlorophyll derivatives in the extracts. This suggests that algal derived lipophilic antioxidants may be the principal constituents responsible for the antioxidant properties from these species. These findings indicate that there may be a potential to further characterize these compounds in such extracts which can be used in pharmaceuticals, foods and cosmetics to act as antioxidants thus enhancing the quality and nutritive value of such products. Although seaweed has a great potential to be used as a source of natural antioxidant in food and cosmetics, their application as a dietary supplement or as a food ingredient should not be based only on in-vitro analysis which is just a preliminary screening tool. More research focusing on mechanisms of antioxidant action and activity against various free radicals will be advantageous in leading to the development of food and medicinal products to protect against certain age-related diseases. The identified lipophilic compounds/extracts should also be screened for their toxicity as well as for bioavailability and bioaccessibility in an in-vivo system prior to their application in commercial products.

**Author Contributions:** G.R. designed the work, performed the experiments, the statistical treatment of the data and wrote the manuscript.

**Funding:** This work was partly supported by Science Foundation Ireland (SFI) [grant number: 14/IA/2548].

**Acknowledgments:** The author would like to thank Prof. John O'Doherty for providing consumables for this work.

**Conflicts of Interest:** The author declares no conflict of interest.

#### **References**


© 2019 by the author. 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*
