Spirulina sp. biomass is from the Biochemical Engineering Laboratory, Federal University of Rio Grande (FURG), Rio Grande, RS, Brazil, cultivated and dried according to Morais et al. (9). The microalga grew in stirred glass fibre tanks (open raceway pond, volume capacity 10 000 L) with water from Mangueira lagoon (S33°30’13” and W53°08’59”), supplemented with 20% (by volume) Zarrouk medium (containing in g/L: NaHCO₃ 16.8, K₂HPO₄ 0.50, NaNO₃ 2.5, K₂SO₄ 1.00, NaCl 1.00, MgSO₄·7H₂O 0.20, CaCl₂ 0.04, FeSO₄·7H₂O 0.01 and EDTA 0.08). The microalgal biomass was filtered through a 200-μm filter after reaching the concentration of 1 g/L.

The Nannochloropsis sp. biomass grew in f/2 medium (culture medium for marine diatoms), salinity 28 PSU (practical salinity units), at 20 °C, light intensity 40 μmol/(m²·s) and photoperiod 12 h light/dark, and dried in the Phytoplankton and Marine Microorganism Laboratory at FURG, Rio Grande, Brazil (1).

The microalgal biomass samples were dried in tray dryers (Q314M; Quimis, São Paulo, Brazil) at 50 °C for 5 h, ground up to 32 mesh, vacuum packaged (AP450; Tecmac, São Paulo, Brazil) and stored at refrigeration temperature until further analyses.

The chemical characterization of microalgal biomass complied with the AOAC methodology: the moisture content was determined using gravimetry through drying in oven (Q314M; Quimis) at 105 °C according to method 930.04 (10), lipids were determined by extraction with petroleum ether according to method 954.02 (11), ash was determined by gravimetry, through incineration of the samples in a muffle furnace (Q318M24; Quimis) at 550 °C, as described in method 930.05 (12). Protein levels were determined by the micro-Kjeldahl method 977.02 (13). The crude fibre was determined by acid--base digestion according to method 978.10 (14). The percentages of the other carbohydrates were estimated by difference, considering the sum of protein, lipid, ash, moisture and crude fibre contents.

Phenolic compounds, soluble in methanol or ethanol, or insoluble (naturally cell wall-bound phenolics) were obtained from the microalgae via the protocols described below. Phenolic compounds soluble in methanol were extracted from 6 g biomass homogenized with 20 mL methanol (p.a., Labsynth®, São Paulo, Brazil) in an orbital shaker (TE-141; Tecnal, Erechim, Brazil) at 25 °C and 200 rpm for 60 min. Agitation was interrupted for 15 min and then restarted, after the addition of 20 mL methanol, for 90 min. Extracts were centrifuged at 3020×g for 10 min (refrigerated centrifuge CT-5000R; Cientec, São Paulo, Brazil) at room temperature, filtered with Whatman filter paper no. 1, evaporated in a rotary evaporator (801-802; Fisatom, São Paulo, Brazil) at 50 °C, diluted in 40 mL sterile distilled water and clarified with 10 mL Ba(OH)₂ (p.a., Labsynth®) 0.1 M and 10 mL ZnSO₄ (p.a., Labsynth®) 5%. Clarified extracts were vacuum filtered (MA 454; Marconi, São Paulo, SP) with sterile membrane (pore size 0.25 μm; Minisart, Goettingen, Germany) (3).

Phenolic compounds soluble in ethanol were extracted from microalgal biomass with 80% ethanol (p.a., 1:3 mass per volume; Labsynth®) in orbital shaker at 300 rpm for 10 min, followed by centrifugation at 3020×g and 10 °C for 15 min (this procedure was performed twice). The soluble fraction was evaporated and resuspended in distilled water.

The solid residue, from the last step, was used for extraction of bound phenolic compounds by adding 10 mL hexane (p.a., Labsynth®), homogenizing in an orbital shaker at 300 rpm for 10 min and separating by centrifugation at 3020×g and 10 °C for 20 min. The solvent was removed by filtration through Whatman no. 1 filter paper.

The solid phase was hydrolyzed with 60 mL 4 M NaOH (p.a., Labsynth®) under forced nitrogen atmosphere in orbital shaker at 250 rpm and 25 °C. After 4 h, 6 M HCl (p.a., Labsynth®) was added to reach pH=1. The suspension was centrifuged at 3020×g and 10 °C for 40 min and then the supernatant was used in a partition with ethyl acetate (p.a., Labsynth®) (three times in 30 mL). The liquid was further separated with Whatman no. 1 filter paper with anhydrous Na₂SO₄ P.A. (p.a., Labsynth®) and evaporated in a rotary evaporator at 60 °C and resuspended in sterile water (15). Fig. 1 shows the analytical sequence for obtaining the three phenolic extracts from the microalgal biomass.

The phenolic acids found in the algal extracts (standards by Sigma-Aldrich, Merck KGaA, St. Louis, MO, USA: caffeic, chlorogenic, coumaric, ferulic, gallic, hydroxybenzoic, protocatechuic, syringic and vanillin) were determined by using an aliquot dried under nitrogen flow, resuspended in methanol/water mixture (1:1) and separated by high-performance liquid chromatography (CLASS-M10A; Shimadzu, Tokyo, Japan) coupled to ultraviolet detector, with the use of a C18 reversed-phase column (4.6 mm×250 mm, 5 μm; Discovery, Bellefonte, PA, USA). The mobile phase flow rate was 0.7 mL/min at 35 °C, with an isocratic elution of solvents composed of HPLC grade methanol (J.T. Baker, Phillipsburg, NJ, USA) and water with 1% acetic acid in the ratio 20:80 (by volume) for 25 min. Phenolic acids in the eluate were measured at 280 nm up to 15 min of elution and at 320 nm up to the final 25 min (15).

The efficiency of the chromatographic separation was estimated by the retention factor (k) which is equal to the ratio of retention time of every phenolic acid on the column and the retention time of an unretained compound (dead time). The separation factor (α) was estimated between two phenolic acids as the ratio of their retention factors (k_A/k_B) (16).

To confirm the results, the artificial extracts were subjected to HPLC analysis in an Alliance 2695 Separations Waters liquid chromatograph (Milford, MA, USA) coupled to a mass detector (LC-ESI-MS/MS) fitted with an autosampler, a membrane degasser and a quaternary pump. Mass spectrometry was performed on a Micromass Quattro Micro^TM API (Waters) with an electrospray ionisation (ESI) interface. Additionally, data were acquired with MassLynx v. 4.1 (17), and a Waters XTerra® MS C18 column, 3.5 μm, 144 Å (50 mm×3 mm i.d.) was used. The mobile phase was methanol, ultrapure water with 0.1% acetic acid (90:10), in an isocratic elution and a flow rate of 0.3 mL/min. The fragmentation conditions were: source temperature 100 °C, desolvation temperature 400 °C, and desolvation gas flow (N₂) and cone gas flow 500 and 50 L/h, respectively. The capillary voltage was 4 kV and the extractor 2 V. Table 1 shows the phenolic compounds and the fragmentation conditions that were studied. The conditions were as determined by Singh et al. (18).

The antiradical activity of the microalgal extracts was estimated by four different methods, with some alterations. In these experiments, only the phenolic extracts extracted with methanol were used, since antifungal tests showed that these extracts have the highest potential (results not shown). Solutions with a mixture of the most abundant phenolic acid standards were also formulated with the same ratios found in the natural extract.

Analysis of variance (ANOVA) determined the results of the antiradical activity, followed by the Tukey’s test with 95% confidence level, using Statistica v. 12 (25) to verify the influence of different phenolic acids found in the microalgal extracts on the antiradical potential.

The approximate composition of microalgal biomass determined in this study (Table 2) is a confirmation of the high protein content reported by other authors in the Spirulina biomass (9) and the high lipid content of Nannochloropsis (26) biomass. These biomolecules are the main focus of the vast range of studies of microalgal biomass, from food formulation to biodiesel production.

The efficiency parameters of the chromatographic method employed to quantify the phenolic acids (Table 3) showed that it has adequate coefficients of determination, in agreement with International Union of Pure and Applied Chemistry (27) recommendations, of R² values equal to or higher than 0.9. Furthermore, the low phenolic acid quantification limit is important to ensure analysis reliability, especially when the objective is to determine trace components in biological matrices (28).

The phenolic acid retention factors ranged from 5.4 to 26.6, which is due to the long analysis time (16). In this study, we determined nine compounds in a 25-minute chromatographic run. Separation factors were above 1 and had appropriate elution sequence, with neither peak superposition nor peak enlargement despite the unusually large retention factor in some runs. The identification of phenolic acids found in microalgae (Table 4) showed that each phenolic fraction (methanolic, ethanolic and bound extracts) had different composition, as it was expected (2). The chromatograms of standards and phenolic extracts of microalgae in Fig. 2 show the efficiency of separation.

The profile of phenolic acids identified in the Spirulina biomass extracts was similar to the one reported by other authors who also found gallic, salicylic, trans-cinnamic and chlorogenic acids (4, 29). However, identification studies of Nannochloropsis biomass phenolic compounds that specifically mention distribution of phenols are sparse (30, 31).

Phenolic compounds are a large chemical family associated with defense mechanism against endogenous and exo-genous factors, such as oxidative process, light, temperature and pathogen invasion. They have been found in photosynthetic microalgae (4, 29), however, their determination in microalgae is not frequent. The study of cells that produce these compounds is even more important than the phenolic compounds themselves. The determination of the profile of phenolic compounds in the cells can identify phenolic sources for application as natural preservatives in the food industry, therefore, they were the target of our experiment.

The ability of phenolic acids to donate hydrogen to the free radical that is formed in the cell determines their antioxidant properties. There are two groups of phenolic acids: derivates of benzoic and cinamic acids, whose chemical structures, especially their hydroxilated forms, determine their potential to protect against oxidants (2). They act by inhibiting the release and propagation of free radicals in some reactions and oxidative and hydrolytic enzymes. Therefore, to evaluate the antioxidant properties of phenolic compounds, it is recommendable to carry out methods that could reflect the antioxidation mechanism.

The main phenolic compound in aqueous extracts from microalgae found in this study was chlorogenic acid, whose antioxidant activity and capacity of inhibiting mutagenicity promoted by chemical compounds has already been shown (32).

Different procedures determined the antioxidant activity of the extracts to estimate their minimal concentration for the protective effect, stability and versatility of action mechanism. The determined radical ABTS inhibition showed the ability of the extracts to hinder the production of free radicals and allowed the estimation of the minimal dose-effect. DPPH is a stable radical with low deterioration and reactivity rates towards most compounds found in a biological system and may serve as indicator of the stability of antioxidant compounds. Only strong reducing agents are able to react with this type of radical stoichiometrically (20) and the determination allowed the evaluation of the stability of the phenolic acid inhibition. The determination of inhibition of peroxidase activity took into account the importance of knowing the prooxidant potential of the extracts, which can be evaluated by the inhibitory effect against the specific enzyme activity of peroxidase. This enzyme acts on reducing compounds of phenolic nature in the presence of hydrogen peroxide, thus yielding quinones depending on the oxidant potential of the substrate (21, 22). Within the cells, this enzyme participates in defense mechanisms but it may also lead to oxidative stress, however, it is expected that in microalgae the phenolic compounds are able to inhibit this process.

The phenolic extracts from microalgal biomass were significantly (p<0.05) more efficient against ABTS radical (Table 5) than synthetic mixtures of phenolic acid, which suggests the ability to prevent free radical production. The higher efficiency of the crude phenolic extracts is promising and more advantageous than using purified phenolic acid compounds. The downstream processing of the crude extract may be difficult and expensive when considering mass production. Thus, using crude microalgal extracts can be more economically viable with more benefits than using a purified phenolic compound mixture.

Natural methanolic extracts from both microalgae had the highest inhibitory capacity against ABTS: 1.2 and 1.7%/(min·μg) from Spirulina sp. and Nannochloropsis sp., respectively. Values of IC₅₀ obtained by the linear regression showed that the lowest concentration needed to capture 50% of the radical ABTS was achieved by Nannochloropsis sp. natural extract (IC₅₀=8.0 μg/mL), and its synthetic mixture was also the most efficient one (IC₅₀=14 μg/mL) among the artificial preparations. The antiradical potential of phenolic extracts refined with methanol is due to the synergy of different phenolic compounds. Therefore, purifying the phenolic acid from a natural phenolic extract, emulating the natural distribution of different kinds of phenolic compounds may not lead to optimal protection in food or pharmaceutical formulations (2).

Table 5 shows the results of the assays of phenolic extracts from microalgae carried out during this study. The natural extract from Spirulina sp. showed the best properties during 180 min of monitoring the DPPH inhibitory activity. The natural extract and the synthetic mixtures of phenols from Nannochloropsis sp. showed a tendency to reduce the protecting effect of phenols, suggesting that they are more adequate to apply in situations where shorter protection intervals are necessary. The inhibitory effect of the phenolic extracts from Nannochloropsis sp. is beyond the capacity of the equivalent synthetic mixture; thus, it is a source of antioxidant compounds that may be explored to protect cells and products in situations where the inhibition of the oxidative process needs to occur much faster.

The compounds with antioxidant activity may act in different phases of the oxidative process (33). Considering our results of the ABTS and DPPH methods, the phenolic extracts from microalgae were able to inhibit the propagation phase. The reduction power stopped for short intervals, indicating that the extracts held to their free radical scavenging capacity, halting the chain reaction.

It is worth mentioning that few studies have determined the inhibitory role of phenolic compounds in oxidative processes promoted by enzymatic action, however, it is very important in biological materials as part of early defense mechanism (21, 22). These are triggered by enzymes, mainly peroxidase, therefore its inhibition is an interesting sign of antioxidant action. The measurement of peroxidase activity in the presence of phenolic extracts enables to investigate the role of these compounds in cell defense mechanisms, choosing those of interest for use as preservatives in food industry. Table 5 shows that the best extract, when regarding this property, was the one from Nannochloropsis sp. It differed significantly (p<0.05) from the others in the shortest study time.

The inhibitory activity of amylase from the extracts may constitute a natural defense mechanism of microalgae against fungal attacks and this may be interesting for protection against endogenous hydrolytic process. This potential was estimated by checking the inhibition capacities of each extract against fungal amylase. The natural Spirulina sp. extract had higher effect on the reduction of the activity of amylase, 0.08%/(min·μg), and differed significantly (p<0.05) from the other extracts. The synthetic mixture of this microalga had a reduction effect of 0.01%/(min·μg). Nannochloropsis sp. natural extract led to a 0.01%/(min·μg) reduction of this enzyme activity.

The inhibitory activity of amylase from microalgal phenolic compounds shows that their application against fungal contamination in the fields or for grain beneficiation may be viable. The production of compounds from microalgae with inhibitory effects against amylases has not been deeply investigated. Therefore, this study is pioneering and may contribute to inferring antifungal activity of phenolic extracts from microalgae, since they survive in environments where they compete with fungi (29).

Fig. 3 shows a schematic representation of the action mechanisms of phenolic acids, taking chlorogenic acid as an example, considering that this phenolic acid is present mostly in microalgae Nannochloropsis sp. and Spirulina sp.