Nutritional Profiling and Preliminary Bioactivity Screening of Five Micro-Algae Strains Cultivated in Northwest Europe

This study aimed to map the nutritional profile and bioactivities of five microalgae that can be grown in Northwest Europe or areas with similar cultivation conditions. Next to the biochemical composition, the in vitro digestibility of carbohydrates, proteins, and lipids was studied for Chlamydomonas nivalis, Porphyridium purpureum, Chlorella vulgaris, Nannochloropsis gaditana, and Scenedesmus species biomass. These microalgae were also assessed for their ability to inhibit the angiotensin-1-converting enzyme (ACE-1, EC 3.4.15.1), which is known to play a role in the control of blood pressure in mammals. Large differences in organic matter solubility after digestion suggested that a cell disruption step is needed to unlock the majority of the nutrients from N. gaditana and Scenedesmus species biomass. Significant amounts of free glucose (16.4–25.5 g glucose/100 g dry algae) were detected after the digestion of C. nivalis, P. purpureum, and disrupted Scenedesmus. The fatty acid profiles showed major variations, with particularly high Ω-3 fatty acid levels found in N. gaditana (5.5 ± 0.5 g/100 g dry algae), while lipid digestibility ranged from 33.3 ± 6.5% (disrupted N. gaditana) to 67.1 ± 11.2% (P. purpureum). C. vulgaris and disrupted N. gaditana had the highest protein content (45–46% of dry matter), a nitrogen solubility after digestion of 65–71%, and the degree of protein hydrolysis was determined as 31% and 26%, respectively. Microalgae inhibited ACE-1 by 73.4–87.1% at physiologically relevant concentrations compared to a commercial control. These data can assist algae growers and processors in selecting the most suitable algae species for food or feed applications.


Introduction
Micro-algae, further referred to as algae, have emerged as a sustainable and nutritious feedstock for food and feed [1,2]. The CO 2 -capturing capacity of autotrophic algae and the possibility of cultivating them without using fertile soil and with limited use of water, enables sustainable algae cultivation. They are an attractive source of nutrients such as proteins and long-chain polyunsaturated fatty acids (PUFAs), including eicosapentaenoic acid (EPA, C20:5) and docosahexaenoic acid (DHA, C22:6) [3]. Specific algae species are, together with fish, the sole sources in the human diet of EPA and DHA, which both serve important physiological functions in the body [4]. Next to nutrient levels, nutrient digestibility and accessibility are also relevant in this context. Although research has been carried out on the digestibility of algae proteins [5][6][7][8] and lipids [9,10], carbohydrate digestibility has received little attention. Carbohydrates are, nevertheless, a major algae constituent [11] and have the potential to provide a significant amount of energy. In some studies [12], the total pool of soluble carbohydrates after digestion was measured. However, this only provides a very rough estimation of carbohydrate digestibility. Indeed, it can be hypothesized that a significant portion of soluble carbohydrates is indigestible. Algae carbohydrates are very diverse in terms of composition and linkage type [11], making them less likely to be (4 h). The total lipid content was measured by chloroform:methanol extraction according to the procedure of Ryckebosch et al. [14]. For the analysis of the fatty acid composition, the fatty acids in the lipid extract were methylated under sequential alkaline and acid conditions following the ISO 12966-2:2011 protocol [15]. Fatty acid methyl esters (FAMEs) were quantified by gas chromatography-flame ionization detection (GC-FID) using a TR-FAME column (100 m length, 0.25 mm internal diameter, 0.20 µm film thickness, Thermo Scientific, Waltham, MA, United States). Amino acid analysis was performed by high-performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD) after acid hydrolysis with phenol-HCl (6 N, 23 h, 110 • C) according to ISO protocol 13903:2005. Amino acid separation was performed using a Dionex AminoPac PA-10 column (2 nm × 250 nm) and a previously detailed elution gradient [16]. Tryptophane was not analyzed due to its liability for acid degradation. Only the sum of glutamate and glutamine was measured, because the latter is converted to the first during acid hydrolysis. Similarly, the sum of aspartate and asparagine was measured, and not their individual levels. The total protein content was calculated as the sum of all amino acids each corrected for water uptake during hydrolysis. The carbohydrate content was measured after a two-step H 2 SO 4 hydrolysis and HPAEC-PAD analysis, as described before [17]. The total carbohydrate level was estimated by summing monosaccharides and uronic acids, each corrected for water uptake during hydrolysis. For mineral analysis, algae (0.4 g) were mixed with AgNO 3 (1 M, 1 mL) and HNO 3 (5 mL, 67-69%), incubated for 2 h at 105 • C and transferred to a new tube. The sample was diluted with ultrapure water to a volume of 15 mL, centrifuged (20 min, 4000 rpm) and the supernatant was transferred to another tube. The washing step was executed two more times by adding 15 mL of water, shaking, centrifugation, and collecting the supernatant. The volume of the combined supernatant phases was adjusted to 50 mL with water, after which, the mixture was analyzed by inductively coupled plasma atomic emission spectroscopy (Thermo iCAP analyzer, Thermo Scientific) for P, K, Ca, Na, and Fe content. The precipitate was incubated with 1 mL of NH 3 (25% m/m) in the dark and shaken. This was repeated until the precipitate turned into a solution. Next, the dissolved material was transferred to the first tube that was used for the AgNO 3 /HNO 3 incubation. After dilution with water, the diluted product was analyzed for Cl content with high resolution inductively coupled plasma mass spectrometry (Thermo Element 2, Thermo Scientific). The latter was performed after calibration using 0-250-500-1000 µg Cl − /L solutions in 0.1% NH 3 (v/v). Following the same protocol as above, a quality control was prepared with 1000 mg Cl − /L to verify that recovery was at least 95%. The Cl analysis was based on the procedure of Naozuka et al. [18].

In Vitro Digestibility
The standardized in vitro digestion protocol proposed by Minekus et al. [19] was used to evaluate nutrient digestibility. Blanks containing digestive fluids, enzymes, and bile but no algae, were included in each experiment to account for nutrients not coming from algae digestion. Digestion tests were preceded by pH adjustment experiments to accurately determine the amount of NaOH and HCl needed to reach the pH values prescribed in the consensus protocol [19]. As recommended by Brodkorb et al. [20], the substrate:fluid ratio for oral phase simulation was adapted in order to obtain a swallowable bolus with a paste-like consistency. In fact, substrate:fluid ratios of 1:5 and 1:2 were used for P. purpureum and the other algae samples, respectively. Water was added to reach the standard mass at the start of digestion.
For the evaluation of protein digestibility, the digestion was stopped by adding a protease inhibitor (1 mM 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride). Subsequently, the digesta was centrifuged (5 min, 5000× g) and supernatant aliquots were examined. Firstly, the total amount of nitrogen in the supernatant was measured with an automated Dumas protein analysis system (Vario EL Cube, Elementar, Langenselbold, Germany) and compared with the amount of nitrogen in the initial algae biomass. Secondly, the degree of protein hydrolysis was estimated by the quantification of the primary amino groups and comparison with the total nitrogen in the sample. Samples were incubated overnight in 2% sodium dodecyl sulfate solution at 4 • C, and added to a freshly prepared mixture of o-phthalaldehyde (OPA) and N-acetyl-cysteine (NAC) in borate buffer, as described before [21]. By using an L-isoleucine calibration series and absorbance measurement at 340 nm (Tecan Infinite 200 PRO, Tecan, Männedorf, Switzerland), the concentration of primary amino groups can be determined [22]. The degree of protein hydrolysis was calculated as the ratio of primary amino groups to the total algae nitrogen. A correction was made for primary amino groups and total nitrogen in the blank.
For the assessment of the lipid digestibility, digestion was ended by adding chloroform:methanol, as proposed by Brodkorb et al. [20], immediately followed by lipid extraction, as described above. In case lipid extraction could not be executed instantly after digestion, a lipase inhibitor was added (4-bromophenylboronic acid in methanol, 5 mM inhibitor/mL digesta), the digesta was snap frozen, stored at −20 • C and extracted with chloroform:methanol on another day. After lipid extraction, the amount of free fatty acids (FFAs) was determined after their derivatization to fatty acid diethylamides and subsequent GC-FID analysis [23]. A famewax column (30 m length, 0.32 mm internal diameter, 0.25 µm film thickness, Restek, Bellefonte, PA, USA) was used for GC separation, pentadecanoic acid (C15:0) as the internal standard, and a lauric acid (C12:0) dilution series for detector calibration.
Carbohydrate digestibility was tested by executing the standard digestion protocol followed by centrifugation (5 min, 5000× g) and incubation with Aspergillus niger amyloglucosidase (A1602 Sigma Aldrich, Munich, Germany), which is needed to simulate the activity of intestinal brush border enzymes [19]. To this end, 1 mL of supernatant was transferred to a new recipient, the pH adapted to 4.5 and A. niger amyloglucosidase (12 U) was added. Next, the mixture was incubated for 60 min at 37 • C, heated (10 min at 95 • C) to inactivate enzymes, and the hydrolysate diluted and analyzed by HPAEC-PAD, as described above.
Finally, the concentration of dissolved orthophosphate in the digesta was measured. After the heat inactivation of enzymes (10 min at 95 • C), the digesta was cooled to room temperature and diluted so that dissolved orthophosphate could be measured with either the LCK348 or LCK 350 total/ortho phosphate kit (Hach Lange, Düsseldorf, Germany) according to the manufacturer's instructions.

Angiotensin-1-Converting Enzyme (ACE-1; EC 3.4.15.1) Inhibition Assay
Assessment for the ability of the microalgae to inhibit the ACE-1 enzyme was carried out using the ACE-I inhibition bioassay using a previously published method [13,24]. The ACE-1 inhibition kit (ACE-1 inhibition kit-WST, Dojindo Laboratories, Kumamoto, Japan) was used. In brief, 20 µL of each sample prepared at a concentration of 1 mg/mL in water was added to 20 µL of the substrate and 20 µL of the enzyme working solution in triplicate. Captopril © was used as a positive control at a concentration of 0.05 mg/mL dissolved in ddH 2 O. Samples were incubated at 37 • C for 1 h. An amount of 200 µL of indicator working solution was then added to each well, and subsequent incubation at room temperature was carried out for 10 min. Absorbance at 450 nm was read using a FLUOstarOmega microplate reader (BMG LABTECH GmbH, Offenburg, Germany). The percentage ACE-1 inhibition was calculated using the following equation: where A 0 is the substrate absorbance at 450 nm in the presence of ACE-I and absence of the inhibitor, and A I is the substrate absorbance at 450 nm in the presence of ACE-I and the inhibitor or Captopril © (positive control).

Statistics
Statistical analyses were performed with Statistica version 12 (Dell Inc., Tulsa, OK, USA, 2015) with 5% as the significance threshold level. One-way ANOVA was used to evaluate the impact of algae type on digestibility parameters. After a positive omnibus test, a post hoc Tukey-test was performed.

Sample Pretreatment and Biochemical Composition
When algae are cultivated for food or feed applications, it is important that nutrients do not remain physically locked in the cell during digestion. In that case, a cell disruption step may be appropriate so that the algae's nutritional potential is exploited to the maximum.
A preliminary test was therefore performed to identify for which algae the cell structure could have such a nutrient-encapsulating effect. Algae were digested in vitro and the amount of soluble organic matter was determined. Since carbohydrates, proteins, and lipids can only be absorbed in the small intestine when they are degraded to smaller building blocks that are soluble in the liquid/micellar phase, a low organic matter solubility after digestion can be seen as an indication for a low combined digestibility of carbohydrates, proteins, and lipids.
For C. nivalis, P. purpureum and C. vulgaris, around 60% of the organic matter was soluble after in vitro digestion, while for N. gaditana and Scenedesmus, only about 20% of the organic matter was soluble after digestion ( Figure 1). This suggests that the majority of the organic fraction of N. gaditana and Scenedesmus biomass cannot be absorbed in the small intestine.
USA, 2015) with 5% as the significance threshold level. One-way ANOVA w evaluate the impact of algae type on digestibility parameters. After a positiv test, a post hoc Tukey-test was performed.

Sample Pretreatment and Biochemical Composition
When algae are cultivated for food or feed applications, it is important th do not remain physically locked in the cell during digestion. In that case, a cel step may be appropriate so that the algae's nutritional potential is exploited t mum.
A preliminary test was therefore performed to identify for which algae th ture could have such a nutrient-encapsulating effect. Algae were digested in v amount of soluble organic matter was determined. Since carbohydrates, prote pids can only be absorbed in the small intestine when they are degraded to sm ing blocks that are soluble in the liquid/micellar phase, a low organic matte after digestion can be seen as an indication for a low combined digestibility drates, proteins, and lipids.
For C. nivalis, P. purpureum and C. vulgaris, around 60% of the organic soluble after in vitro digestion, while for N. gaditana and Scenedesmus, only ab the organic matter was soluble after digestion ( Figure 1). This suggests that t of the organic fraction of N. gaditana and Scenedesmus biomass cannot be abso small intestine. This accords with earlier observations showing that none or only a negligi of the lipids and proteins are bio-accessible in non-disrupted Nannochloropsis and that cell disruption is crucial to make lipids [9,10] and carotenoids [9] bi from Nannochloropsis cells. Scenedesmus sp. have a notoriously recalcitrant ce heating to 90 °C [25], ultrasound treatment [26], or a combination of a mech enzymatic treatment [27] were previously needed to significantly disrupt Sc cells and to solubilize the cell content. When the Scenedesmus and N. gadita were disrupted by bead milling before digestion in the current study, the solu matter fraction increased to 68.9 ± 3.5% and 70.5 ± 2.2%, respectively (Figure data indicate that a disruption step is needed to make considerable levels o susceptible to digestion and available for absorption, it was decided to only c the disrupted N. gaditana and Scenedesmus samples in the further course of the This accords with earlier observations showing that none or only a negligible fraction of the lipids and proteins are bio-accessible in non-disrupted Nannochloropsis oculata [10] and that cell disruption is crucial to make lipids [9,10] and carotenoids [9] bio-accessible from Nannochloropsis cells. Scenedesmus sp. have a notoriously recalcitrant cell wall, and heating to 90 • C [25], ultrasound treatment [26], or a combination of a mechanical and enzymatic treatment [27] were previously needed to significantly disrupt Scenedesmus cells and to solubilize the cell content. When the Scenedesmus and N. gaditana biomass were disrupted by bead milling before digestion in the current study, the soluble organic matter fraction increased to 68.9 ± 3.5% and 70.5 ± 2.2%, respectively ( Figure 1). As these data indicate that a disruption step is needed to make considerable levels of nutrients susceptible to digestion and available for absorption, it was decided to only characterize the disrupted N. gaditana and Scenedesmus samples in the further course of the study.
Due to the high sample amounts needed, it was decided to use a new batch of disrupted N. gaditana from this point onwards. This new batch had a soluble organic matter after digestion of 62.8 ± 0.8%.
As shown in Table 1, carbohydrates were important algae constituents for C. nivalis, P. purpureum, and disrupted Scenedesmus and the sum of all monosaccharides and uronic acids ranged between 38% and 51% of the dry matter (dm). Proteins were the main constituent of disrupted N. gaditana and C. vulgaris biomass. Lipid levels ranged between 13.2% (P. purpureum) and 31.9% (disrupted N. gaditana). The digestibility of these macronutrients is discussed below.  1 33 Carbohydrates were calculated as the sum of all monosaccharides and uronic acids, while proteins were calculated as the sum of all amino acids, each corrected for water uptake during hydrolysis.

Carbohydrate Composition and Digestibility
The monosaccharide composition analysis ( Figure 2a) shows that glucose was the main monosaccharide in all algae samples, with concentrations ranging between 7.1% and 38.1% on a dm basis. Other monosaccharides, mainly galactose, xylose, and arabinose, were present in lower concentrations ( Figure 2a).
Carbohydrates serve various functions in the algae cell, including building up the cell structure and serving as an energy stock. Glucose can be part of the structural carbohydrates, such as the extracellular polysaccharides in Porphyridium sp. [28,29] or as cellulose as in N. gaditana [29,30]. Additionally, glucose is also part of glucose-based storage polysaccharides. C. nivalis is known to accumulate starch [31,32], and starch levels up to 35% and 39% on a dm basis were reported for Chlamydomonas reinhardtii [33,34] and heterotrophically cultivated Scenedesmus sp. [35,36], respectively. Only 2.1% of floridean starch was found in a commercial P. purpureum sample [29] but levels might be higher under different growth conditions, as seen for C. vulgaris [37,38]. Indeed, low levels (0-14%) of α-linked glucose were found in commercial Chlorella sp. [6] and C. vulgaris [29] products but starch levels of 30-35% [39] and 50% [40] were obtained for C. vulgaris cultivated under conditions favoring starch accumulation. N. gaditana differs from the other algae species in that it contains no or negligible amounts of starch [41], while it synthesizes chrysolaminarin, a storage polymer composed of β-linked glucose units [42].
When the algae fractions underwent in vitro digestion, significant amounts of free glucose (16.4-25.5 g glucose/100 g dry algae) were detected in the case of C. nivalis, P. purpureum, and disrupted Scenedesmus (Figure 2b). The digestion of C. vulgaris and disrupted N. gaditana yielded only 3.6 g free glucose/100 g dm and 1.8 g free glucose/100 g dm, respectively. This was expected, since these algae had a much lower initial glucose content.
These data do not allow a precise prediction of the in vivo glucose release, since a detailed kinetic study of starch degradation was not performed and because of the limitations of in vitro digestion studies [20]. Yet, these results do suggest that considerable glucose levels can become available for intestinal uptake during the digestion of C. nivalis, P. purpureum and disrupted Scenedesmus, and that glucose will contribute significantly to the energy content. It is noteworthy that algae biomass was not heated before digestion. One would expect that algal starch was not yet gelatinized and hence, partly resistant to hydrolysis by intestinal enzymes, as seen for other starch types such as cereal starches [43]. Indeed, in vitro digestion was performed at 37 • C, so below the gelatinization temperature of common food starches (50-70 • C) [44]. Yet, it is plausible that gelatinization occurs at lower temperatures for P. purpureum since the gelatinization of floridean starch in red macroalgae was reported to start already at 41-42 • C [45]. For disrupted Scenedesmus, the disruption step could have affected the degree of starch (semi-)crystallinity, which may increase its susceptibility to enzymatic degradation, as is the case, for example, for ball-milled microcrystalline cellulose [46]. Moreover, it cannot be excluded that endogenous algae enzymes also played a role in degrading (non-gelatinized) starch. For disrupted N. gaditana, the limited glucose release is not surprising, since the glucose units in chrysolaminarin are connected by β-linkages, which are not susceptible to degradation by human α-amylases or disaccharidases [47].  Table S1.
Carbohydrates serve various functions in the algae cell, including building up the cell structure and serving as an energy stock. Glucose can be part of the structural carbohydrates, such as the extracellular polysaccharides in Porphyridium sp. [28,29] or as cellulose as in N. gaditana [29,30]. Additionally, glucose is also part of glucose-based storage polysaccharides. C. nivalis is known to accumulate starch [31,32], and starch levels up to    Table S1. The concentrations of the other free monosaccharides (<2%) were much lower than that of glucose (Figure 2b). Galactose, fructose, and mannitol will have a negligible impact on energy supply. However, the latter two can still be relevant, as they belong to the socalled FODMAPs. FODMAPs are fermentable oligo-, di-and monosaccharides and polyols that are not or are incompletely absorbed in the small intestine. These compounds can cause gastro-intestinal problems with sensitive subjects upon consumption. Clinical trials indicated that eliminating FODMAPS from the diet relieves the symptoms of 50-80% of patients with irritable bowel syndrome with improvements in bloating, flatulence, diarrhea and general symptoms [48,49]. Yet, the fructose and mannitol amounts released during digestion of the studied algae (Figure 2b) are low compared with the levels previously reported to be mal-absorbed [49] and are probably too low to induce adverse effects.

Lipid Composition and Digestibility
To assess the nutritional value of the lipid fraction, the fatty acid composition was first evaluated ( Table 2). The discussion focuses on the essential fatty acids linoleic acid (LA, (cis,cis-9,12)) and α-linolenic acid (ALA, C18:3 (cis 9,12,15)) on the one hand, and on EPA (C20:5) and DHA (C22:6) on the other hand. Since EPA and DHA synthesis by the human body is limited, several food authorities have set recommendations on their intake as well [4]. Table 2. Fatty acid concentrations (mg/g dry algae) in non-digested algae biomass. LA = linoleic acid, ALA = α-linolenic acid, EPA = eicosapentaenoic acid, DHA = docosahexaenoic acid, SFA = saturated fatty acids, MUFA = mono-unsaturated fatty acids, PUFA = poly-unsaturated fatty acids.
Lipid digestibility was evaluated by comparing the concentration of free fatty acids (FFAs) after digestion with the total fatty acid content of the sample. It should be noted that this is not the same as the lipid bio-accessibility, i.e., the fraction available for uptake in the small intestine. Indeed, not only FFAs but also other lipid degradation products such as monoacylglycerols participate in the formation of micelles that are taken up by enterocytes in the gut. On the other hand, the incorporation of entirely hydrolyzed lipids in mixed micelles may be limited due to the presence of other cell compounds or due to the lipid localization in the cell [9].
P. purpureum and C. nivalis displayed the highest lipid digestibility, with 67.1 ± 11.2% and 58.9 ± 11.7% of the fatty acids liberated as free fatty acids after digestion, respectively ( Figure 3a). Lower average values were observed for C. vulgaris, disrupted N. gaditana and disrupted Scenedesmus (36.6 ± 18.2%, 33.3 ± 6.5% and 42.8 ± 5.3%). C. nivalis had the highest absolute FFA content after digestion (69.6 ± 7.0 mg FFA/g dry algae, Figure 3b). Due to the stereoselectivity of pancreatic lipases, only a limited fraction of the fatty acids can be liberated. Bernaerts et al. [9] estimated that in the case of Nannochloropsis, only 55-65% of the fatty acids can be released where this maximal hydrolysis value depends on the lipid classes (triacylglycerols, phospholipids, glycolipids) present in the biomass.
To our knowledge, no literature is available on the lipid digestibility of C. nivalis and P. purpureum. Sonication experiments with Chlamydomonas reinhardtii suggested that carotenoids are already fairly bio-accessible in C. reinhardtii without pre-treatment and that sonication has no additional effect [60]. Accordingly, a high lipid digestibility was found in this study for C. nivalis (58.9 ± 11.7%).
Regarding C. vulgaris, several studies, listed in a recently published review [11], showed that carotenoid bio-accessibility can be increased by pretreating the biomass by cell disruption. Yet, an enzymatic disruption could not increase chlorophyll release in C. vulgaris biomass [61]. Similarly, a mechanical disruption treatment was unnecessary when chlorophyll bio-accessibility was considered in C. vulgaris biomass, with bio-accessibility values already ranging between 77-84% for untreated biomass [62]. No literature was found on fatty acid release during C. vulgaris digestion.
The lower value observed here for disrupted N. gaditana (33.3 ± 6.5%) and disrupted Scenedesmus (42.8 ± 5.3%) can be due to an incomplete disruption of the biomass. The ball milling of Nannochloropsis oculata previously increased lipid digestibility from 0% (un-  Due to the stereoselectivity of pancreatic lipases, only a limited fraction of the fatty acids can be liberated. Bernaerts et al. [9] estimated that in the case of Nannochloropsis, only 55-65% of the fatty acids can be released where this maximal hydrolysis value depends on the lipid classes (triacylglycerols, phospholipids, glycolipids) present in the biomass.
To our knowledge, no literature is available on the lipid digestibility of C. nivalis and P. purpureum. Sonication experiments with Chlamydomonas reinhardtii suggested that carotenoids are already fairly bio-accessible in C. reinhardtii without pre-treatment and that sonication has no additional effect [60]. Accordingly, a high lipid digestibility was found in this study for C. nivalis (58.9 ± 11.7%).
Regarding C. vulgaris, several studies, listed in a recently published review [11], showed that carotenoid bio-accessibility can be increased by pretreating the biomass by cell disruption. Yet, an enzymatic disruption could not increase chlorophyll release in C. vulgaris biomass [61]. Similarly, a mechanical disruption treatment was unnecessary when chlorophyll bio-accessibility was considered in C. vulgaris biomass, with bio-accessibility values already ranging between 77-84% for untreated biomass [62]. No literature was found on fatty acid release during C. vulgaris digestion.
Another possible explanation is that a certain degree of lipid hydrolysis by algae lipases, for instance, immediately after harvesting or during the cell disruption treatment, is needed before digestion to facilitate complete lipid hydrolysis by pancreatic lipases. Indeed, it is known that FFA concentrations increase quickly after HPH when Nannochloropsis biomass is not immediately cooled [63]. It cannot be inferred what caused the incomplete lipid digestibility for disrupted N. gaditana in this study, but it is clear from the literature that cell disruption favors digestibility [9,10]. The same can be expected for the Scenedesmus case, since its recalcitrant cell wall is known to limit the availability of other nutrients, such as carbohydrates [27]. It can be concluded that the studied algae biomass differs significantly in terms of fatty acid composition and that differences in lipid digestibility will also contribute to differences in nutritional value.

Protein Composition and Digestibility
The highest protein levels were found in C. vulgaris and disrupted N. gaditana (45.5 ± 1.3% and 46.2 ± 0.3%, respectively on dm basis) and the lowest in P. purpureum (22.2% ± 0.1%). Amino acids essential to humans ranged between 26 and 28% of the total amino acid content (Table 3).
To examine protein digestibility, three different approaches were followed. First, the concentration of free amino acids after digestion was determined. Yet, the HPAEC profiles of digesta samples contained several interfering peaks, making reliable amino acid quantification impossible. Therefore, a second approach was followed, where the nitrogen fraction that is soluble after digestion was determined (Figure 4a).  The nitrogen solubility after digestion was the lowest for P. purpureum (51.1 ± 0.1%) and significantly higher values were noted for C. nivalis and C. vulgaris (64.5 ± 1.5% and 65.5 ± 3.4%, respectively) and disrupted Scenedesmus and disrupted N. gaditana (   The nitrogen solubility after digestion was the lowest for P. purpureum (51.1 ± 0.1%) and significantly higher values were noted for C. nivalis and C. vulgaris (64.5 ± 1.5% and 65.5 ± 3.4%, respectively) and disrupted Scenedesmus and disrupted N. gaditana (70.4 ± 1.3% and 71.5 ± 1.8%, respectively). The highest absolute soluble nitrogen levels after digestion were observed for C. vulgaris and disrupted N. gaditana (5.7 mg soluble N/100 g dry algae and 6.0 mg soluble N/100 g dry algae, respectively. The lower N solubility after the digestion of P. purpureum might be related to its high polysaccharide content [64], of which some are known to increase medium viscosity [65], possibly affecting protein solubilization and protein-protease interactions. The observed value for C. vulgaris (65.5 ± 3.4%) lies within the range found in the literature for heterotrophically cultivated C. vulgaris (43-49%) [56,66], 2 commercial Chlorella products (60-63%) [56], 11 commercial Chlorella products (mean 51 ± 9%) [67] and a commercial C. vulgaris product (76%) [12]. It should be noted that in the last study , an in vitro pig digestion model was used, while the first three studies followed the human digestion consensus protocol also used here. Slightly lower values were reported for untreated N. oceanica (50%) by Niccolai et al. [12] and for four untreated Nannochloropsis samples (48-59%) by Wild et al. [6], who also used an in vitro pig digestion protocol. Yet, Wild et al. observed a higher value after cell disruption by ball milling (78-80%). Similar values were found in this study for disrupted N. gaditana (71.5 ± 1.8%) and disrupted Scenedesmus (70.4 ± 1.3%).
Next, the degree of protein hydrolysis was assessed. It should be noted that a 100% conversion to primary amines cannot be expected, since algae contain low levels of nonprotein nitrogen, mainly as inorganic nitrogen, nucleic acids, and chlorophylls [68]. For example, in the case of C. vulgaris, 10% of all nitrogen was reported to be non-protein nitrogen [69]. When protein hydrolysis was assessed in digested P. purpureum, negative values were obtained. This was probably linked to the viscosity-enhancing effect of P. purpureum biomass, which may have affected substrate-enzyme interactions and/or hindered the complete removal of suspended particles by centrifugation. The latter biases the absorption measurement-based analysis due to light scattering from colloidal particles. The degree of protein hydrolysis of the four other algae ranged between 24.8 ± 2.6% (C. nivalis) and 30.8 ± 4.1% (C. vulgaris), with no statistical difference among samples (p = 0.11, Figure 4b). Using a similar protocol, Cavonius et al. [10] obtained a protein hydrolysis degree of ±35% for disrupted Nannochloropsis oculata, while in the untreated biomass, only 3% of the peptide bonds was hydrolyzed. Although this analysis provides only a first indication, it seems that the proteins of the algae shown in Figure 4b are fairly digestible. About 30-35% of all proteins were insoluble after digestion (Figure 4a) and thus indigestible, while 25-30% of all nitrogen was present as free amino groups after digestion (Figure 4b). It can be speculated that at least part of the remaining solu-ble fraction is also bio-accessible, since brush-border peptidases are not included in the standard digestion protocol used in this study. Brush-border peptidases are required to complete protein and peptide degradation [19] and play an important role in the activation of trypsinogen [20].

Mineral Content
Biomass mineral concentrations (P, K, Ca, Na, Cl, Fe) are listed in Table 4. For all algae, P, along with K, Na, and Cl, is always one of the main minerals. Na and Cl levels vary largely between different algae types, probably due to the use of different addition levels in the growth medium (cfr. Section 2.1, Algae Biomass) and due to the different harvesting types. Fe concentrations are relatively low (<1.5 g Fe/kg) but can still be nutritionally relevant. Indeed, when defatted Nannochloropsis oceanica containing 2.6 g Fe/kg biomass was included in pig feed at a 0.5% dosage, it alleviated the anemic status of weanling pigs [70], suggesting that the algae Fe bio-availability was high. Phytic acid (or inositol hexakisphosphate) can play an important role in this regard, since the bioaccessibility of Fe and other minerals is known to be limited by the presence of phytic acid in many staple foods. In cereals, legumes, nuts, and oilseeds, about 60-90% of total phosphorus is part of phytic acid that can bind to cations such as Ca, Fe, K, Mg, Mn, and Zn, making them insoluble and unavailable for uptake [71]. Not only phytic acid, but also other phosphate forms can form insoluble complexes with minerals such as magnesium ammonium phosphate [72]. In the present study, between 45% (disrupted Scenedesmus) and 67% (C. nivalis) of all phosphorus was present as soluble orthophosphate (PO 4 3− ) after in vitro digestion (Table 4). At least this fraction is not made insoluble and can potentially be taken up by the body. Accordingly, Wild et al. did not detect inositol phosphate isomers in 16 commercial algae samples, including Arthrospira, Chlorella, Nannochloropsis, and Phaeodactylum samples [6]. For P. purpureum, the observed soluble orthophosphate value was slightly lower than that of the control. Again, this is possibly linked to the viscosity enhancing effect of P. purpureum biomass, as already mentioned in the discussion of the protein hydrolysis results (Section 3.4).

Bioactivity Screening-ACE-1 Inhibition
ACE-1 is known to regulate blood pressure and the salt-water balance within many mammalian species. ACE-1 inhibitors work within the renin angiotensin aldosterone system (RAAS). The kidneys produce renin in response to low blood volume and low sodium and high levels of potassium. The enzyme renin, which is the rate limiting step in the RAAS, acts on angiotensinogen and forms the vasodilator Angiotensin I. The enzyme ACE-1 converts Angiotensin I to Angiotensin II, a vasoconstrictor, and this action causes high blood pressure. The inhibition of ACE-1 prevents the formation of the vasoconstrictor Angiotensin II and subsequent high blood pressure or hypertension. An ACE-I inhibition bioassay was performed on the algal biomass. As shown in Figure 5, all microalgae tested inhibited ACE-1 by between 73.4% and 87.1% when assessed at a concentration of 1 mg/mL compared to the positive control Captopril©, which inhibited ACE-1 by 93.48% when tested at a concentration of 0.05 mg/mL. Disrupted N. gaditana biomass was found to inhibit ACE-1 by 87.1%, compare control. C. nivalis, which inhibited ACE-1 least significantly (73.4%). The percenta 1 inhibition values observed in this study are greater than values reported previo Spirulina sp. (47.6% ACE-1 inhibition) [13]. ACE-1 is a key enzyme that helps to the salt-water balance and blood pressure. The inhibition of ACE-1 is a bioactiv could be exploited in the development of functional foods for the maintenance of tensive blood pressure. Algae proteins are likely to be responsible for ACE-1 in since high ACE-1 inhibition activity was also observed with protein hydroly amongst others C. vulgaris and N. oculata, as reviewed by Bleakley and Hayes [64 Although the ACE-1 inhibition assay is an in vitro method, the inhibition o was previously found to directly relate to an observed antihypertensive effect in studies. For example, Fitzgerald et al. [73] previously identified an ACE-1 inhibito capeptide from the macro-alga Palmaria palmata. This peptide and the whole a duced an antihypertensive effect in spontaneously hypertensive rats, when assaye same concentrations [74]. The promising data of the current study now justify con the necessary experiments to prove anti-hypertensive effects in vivo.

Conclusions
This study aimed to make an overview of the nutritional characteristics of micro-algae that can be cultivated in the same climatic area. Large differences in matter solubility after digestion suggested that for N. gaditana and Scenedesmus b a prior cell disruption step is needed to avoid the majority of nutrients not being d To our knowledge, this is the first study indicating that significant amounts of free can be present after the digestion of C. nivalis, P. purpureum, and disrupted Scen which will contribute to the energy content of the biomass. Moreover, it confir micro-algae are rich in both essential fatty acids and amino acids, though both co Disrupted N. gaditana biomass was found to inhibit ACE-1 by 87.1%, compared to the control. C. nivalis, which inhibited ACE-1 least significantly (73.4%). The percentage ACE-1 inhibition values observed in this study are greater than values reported previously for Spirulina sp. (47.6% ACE-1 inhibition) [13]. ACE-1 is a key enzyme that helps to regulate the salt-water balance and blood pressure. The inhibition of ACE-1 is a bioactivity that could be exploited in the development of functional foods for the maintenance of normotensive blood pressure. Algae proteins are likely to be responsible for ACE-1 inhibition, since high ACE-1 inhibition activity was also observed with protein hydrolysates of amongst others C. vulgaris and N. oculata, as reviewed by Bleakley and Hayes [64].
Although the ACE-1 inhibition assay is an in vitro method, the inhibition of ACE-1 was previously found to directly relate to an observed antihypertensive effect in animal studies. For example, Fitzgerald et al. [73] previously identified an ACE-1 inhibitory tridecapeptide from the macro-alga Palmaria palmata. This peptide and the whole alga produced an antihypertensive effect in spontaneously hypertensive rats, when assayed at the same concentrations [74]. The promising data of the current study now justify conducting the necessary experiments to prove anti-hypertensive effects in vivo.

Conclusions
This study aimed to make an overview of the nutritional characteristics of a set of micro-algae that can be cultivated in the same climatic area. Large differences in organic matter solubility after digestion suggested that for N. gaditana and Scenedesmus biomass, a prior cell disruption step is needed to avoid the majority of nutrients not being digested. To our knowledge, this is the first study indicating that significant amounts of free glucose can be present after the digestion of C. nivalis, P. purpureum, and disrupted Scenedesmus, which will contribute to the energy content of the biomass. Moreover, it confirms that micro-algae are rich in both essential fatty acids and amino acids, though both concentrations and the degree of digestibility differ among species. The inhibition of ACE-1 was observed for all algae at physiologically relevant concentrations when assayed in vitro against the known ACE-1 inhibitor Captopril ® . However, future work should include the identification of the active compound responsible for these observed bioactivities (fatty acids, peptides) and confirmation of the potential antihypertensive effect in vivo using an animal model, such as spontaneously hypertensive rats. The results of this study can support algae cultivators and processors in selecting the most appropriate algae species for a particular food or feed application.