CHEMICAL COMPOSITION , ANTIOXIDANT ACTIVITY AND 1 SENSORY EVALUATION OF FIVE DIFFERENT SPECIES OF 2 BROWN EDIBLE SEAWEEDS . 3

21 The chemical and volatile composition as well as sensory profile of five brown edible 22 seaweeds collected in the United Kingdom, was evaluated. The ash content was 190‒ 23 280 mg/g, NaCl 35.1‒115.1 mg/g, protein 2.9‒6.0 g/g, and fat 0.6‒5.8 g/g (dry basis). 24 Fucus vesiculosus, Fucus spiralis. and Ascophyllum nodosum showed higher 25 antioxidant activities (DPPH and FRAP). Nucleotide concentrations were of the same 26 order of magnitude as reported in other foods such as tomatoes or potatoes, except for 27 Fucus vesiculosus where levels of nucleotides were 10 times higher. The fatty acids 28 profile was dominated by oleic acid (21.9‒41.45 %), followed by myristic (6.63‒26.75 29 %) and palmitic (9.23‒16.91 %). Glutamic and aspartic acid (0.15‒1.8 mg/g and 0.05‒ 30 3.1 mg/g) were the most abundant amino acids. Finally, sensory and volatile analyses 31 illustrated that Laminaria sp. had the strongest seaweed and seafood-like aroma and 32 taste. 33 34


Introduction. 44
Due to their low content of lipid, high concentration of polysaccharides, natural richness 45 in minerals, polyunsaturated fatty acids and vitamins as well as their high content of 46 bioactive molecules, marine algae have, in recent years, received great attention (Gupta 47 Particularly, in the UK, the market for seaweed (therapeutic, biotechnology, bio-fuel 69 seaweeds based, or foods) is mostly imported, whereas there is abundance of growing 70 seaweeds around the islands, with some local producers already harvesting them for 71 commercial purposes. Particularly, in the coast of Scotland there are dozens of different 72 kinds of edible seaweed, being the red seaweed dulse (Palmaria palmata), as well as the 73 brown seaweeds: kelp (Laminaria sp.) and different wracks (Fucus sp., Ascophyllum 74 nodosum, Pelvetia canaliculata) the most generally harvested (due to their abundance 75 and accessibility). 76 The use of brown seaweeds, as ingredient or as a whole food, has already been reported 77 by numerous authors to be beneficial in different aspects. For instance, as an alternative 78 source of protein, with some brown species having higher protein content than 79 soybeans. Their fat content accounts for 1 to 6 g/100 g dry weight with some varieties, 80 as Laminaria sp. generally between 1.5 and 3.3% of dry weight (Fleurence, Gutbier, 81 Mabeau, & Leray, 1994), and some of these species are also characterised by a high 82 level of eicosapentaenoic acid (up to 24% of the total fatty acid fraction) (Fleurence, 83 2004). Antioxidants are also other important metabolites in brown seaweeds including 84 fucoxanthin, polyphloroglucinol, phenolic compounds or bromophenols, that have been 85 isolated from species such as Fucus and Laminaria (Xu et al., 2004a;2004b;Gupta & 86 Abu-Ghannam, 2011b; Fleurence et al., 2012) 87 In addition, there are recent projections in the functional effects of seaweeds as means 88 to improve the fibre content and reduce the salt content of food products. This is mainly 89 due to their high content in umami compounds such as nucleotides or some amino acids. 90 The aim of this study was to characterise five different brown edible seaweeds locally 91 produced on the west coast of Scotland (Isle of Bute), UK, in terms of chemical composition as well as sensory and volatile analyses; this information might be useful to 93 evaluate their use as food ingredients and their potential contribution to the diet. 94

Fatty acids. 118
The fatty acid composition was analysed by GC-FID after transesterification to methyl 119 esters (FAMEs) with a mixture BF 3 methanol at 20 ºC according to the IUPAC standard 120 method (IUPAC, 1992, Yaich et al., 2011. 121 Fat (10 mg), hexane (0.2 mL) and BF 3 (0.5 mL) were heated at 70 ºC for 1.5 h. After 122 transesterification, saturated salt solution (0.5 mL, 25 % NaCl), H 2 SO 4 (0.2 mL, 10%) 123 and hexane (7 mL) were added to the reaction medium. For the DPPH and FRAP assay calibration curves of Trolox (0-1000 mM) were 155 prepared and results were expressed as the number of equivalents of Trolox (mmol eq of 156 Trolox/g dry weight). Gallic acid (0-1000 mM) was used for TPC and results expressed 157 as the number of equivalents of gallic acid (mmol eq of gallic acid/g dry weight of 158 seaweed powder). . The sensory attributes studied, which 217 had been previously described by 4 assessors, were: honey-like odour, herbal odour, 218 seaweed-like odour, seafood-like taste, saltiness, astringency, bitterness, green tea-like 219 taste, and salmon-like taste. 10 mL of each seaweed extract at room temperature was 220 served to each panellist. Continuous non-structured scales were used for evaluation. The 221 left side of the scale corresponded to the lowest intensity (value 0) and the right side to 222 the highest intensity (value 10). Each panellist rinsed their mouth with mineral water 223 and ate a piece of plain cracker between samples. 224

Statistics 225
Analysis of variance (ANOVA) and the Friedman test (p-value < 0.05) were carried out 226 using SPSS to estimate the differences in composition of the seaweed varieties 227 investigated in this study.

Antioxidant activity 255
The antioxidant activity of the ethanolic extracts of the seaweed samples was analysed 256 by two different methods to accurately reflect all the antioxidants in the samples (Table  257 1). The FRAP reagent can react with iron (II) and thiol groups (Benzie & Szeto, 1999), 258 while DPPH is expected to react with organic radicals (Chandrasekar, Madhusudhana, 259 Ramakrishna & Diwan, 2006). The values for the total phenolic content are also 260 presented in Table 1 (mmol equivalents of gallic acid/g dry weight). The estimation of 261 the antioxidant potential using different methods enables a better understanding of the 262 mechanism(s) of antioxidative action of the seaweed extracts. 263 Table 1.

265
Composition of the seaweed samples: moisture (x w %), ash (% dry weight), NaCl (mg / g dry weight), 266 protein (g / g dry weight) and fat content (g / g dry weight), antioxidant activity (DPPH and FRAP 267 mET/100g of dry weight), total phenolic content (mEG /100g of dry weight), fatty acids composition 268 (g/100g of total fat), and homogeneous groups obtained from the statistical analysis for the different 269 species of seaweeds and the different batches used (n=3).

Fatty acid composition 316
The fatty acid composition of the two batches of seaweed samples is given in Table 1.

Free amino acids, nucleotides and umami contribution 338
The free amino acid composition (mg/ g of dry weight) is illustrated in Table 2. It is 339 important to point out, the high alanine content in the seaweeds collected in July and 340 August of L. digitata (4.1 ± 0.2 mg/ g of dry weight) compared to those collected earlier 341 for the same species, but also compared to the others. Glutamic acid was particularly 342 high in P. canaliculata and F. spiralis, while aspartic acid was the highest amino acid in 343 F. spiralis. 344 345  The nucleotide composition (μg/ g of dry weight) for the five seaweeds samples is given 362 in Table 2. These values ranged from 0.20 ± 0.02 to 364.3 ± 13.2 µg/ g of dry weight, 363 and were of the same order of magnitude as reported in other foods such as tomatoes, 364 potatoes or some varieties of mushrooms (60 to 300 µg / g of dry weight) (

Volatiles analysis 381
A total of 23 compounds were detected and identified in the aqueous extracts of the 5 382 seaweeds. Volatile compounds identified in the different seaweed samples are presented 383 in Table 3 and can be classified as aldehydes, alcohols, esters, ketones, acids and 384 aromatic compounds. Five key compounds, (hexanal, heptanal, nonanal, 1-octen-3-ol 385 and 2,4-heptadienal), which have previously been described as giving rise to fishy notes 386 (Ganeko, et al., 2008; Giri, Osako & Ohshima, 2010) were studied in more detail. They 387 were quantified using external calibration curves and the Friedman test was applied to 388 study any differences in their concentrations between the aqueous seaweed extracts 389 (Fig. 1).

437
Despite the fact that Laminaria showed the highest score for saltiness, as could be 438 expected due to its high concentration in NaCl compared to the other seaweeds, the 439 difference was not significant. The results suggest that the panellists did not associate 440 umami taste with seafood taste or seaweed aroma, as Laminaria had the lowest EUC 441 (Table 2). This could be due to the assessors used were untrained subjects unfamiliar 442 with the characteristics of the typical umami taste, however, this type of panel has 443 previously been used for that kind of assessment and though the performance of the 444 untrained panels would not be as good as if they had been trained, they were able to distinguish 445 between samples, (Claperton and Piggott, 1979; Husson & Pagés, 2003)). Therefore its 446 sensory attributes could be mainly due to its high salt content together with high levels 447 of the volatile compounds, hexanal, heptanal, nonanal and 2,4-heptadienal.