Assessment of Arabian Gulf Seaweeds from Kuwait as Sources of Nutritionally Important Polyunsaturated Fatty Acids (PUFAs)

The fatty acid (FA) compositions of ten seaweeds representative of Chlorophyta, Rhodophyta, and Ochrophyta from Kuwait in the Arabian Gulf region were determined and are discussed in the context of their potential nutritional perspectives for seaweed valorization. All the seaweeds had higher saturated fatty acid (SFA) and lower monounsaturated (MUFA) and polyunsaturated fatty acid (PUFA) contents than those typical of tropical environments. Palmitic, myristic, stearic, oleic, linoleic, α-linolenic, and stearidonic acids were the major FAs detected. Arachidonic, eicosapentaenoic, and docosahexaenoic acids were detected in minor amounts. Conserved fatty acid patterns revealed phylogenetic relationships among phyla, classes, and orders matching the molecular phylogenies at higher taxonomic ranks. Hierarchical clustering analyses clearly segregated different seaweeds (except Codium papillatum and Iyengaria stellata) into distinct groups based on their FA signatures. All but one species (Chondria sp.) had health-beneficial n6/n3 PUFAs (0.33:1–2.94:1) and atherogenic (0.80–2.52) and thrombogenic indices (0.61–5.17). However, low PUFA/SFA contents in most of the species (except Ulva spp.) may limit their utilization in the formulation of PUFA-rich functional foods. Ulva spp. had substantially high PUFAs with PUFA/SFA > 0.4, n6/n3 (0.33–0.66) and atherogenic (0.80–1.15) and thrombogenic indices (0.49–0.72), providing substantial potential for their utilization in food and feed applications.


Introduction
Seaweeds are photosynthetic, multicellular marine macroalgae that have been utilized for food, animal feed, phycocolloids, and bioactive compounds of pharmacological importance for centuries. In fact, they are considered one of the most important food sources for the coastal communities especially in Asian countries such as Japan, China, and Korea [1]. Today the global seaweed industry is worth USD 6 billion per annum, 85% of which comprise food products for human consumption [1]. These seaweeds are rich sources of essential nutrients and health-promoting compounds including proteins, carbohydrates, polyunsaturated fatty acids (PUFAs), antioxidants, minerals, dietary fibers, and vitamins [2][3][4][5]. As a matter of fact, it is often pointed out that the Japanese, who have eaten seaweeds regularly in their daily cuisines for centuries, have one of the highest life expectancies in the world [6]. Seaweed-digesting enzymes such as porphyranases and agarases were discovered in the Japanese gut bacteria a decade ago, but were absent from American populations [7]. Moreover, increasing awareness of beneficial impacts

Fatty Acid Extraction and Methyl Ester Preparation
Fatty acids were extracted and converted into the respective methyl esters from freeze-dried samples by the base-catalyzed direct transmethylation method modified after Christie and Han [40]. Briefly, 0.3 g of freeze-dried seaweed samples (in triplicates) were homogenized in a mortar and pestle and transferred to Oakridge™ centrifuge tubes (15 mL), to which 3 mL of KOH-MeOH solution (0.2 M) was added. The mixture was heated at 75 • C for 1 h. After cooling to room temperature, 3 mL of n-hexane was added and mixed thoroughly using a vortex. The organic layers containing fatty acid methyl esters (FAMEs) were collected in GC vials and stored at −20 • C until analysis.

Gas Chromatographic (GC) Analysis
For analysis of FAMEs, 1 µL of esterified sample was injected into a gas chromatograph (Shimadzu GC 2010, Tokyo, Japan) coupled with a flame ionization detector (FID). A cyanopolysiloxane (CP-Sil 88 for FAME, part number 839171) capillary column (100 m × 0.25 mm, 0.20 µm (J&W, Varian, Chrompack, São Paulo, Brazil) was used for the FAMEs separation under the following instrumental conditions: injector and FID detector temperatures were 250 and 270 • C, respectively, with an injector split ratio of 1:50, and carrier gas helium with a constant flow rate of 1.0 mL/min. The initial oven temperature was 80 • C, at which it was held for 5 min, followed by an increase to 220 • C at a rate of 4 • C/min; then, it was held for 5 min, and finally the temperature was increased to 240 • C at a rate of 1 • C/min and was held for an additional 10 min. FAME peaks were identified by comparison of their retention times with those of external standard (FAME Mix C4-C24; Sigma-Aldrich, Laramie, WY, USA) and quantified by area normalization using postrun analysis, GC LabStationsTM software v. 5.96 (Shimadzu, Tokyo, Japan). The content of individual fatty acid was finally reported as relative percentage of the total fatty acid methyl esters (TFAs).

Statistical Analysis
All analytical determinations were performed in triplicate (n = 3) and the mean values were recorded. The fatty acid contents of different seaweed species were compared by analysis of variance (ANOVA) followed by Tukey's HSD post-hoc test with differences considered significant at p < 0.01 using SPSS v 22. All multivariate analyses were performed after log-transformation and pareto scaling (mean-centered and divided by the square root of standard deviation of each value) of FA and nutritional data matrices (Supplementary datasheet S1-S2) using the web-based software Metabo-Analyst v 5.0 (https://www.metaboanalyst.ca accessed on 26 September 2021). This data pre-processing was carried out to give equal weight to all variables, regardless of their absolute value as the detected fatty acid levels were of different orders of magnitude. The principal component analysis (PCA) was performed on data matrices without rotation and the principal components were extracted based on scree plot. The dendrogram was obtained by hierarchical clustering based on Ward linkage with Euclidean distance [42]. Additionally, the normalized data matrices obtained in MetaboAnalyst were exported to SPSS v 22 for Kaiser-Meyer-Olkin (KMO) test for measuring sampling adequacy for PCA analysis and Bartlet's test of sphericity to assess the equality of variance in the data matrices.
Overall, a substantial variation was observed in the individual FA contents of the same and related species of the same genus among all green, red, and brown seaweeds in our study, which is also reflected in the literature from different regions of the world [4,24,39,45,47,48,52]. These variations are due to species-specific variations, different geographical locations, and environmental factors (temperature, light, salinity, nutrients) [25,28,30,46,59]. Thus, it becomes necessary to screen different seaweeds (both wild and cultivated) from different regions for their FA contents and to monitor them across different seasons to determine the suitable period of harvest for seaweed valorization. The effect of different seasons or other environmental factors on FA composition of seaweeds were not studied in the present study but will be an objective of our future research. Additionally, there can be variations in FA contents of same or related species in literature due to different extraction and derivatization methods employed by researchers [25,60], but it is beyond the scope of this study to compare such FA variations.

Fatty Acid Chemotaxonomy
Hierarchical clustering was performed on the FA data matrix (Supplementary datasheet S1) to evaluate the chemotaxonomic relationships between different species at different taxonomic levels. A few FAs, namely, 10:0, 13:0, 24:0, 9c-14:1, 10c-15:1, 10c-17:1, 9t-18:1, 11c-20:1, 13c-22:1, 15c-24:1, 9t12t-18:2, 13c16c-22:2, and 11c14c17c-20:3 were excluded from this FA data matrix due to their insignificant amounts and lack of correlation with the data matrix since such variables often lead to misclassification of species. The dendrogram obtained from Ward hierarchical clustering grouped the seaweed samples into three demarcated clusters (Figure 1). Thus, our study displayed that FA traits are conserved in seaweeds at higher ordinal levels of families, orders, and phyla, in line with the previous findings [4,11,14,16,17,45]. FA signatures could be potential tools for understanding the chemotaxonomic relationships among different seaweed species, but require proper sampling. Otherwise, higher variations in FA contents at the levels of genus or species may pose difficulty in discriminating species in the absence of adequate taxon sampling and replicates, as observed in Ward linkage is an agglomerative clustering algorithm which starts with n singleton clusters (each consisting of one element of the data set) and merges two clusters based on similarity measure. All of the Ulva species (Ulvales) were grouped together in group I while C. papillatum (Bryopsidales) was grouped with I. stellata (Ectocarpales) in group II. The single red alga investigated in this study, Chondria sp., belonging to Ceramiales, was grouped together with brown seaweeds P. boergesenii (Dictyotales), S. aquifolium (Fucales), and F. indica (Ectocarpales) in group III. Kumari et al. [14] also showed that Bryopsidales are grouped separately from Ulvales and the latter generally aligns with Ulotrichales, forming the Ulvales-Ulotrichales clade [14]. However, a greater number of replicates as well as species belonging to the genus Codium and Iyengaria are required to resolve their misclassification based on FAs, as observed in our study. Further, group I can be subdivided into two sub-groups, consisting of U. tepida in one, and U. ohnoi, Ulva sp. and U. chaugulii in another sub-group. Similarly, group III can be further sub-divided into three sub-groups, the first comprising of F. indica and S. aquifolium, the second of P. boergesenii, which was closely related to Chondria sp., forming the third sub-group. Similarly, the species belonging to the genera Padina and Sargassum were grouped in different sub-groups based on their FA profiles [11,14,45] as well as different clades based on their molecular data [61]. However, for adequate comparison of inter-relationships between different groups deduced from FA composition with the clades inferred from genomic data, extensive sampling effort with samples belonging to the same genus as well as same class or orders are imperative.
Thus, our study displayed that FA traits are conserved in seaweeds at higher ordinal levels of families, orders, and phyla, in line with the previous findings [4,11,14,16,17,45]. FA signatures could be potential tools for understanding the chemotaxonomic relationships among different seaweed species, but require proper sampling. Otherwise, higher variations in FA contents at the levels of genus or species may pose difficulty in discriminating species in the absence of adequate taxon sampling and replicates, as observed in our study for C. papillatum and I. stellata.

Nutritional Assessment for Seaweed Valorization
Our study revealed that Ulva species are rich sources of nutritionally important PUFAs with their unsaturation indices (UI) varying from 119.21 ± 0.45 (U. ohnoi) to 133.28 ± 1.65 (U. tepida) ( Table 2) in congruence with the UI values reported in the literature for different species of the genus Ulva [11,14,62]. The UI values for all other species in our study were low, varying from 28.63 ± 0.84 (Chondria sp.) to 78.91 ± 0.35 (F. indica) in agreement with lower PUFA contents in these species.
Further, we conducted a principal component analysis (PCA) (without rotation) and hierarchical clustering based on and the nutritional indices data matrix (Supplementary datasheet S2) to identify potential seaweeds that can be valorized for nutritional and functional food applications. We obtained a KMO value of 0.726 and a significant level for the Bartlett's test (Supplementary Table S1) for the nutritional indices data matrix, suggesting that nutritional indices variables were highly correlated.
The principal components were extracted based on scree plot (Supplementary Figure S1) and the first two principal components, which also presented the maximum explained variance, were used for generating scores and loading plot. PCA of nutritional indices data matrix explained 98.4% of variations (PC1-94.4% and PC2-4%) (Figure 2a). The discriminant variables along PC1 were PUFA/SFA, UI, and TI, and along PC2 were AI and n6/n3 PUFA (Figure 2b). The loadings plot displayed that PUFA/SFA and UI were highly positively correlated, while both these were negatively correlated with TI. Similarly, AI was negatively correlated with n6/n3 PUFAs. Further, PC2 (Y-axis) separated all the Ulva species from brown and red seaweeds owing to their higher loadings of PUFA/SFA and UI, while C. papillatum was positioned along with the brown seaweeds due to its lower contents of UI and PUFA/SFA. S. aquifolium was found to be the outlier, separated from rest of the brown seaweeds by X-axis due to its high loadings of n6/n3 PUFAs. Chondria sp. was separated from the rest of green and brown seaweeds due to its higher loadings of TI and AI in line with its exceptionally high TI and AI contents ( Table 2). species from brown and red seaweeds owing to their higher loadings of PUFA/SFA and UI, while C. papillatum was positioned along with the brown seaweeds due to its lower contents of UI and PUFA/SFA. S. aquifolium was found to be the outlier, separated from rest of the brown seaweeds by X-axis due to its high loadings of n6/n3 PUFAs. Chondria sp. was separated from the rest of green and brown seaweeds due to its higher loadings of TI and AI in line with its exceptionally high TI and AI contents ( Table 2).  The dendrogram obtained from hierarchical clustering of nutritional indices data revealed three demarcated clusters (Figure 2c). All of Ulva spp. (containing high UI and PUFA/SFA) were clustered together in Group I, like the Ulvales clade deduced from the FA data matrix (Figure 1). Contrary to our previous results, where Chondria sp. was grouped with other brown seaweeds in group III (Figure 1), here, Chondria sp. formed a separate clade, group II. C. papillatum was grouped with brown seaweeds in group III, sharing the sub-clade with I. stellata and P. boergesonii.
PUFAs are essential biomolecules to human health since their consumption is associated with decreased risk of cardiovascular and inflammatory diseases as well as cancer [18][19][20]63,64]. 9c12c15c-18:3 is a precursor of 5c8c11c14c17c-20:5 as well as 4c7c10c13c16c19c-22:6 and has anticancer, antiosteoporotic, antioxidant, anti-inflammatory, as well as coronary and neuronal protective effects [65]. 4c7c10c13c16c19c-22:6 is essential for visual and neurological development in infants while 5c8c11c14c-20:4 and 5c8c11c14c17c-20:5 are precursors of prostaglandins, thromboxanes, and other eicosanoids that influence inflammation processes and immune reactions [18,63]. Free PUFAs also have biological effects including induction of an oxidative burst, oxylipin biosynthesis, and induction of resistance against pathogens in seaweeds such as the brown algal kelps Laminaria digitata and Macrocystis pyrifera [66]. The PUFA/SFA ratio, which is an important parameter to assess the nutritional quality of the lipid fraction of food, should be ≥0.4 [67]. In this study, the PUFA/SFA values were in accordance with the nutritional guidelines only for Ulva species (0.57-0.80). However, much higher PUFA/SFA values (≥0.4) have been reported for species of the genera Ulva, Codium, Sargassum, and Padina in previous reports [4,14,15,24,25,34,43,46,49]. The low PUFA/SFA ratio in our study may be due the warm environment of the Arabian Gulf, in agreement with the reports that seaweeds of temperate regions tend to feature a higher degree of unsaturation in their fatty acid composition [46,59,62,68]. High SFA content in tropical seaweeds may be related to their physiological adaptation to warm temperatures, while high PUFA content in cold water may facilitate thermo-adaptive regulation of membrane lipid fluidity [59,69]. All the species investigated in the present study had health-promoting n6/n3 ratios ranging from 0.33 ± 0.02:1 (Ulva sp.) to 2.94 ± 0.03:1 (S. aquifolium) ( Table 2) in line with the World Health Organization (WHO) recommendations of an n6/n3 ratio of 5:1 [19,20,70]. The atherogenic indices (AI) varied from 0.8 ± 0.01 (U. tepida) to 2.52 ± 0.04 (P. boergesenii), while thrombogenic indices (TI) varied from 0.49 ± 0.01 (U. tepida) to 5.17 ± 0.15 (I. stellata), except for Chondria sp., which had higher AI and TI values ( Table 2). Low AI and TI < 3 have been reported for different green, red, and brown seaweeds in the literature [2,14,24,25,49]. Recently, Chen and Liu [23] compared the nutritional indies of numerous seaweeds reported in the literature with those of plant oils, fish, and dairy products. Accordingly, AI and TI values obtained in our study for all seaweeds except Chondria sp. (Table 2) were comparable to those of fish (AI-0.37-1.22, TI-0.14-0.87), shrimps (AI-0.71-0.82, TI-0.21-0.30), and dairy products (AI-1.42-5.13, TI-0.39-5.04) [23]. There is no recommended level of AI and TI in food products, but the consumption of foods with low AI and TI indices is helpful in reducing the risk of coronary heart diseases [23].
Overall, most of the seaweeds investigated in our study had health-beneficial n6/n3, AI, and TI values, but only Ulva spp. had higher UI and the recommended PUFA/SFA ratio. The multivariate analysis of nutritional indices clearly supported our findings and helped in assessing the nutritional potential of seaweeds from Arabian Gulf. Nevertheless, Ulva spp. are not only rich in essential PUFAs, but also contain high amounts of macroand micronutrients as reported previously [71]. In addition, the nutritional value of Ulva species in terms of carbohydrates, protein, and fatty acids (especially PUFA content) has been reported to be comparable to some vegetables, nuts, and grains [2,23,72] and it has been consumed traditionally for centuries in many Asian countries [1,2,4,25,72].

Final Conclusions
Our study revealed that seaweeds from the Arabian Gulf exhibit typical FA profiles of warm waters with relatively high SFA and low PUFA contents. The green, red, and brown seaweeds exhibit species-specific significant differences in FA contents, but trends of FA profiles were conserved at different taxonomic ranks of genus, class, and order within each phyla. Among all the species investigated, Ulva spp. are the most suitable candidates for developing low-fat foods with PUFA-rich nutraceuticals or utilization in functional food for human consumption and animal feed due to their health beneficial PUFA/SFA, n6/n3, AI, and TI values. However, proper valorization of Ulva species for commercial utilization will require a temporal, spatial, and seasonal consistency in FA contents. Future studies for understanding the environmental and seasonal impacts on FA profiles of Ulva spp. from the Kuwait region will facilitate selecting the correct harvest time for obtaining high PUFA yields.