Active aroma compounds assessment of processed and non-processed micro- and macroalgae by solid-phase microextraction and gas chromatography/mass spectrometry targeting seafood analogs

TYPE Original Research PUBLISHED 26 October 2022 DOI 10.3389/fsufs.2022.1011020 OPEN ACCESS EDITED BY Guadalupe Virginia Nevárez-Moorillón, Autonomous University of Chihuahua, Mexico REVIEWED BY Raul Avila-Sosa, Meritorious Autonomous University of Puebla, Mexico Adriano Gomes Cruz, Adriano Gomes da Cruz, Brazil *CORRESPONDENCE Cristina M. R. Rocha [email protected] SPECIALTY SECTION This article was submitted to Sustainable Food Processing, a section of the journal Frontiers in Sustainable Food Systems RECEIVED 03 August 2022 ACCEPTED 15 September 2022 PUBLISHED 26 October 2022 CITATION Moreira C, Ferreira-Santos P, Teixeira JA and Rocha CMR (2022) Active aroma compounds assessment of processed and non-processed microand macroalgae by solid-phase microextraction and gas chromatography/mass spectrometry targeting seafood analogs. Front. Sustain. Food Syst. 6:1011020. doi: 10.3389/fsufs.2022.1011020 COPYRIGHT ©2022 Moreira, Ferreira-Santos, Teixeira and Rocha. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Active aroma compounds assessment of processed and non-processed microand macroalgae by solid-phase microextraction and gas chromatography/mass spectrometry targeting seafood analogs Catarina Moreira1,2, Pedro Ferreira-Santos1,2, José António Teixeira1,2 and Cristina M. R. Rocha1,2* 1CEB-Centre of Biological Engineering, University of Minho, Braga, Portugal, 2LABBELS—Associate Laboratory in Biotechnology, Bioengineering and Microelectromechanical Systems, University of Minho, Braga, Portugal The odor of four algae was investigated and compared to evaluate the potential of these algae to mimic shrimp aroma. Solid-phase microextraction followed by gas chromatography analysis coupled with sensory analysis was used for performance assessment. The volatile organic compounds were determined in non-processed, raw samples (r), and processed cooked (c) and cooking water (w) samples for two microalgae [Nannochoropsis oceanica (NO) and Tetraselmis chuii (TC)], two macroalgae [Ulva rígida (UR) and Saccharina latíssima (SL)], and shrimp Vannamei cong (SH). The results showed significant differences in the composition of volatile compounds between macroalgae and microalgae. The key odorants in macroalgae were octanal, 2-octenal, nonanal, and β-ionone, and in microalgae were 1,5-octadien-3-ol, hexanal, 2,4-decadienal, 2-octenal, octanal, nonanal, 3,5-octadien-2-one, and terpenes. The PCA analysis of GC-MS data showed odor similarities between the studied samples, which were divided into five main groups: (1) TC(c) and TC(w); (2) TC(r) and NO(c); (3) NO(r), NO(w), and SL(w); (4) SL(c), UR(r), UR(c), and UR(w); and (5) SL(r). The data from the sensory analysis show bigger similarities between the macroalgae and the shrimp odor. Overall, the data provided indicate that the cooking water and cooked samples are very similar in key components of odorants. These features allow the possibility to use algae and their processed resulting products as a shrimp flavor replacement in non-animal-based food formulations, thus decreasing Frontiers in Sustainable Food Systems 01 frontiersin.org Moreira et al. 10.3389/fsufs.2022.1011020 the pressure on seafood crops and aquaculture-associated issues leading to more sustainable livestock. Furthermore, circularity and waste reduction may be further enabled by the use of otherwise wasted cooking water as an odorant agent. KEYWORDS algae, GC-MS, odorants, solid-phase microextraction, volatile compounds, seafood processing, sensory evaluation, sustainability Introduction The Food and Agriculture Organization (FAO) claims that global meat production is expected to double by 2050, due to the increase in world population which is estimated to reach 10 billion people within the next 30 years. This fact makes the search for new alternative proteins an objective of the utmost importance (Dopelt et al., 2019). Due to the enormous increase in the world population, awareness of animal welfare, health, and ecological impact, the study and investment in plant-based alternatives for meat and dairy products have been rapidly increasing (van Vliet et al., 2020), yet fish and seafood plantbased alternatives have been neglected (Koyande et al., 2019). Aquatic animal food consumption, such as finfish, mollusks, crustaceans, and cephalopods, provides 17% of human dietary intake of animal protein, although large disparities are ostensible across regions (FAO, 2018). Yet, dietary guidelines recommend fish as part of sustainable healthy diets, up to 28 g/d (Willett et al., 2019). It is known that dietary fish consumption has been linked to human health outcomes associated with the prevention of stroke and cardiovascular disease, as well as fish consumption during pregnancy is linked to the neurodevelopment of offspring (Iannotti et al., 2021). The replacement of meat and fish proteins for plant proteins is one promising strategy to reduce meat consumption, although most of the products are not well-accepted by meat consumers due to their low sensory appeal. Due to this issue, industries have been moving toward the development of plant-based products with meat-like sensory attributes, commonly referred to as meat analogs (Fiorentini et al., 2020). Microand microalgae are gaining an increasing interest in the EU bioeconomy. Besides being ecological, vegan, and healthy products (Geada et al., 2021), they also have an interesting nutritional profile: microalgae protein can represent up to 70% of total algae biomass and in macroalgae up to 22% (Bleakley and Hayes, 2017). Although algae-based products are trendy and dozens of food products have been already launched into the market in the food sector, one of the major reported issues is their undesirable strong marine taste and odor (Koyande et al., 2019;Lafarga, 2019,2020). It is proven that flavor can be the decisive factor in the choice of a particular product by the consumer, and algae’s odor can act as a pro or con in decision-making. The classic and characteristic fishy and seafood algae’s odor comes from a complex mixture of different volatile organic compounds (VOCs) and their concentrations. These products are mainly composed of polyunsaturated fatty acids, aldehydes (e.g., hexanal, 2-heptenal, 2-octenal, and 2,6-nonadienal), alcohols (e.g., 3,5-octadien-2-ol and 1-octen-3-ol), and ketones (e.g., 1-octen-3-one and 3,5-octadien-2-one), and also sulfur compounds and nitrogen-containing compounds, such as trimethylamine, although there can be significant differences between the algal species (Varlet and Fernandez, 2010;Isleten Hosoglu, 2018;Coleman et al., 2022). VOCs are molecules with high vapor pressure, moderate hydrophilicity, and low molecular weight (Garicano Vilar et al., 2020), and are produced by algae depending on the species, process, culture, and environmental conditions. Besides these VOCs, the odor activity value (OAV) also plays a major role in identifying the key odorants because it estimates the importance of a flavor compound in food, based on the ratio of its concentration to its odor threshold concentration in that food. This means that only compounds which exceed the threshold level in food will impact the flavor and thus sensory perception. Therefore, the larger the OAV, the greater the referred compound will contribute to the overall odor (Garicano Vilar et al., 2020). Although algae have been reported to be used as a flavor ingredient in plant-based seafood alternatives and some studies have found similar VOCs between algae and seafood (Coleman et al., 2022;Moran et al., 2022), to the best of the authors’ knowledge, there are no studies on the VOCs present in algae in different matrices resulting from food processing, such as raw, cooked, and cooking water samples. Therefore, the objective of this study was to investigate the potential of microand macroalgae to be used as flavor ingredients in plant-based shrimp alternatives. The study focused on cooked algae and/or their cooking water to benchmark with shrimp-based foods, and their VOCs were compared with the freeze-dried matrix. Two microalgae [Nannochoropsis oceanica (NO) and Tetraselmis chui (TC)] and two macroalgae [Ulva rígida (UR) and Saccharina latíssima (SL)] were studied to determine the VOCs present in the processed and non-processed samples and their similarity to shrimp odor was analyzed using a combination of sensory evaluation and chemical profiling techniques by headspace solid-phase microextraction (HS-SPME) followed by gas chromatography coupled to mass spectrometry (GC-MS). Frontiers in Sustainable Food Systems 02 frontiersin.org Moreira et al. 10.3389/fsufs.2022.1011020 Materials and methods Algae and shrimp processing The raw biomass samples of microalgae TC and NO were obtained as a power, kindly provided by AllMicroAlgae, Portugal. The raw biomass samples of macroalgae UR and SL were also obtained as power from a local producer (AlgaPlus, Portugal) and milled before analyses. The shrimps (SH), species vannamei cong, were bought frozen at a local supermarket raw with head and carapace and used after thawing. To obtain the processed samples, 2 g of dried microalgae, macroalgae, or shrimp was placed in a 50 mL glass bottle with 20 mL of water and immersed in a water bath at 98◦C for 5 min. Then the mixtures were centrifuged at 330 g for 15 min, and the two phases were separated (liquid and solid). The solid phase was considered as cooked samples (c) and the liquid phase as cooking water (w). To obtain the non-processed samples, 0.5 g of powder microalgae and macroalgae or thawed shrimp was directly placed into a 20 mL vial and prepared for HS-SPME extraction. Headspace solid-phase microextraction of volatile compounds The VOCs of r, c, and resulting w were extracted using an HS-SPME method developed by Zhang and coworkers (Zhang et al., 2020) with slight modifications. This microextraction technique is solvent-free, cheap, easy to use, relatively fast, needs a low volume of sample, and is also sensitive enough for quality control purposes (Iranmanesh et al., 2018). For extraction and absorption of compounds, 0.5 g of solid sample (r and c) or 0.5 mL of w samples was transferred to a 20 mL vial with a three-phase SPME fiber (divinylbenzene (DVB)/carboxen (CAR)/polydimethylsiloxane (PDMS), 50/30 µm) from Supelco (Bllefonte, PA, USA). Moreover, 5 µL of internal standard (IS), 6-methyl-5-hepteno-2-one (from Sigma-Aldrich, Missouri, USA), in n-hexane at concentrations of 0.1 ug/mL (for TC, NO, and SL) and 0.02 ug/mL (for SH and UR) were added to each sample prior to incubation. The vial was incubated in an oven at 60◦C for 40 min. The absorption conditions of the compounds were selected taking into account the previous results (data not shown) and studies reported in the literature (Zhang et al., 2020). Before incubation, fiber was conditioned into the injector of a gas chromatograph for 20 min at 250◦C and was used immediately to prevent contamination. After incubation, the volatile compounds were analyzed in GC-MS. The analyses were done in triplicate. Gas chromatography-mass spectrometry analysis The samples were analyzed in a GC coupled with a mass detector (MS) equipped with an ion-trap analyzer (Shimadzu QP2020 NX, Kyoto, Japan). The volatile compounds were desorbed by inserting the fiber into the injection port, which remained in the injector for 5 min at a temperature of 250◦C. The split injection of samples was set at a ratio of 1:10, while helium (99.999% purity) was used as a carrier gas at a flow rate of 1.2 mL/min. Compounds were separated by using a Stabilwax capillary column (30 m ×0.25 mm ×0.25 µm film thickness, Agilent). The column temperature was programmed as follows: it was held at 40◦C for 3 min, and then increased to 100◦C at a rate of 5◦C/min, to 180◦C at a rate of 2◦C/min, to 250◦C at a rate of 10◦C/min, and then maintained at 250◦C for 5 min (Zhang et al., 2020). The MS ionization energy (EI) was 70 eV, the ion source temperature was 230◦C, and the interface detector temperature was 250◦C. The mass scanning range (m/z) was 40–450 at a scanning rate of 1.8 scans/s as described previously (Zhang et al., 2020). Identification and quantitative analysis of volatile compounds The volatile compounds were identified based on mass spectral interpretation and comparison with the National Institute of Standards and Technology (NIST) and WILEY libraries installed in GC–MS (match quality >80%). A semi-quantitative determination of the volatiles was done by directly comparing them with IS (Gu et al., 2013). The area of the chromatographic peak of each identified volatile was divided by the area corresponding to the IS. To semiquantitatively calculate the concentrations of each volatile, the obtained responses were multiplied by the concentration of IS in the sample, assuming that all of the response factors were equal to one (Coleman et al., 2022). Odor activity values (OAVs) were calculated as the ratio of their concentration to odor thresholds in water. Only aroma compounds exceeding their odor threshold concentration are suggested to contribute to the overall aroma of food. Odor thresholds in water were determined following a previously published protocol (van Gemert, 2011). Three replicate analyses were performed on each sample. Sensory analysis To evaluate the similarity of smell between algae and shrimp samples, a non-trained panel consisting of 50 assessors (17 men and 33 women, aged between 22 and 65 years) was used. A Frontiers in Sustainable Food Systems 03 frontiersin.org Moreira et al. 10.3389/fsufs.2022.1011020 TABLE 1 Odor activity values (OAVs) of volatiles within a specific group. Nannochloropsis oceanica Tetraselmis chuii Concentration (ng/g) OAVaConcentration (ng/g) OAVa Name Chemical formula DescriptorbRaw Cooked Water Raw Cooked Water Raw Cooked Water Raw Cooked Water Acids Acetic acid C2H4O2Sour 73.70 44.56 - <1<1 - 133.89 - 3.810 <1 - <1 Octanoic acid C8H16O2Sweat and cheese 20.30 - - <1 - - - - - - - - Alcohols 2-ethyl-1-hexanol C8H18O Rose and green 56.50 7.51 1.79 <1<1<1 - 1.68 - - <1 - Trans-2-hexen-1-ol C6H12O Leaf, green, and fruit 24.06 - - <1 - - 7.25 0.00 - <1<1 - 1,5-octadien-3-ol C8H14O Earth and herb 128.57 31.01 12.76 12.86 3.10 1.28 22.17 8.13 - 2,22 <1 - 3,5-octadien-2-ol C8H14O Green 34.39 7.04 2.90 <1<1<1 7.95 9.00 - <1<1<1 1-octen-3-ol C8H16O Fishy 168.63 50.77 9.72 <1<1<1 37.22 10.90 2.83 <1<1<1 Trans-2-octen-1-ol C8H16O Mushroom 105.38 53.94 15.69 5.27 2.70 <1 16.85 - 3.41 <1 - <1 2,3,6-trimethyl-7-octen-3-ol C11H22O Mushroom 51.45 - - ND - - 44.28 - - ND - - 1-penten-3-ol C5H10O Pungent 107.00 9.93 3.83 <1<1<1 17.40 1.35 <1 - <1 Aldehydes Benzaldehyde C7H6O Almond and burnt sugar 785.10 83.02 48.64 1.05 <1<1 - 31.57 34.94 - <1<1 Decanal C10H20O Soap, orange peel, and tallow - - 2.45 - - <1 - 3.97 4.61 - <1<1 2,4-decadienal C10H16O Fishy/ macroalgae 6.00 6.99 - 77.92 90.78 - 4.99 - - 64.84 - - 2,4-heptadienal C7H10O Fried - 20.33 10.29 - <1<1 48.04 6.39 4.08 <1<1<1 Hexanal C6H12O Grass, tallow, and fat 66.67 17.58 3.55 13.33 3.52 <1 8.83 2.42 2.28 1.77 <1<1 2-hexenal C6H10O Apple, green, fat, and rancid 17.45 4.65 3.23 <1<1<1 - - - - - - 2,4-hexadienal C6H8O Green 11.62 - 7.71 <1 - <1 14.23 - - <1 - - 2-heptenal C7H12O Soap, fat, almond, and fishy - 5.38 - - <1 - - - - - - 2-methyl-2-pentenal C6H10O Strawberry, fruit, and tomato 24.73 3.95 - <1<1 - - - - - - - Octanal C8H16O Fat, soap, lemon, and green 4.29 3.83 - 7.31 6.52 - 4.15 3.97 - 7.07 6.76 - 2-octenal C8H14O Sweat, green, nut, and fat 9.96 8.84 2.10 49.79 44.20 10.48 - 2.81 - - 14.05 - Trans-2,cis-6-nonadienal C9H14O Cucumber, wax, and green 12.86 10.07 - 16.08 12.58 - - - - - - - Nonanal C9H18O Fat, citrus, and green 4.69 9.93 5.44 4.27 9.02 4.95 4.64 7.65 5.11 4.22 6.95 4.65 Terpenes α-ionone C13H20O Wood and violet 4.29 161.06 - 7.15 268.43 - 247.08 88.73 16.12 411.80 147.88 26.87 β-ionone C13H20O Macroalgae, violet, flower, and raspberry 86.57 182.68 20.36 865.70 1826.80 203.60 293.44 103.71 23.32 2934.40 1037.15 233.21 Ester Butyl acetate C6H12O2Pear 36.51 47.14 - <1<1 - 49.02 52.34 97.90 <1<1<1 Hidrocarbon Dodecane C12H26 Alkane 48.85 5.16 - <1<1 - 10.89 0.00 0.00 <1 - - (Continued) Frontiers in Sustainable Food Systems 04 frontiersin.org Moreira et al. 10.3389/fsufs.2022.1011020 TABLE 1 (Continued) Nannochloropsis oceanica Tetraselmis chuii Concentration (ng/g) OAVaConcentration (ng/g) OAVa Name Chemical formula DescriptorbRaw Cooked Water Raw Cooked Water Raw Cooked Water Raw Cooked Water Ketone 3-buten-2-one C13H20O2Flower and green 526.45 145.29 55.10 1.05 0.29 0.11 91.14 33.95 26.25 <1<1<1 3,5-octadien-2-one C8H12O Fruit, fat, and mushroom 499.98 266.62 115.39 5.00 2.67 1.15 179.12 33.79 29.30 1.79 <1<1 2-undecanone C11H22O Orange, fresh, and green 9.96 - - ND - - - - - - - - Miscellaneous Ethylbenzene C8H10 Phenol and spice 73.85 82.00 - <1<1 - 78.01 88.31 134.09 <1<1<1 2-pentylfuran C9H14O Green bean and butter 27.30 5.78 - 4.71 1.00 - 5.16 - - <1 - - Dihydroactinidiolide C11H16O2Caramel 274.50 80.55 28.06 - - - 123.35 20.46 16.70 - - - Aroma profiles of raw, cooked, and cooking water of raw Nannochloropsis oceanica and Tetraselmis chuii were examined (n =3). ND, odor threshold non-described; (-), non-detected. aOdor thresholds in water (ng/ml) from van Gemert (2011). bOdor descriptor from: http://www.flavornet.org/flavornet.html. difference test (used to determine if a sensory difference, such as odor, taste, and texture, exists between samples) was performed using the shrimp sample as a control (Meilgaard et al., 2016). Using only the olfactory tract, the panelist was asked to classify the algal samples from 1 (no similarity) to 7 (equal odor) by comparing the shrimp smell with raw, cooked, and water samples. Statistical analysis Principal component analysis (PCA) was carried out to visualize the differences between the concentration of different samples based on GC-MS and sensory analysis data using IBM SPSS statistics (version 27; Granato et al., 2018). In the case of the GC-MS results, the concentration of 42 volatile compounds detected by the GC-MS was taken into consideration when building the matrix data set. The mean of at least three trials was used, and the results were presented in ng/g. Furthermore, SPSS statistics used a correlation matrix during data processing with the direct oblimin rotation method with Kraiser normalization. Results and discussion In today’s reality, the importance of fisheries cannot be underestimated, especially due to the huge demand for animal protein and the importance of fish nutrition. Aquaculture has been contributing to the growth of seafood and fish supply intended for the human supply, although the literature has highlighted some harmful results of aquaculture production and its environmental and ecological impacts. It is known that the rapid growth in shrimp farming is a key driver of mangrove forest degradation and the reduction of natural habitats and biodiversity. Also, aquaculture production may lead to a decrease in biodiversity and nutrition diversity, as it usually focuses on a few selected species (Sampantamit et al., 2020). Therefore, sustainable alternatives to seafood and seafood crops are of upcoming importance. However, for a meat/fish/seafood replacer to succeed in the market, it must be first accepted by the public in terms of overall liking, and the odor is the key in this aspect (Fiorentini et al., 2020). The results comprising the identification of VOCs, their relative concentration, and OAVs of the different processed and non-processed microalgae, macroalgae, and shrimp, are presented in Tables 1–3, respectively. Figures 1A–Cshow the chromatograms of rNO, cNO, and wNO as an example. Aroma-active compounds in microalgae Following the volatile analysis, a total of 103 compounds were primarily identified in microalgae NO and TC. Of these, Frontiers in Sustainable Food Systems 05 frontiersin.org Moreira et al. 10.3389/fsufs.2022.1011020 TABLE 2 Odor activity values (OAVs) of volatiles within a specific group. Saccharina latissima Ulva rígida Concentration (ng/g) OAVaConcentration (ng/g) OAVa Name Chemical formula DescriptorbRaw Cooked Water Raw Cooked Water Raw Cooked Water Raw Cooked Water Acids Acetic acid C2H4O2Sour 307.30 3.02 3.71 <1<1<1 13.57 0.12 - <1 - - Alcohols 1-octen-3-ol C8H16O Fishy 17.49 4.50 2.79 <1<1<1 6.30 0.46 0.80 <1<1<1 Trans-2-octen-1-ol C8H16O Fishy 31.70 3.10 1.10 1.59 <1<1 3.25 - - <1 - - 2,3,6-trimethyl-7-octen-3-ol C11H22O Mushroom 17.47 - - ND - - 5.90 - - ND - - Aldehydes Hexanal C6H12O Grass, tallow, and fat 3.89 2.38 2.32 <1<1<1 1.54 8.09 - <1 1.62 - 2-hexenal C6H10O Apple, green, fat, and rancid - - - - - - 1.00 13.60 - <1<1 - Octanal C8H16O Fat, soap, lemon, and green 3.42 - - 5.84 - - 1.50 0.14 0.98 2.56 <1 1.70 2-butyl-1-octanol C12H26O Mushroom 6.31 - - ND - - 4.32 11.01 - ND ND - Nonanal C9H18O Fat, citrus, and green 5.49 5.14 5.80 4.99 4.68 5.27 2.11 3.30 20.20 1.92 3.00 18.37 2-octenal C8H14O Sweat, green, nut, and fat 6.85 3.67 - 34.26 18.33 - 5.51 2.30 - 27.54 11.498 - 2,4-heptadienal C7H10O Fried 8.90 1.85 4.93 <1<1<1 6.73 2.65 1.88 <1<1<1 Butyl acetate C6H12O2Pear - 16.93 23.67 - <1<1 6.70 - - <1 - - Decanal C10H20O Soap, orange peel, and tallow 5.49 10.81 9.74 <1<1<1 2.56 - 1.85 <1 - <1 Benzaldehyde C7H6O Almond and burnt sugar 16.27 18.34 18.04 <1<1<1 24.91 15.71 7.84 <1<1<1 2-nonenal C9H16O Green 15.41 4.10 - ND ND - 7.89 2.75 - - ND - Hydrocarbon 6-methyltridecane C14H30 Alkane 1.71 - - ND - - 0.58 - - ND - - Tetradecane C14H30 Alkane 84.50 - - ND - - 17.66 - - ND - - 2,6,10,15tetramethylheptadecane C21H44 Alkane 28.19 - - ND - - - - - - - - Terpenes α-ionone C13H20O Wood and violet - - - - - - 22.90 20.66 3.71 38.17 34.44 6.18 β-ionone C13H20O Macroalgae, violet, flower, and raspberry 23.60 14.73 7.13 235.95 147.30 71.33 33.45 27.12 5.46 334.50 271.16 54.60 Ketone 3,5-octadien-2-one C8H12O Fruit, fat, and mushroom 22.28 4.00 3.87 <1<1<1 5.95 1.87 1.40 <1<1<1 5-ethyl-2(5H)-furanone C6H8O2Spice 5.73 - - <1 - - - - - - - - Gamma-nonalactone C9H16O2Coconut 9.09 13.80 2.91 - - ND 2.35 - - ND - - 3-buten-2-one C13H20O2Wood 19.15 44.29 10.07 <1<1<1 112.37 36.08 190.18 <1<1<1 3-nonen-5-one C9H16O Mushroom - 5.04 - - <1 - - 2.62 53.78 - <1<1 Miscellaneous Citronellol C10H20O Rose - - - - - - - 19.43 0.00 - ND ND Dihydroactinidiolide C11H16O2Caramel 35.93 75.11 7.85 ND ND ND 43.18 27.77 82.82 ND ND ND ND, odor threshold non-described; (-), non-detected. aOdor thresholds in water (ng/ml) from van Gemert (2011). bOdor descriptor from: http://www.flavornet.org/flavornet.html. Frontiers in Sustainable Food Systems 06 frontiersin.org Moreira et al. 10.3389/fsufs.2022.1011020 TABLE 3 Odor activity values (OAVs) of volatiles within a specific group. Shrimp Concentration (ng/g) OAVa Name Chemical formula DescriptorbRaw Cooked Water Raw Cooked Water Alcohol 2-ethyl-1-hexanol C8H18O Rose and green 44.92 5.98 4.37 <1<1<1 Aldehydes Nonanal C9H18O Fat, citrus, and green 18.52 32.82 55.05 16.83 29.84 50.04 Decanal C10H20O Soap, orange peel, and tallow 6.22 0.00 12.41 <1<1<1 Benzene Ethylbenzene C8H10 Phenol and spice 269.15 551.70 - <1<1 - Esters Butyl acetate C6H12O2Pear 123.67 358.55 - ND ND - Ketone 3,5-octadien-2-one C8H12O Fruit, fat, and mushroom 9.92 - 32.79 <1 - <1 Aroma profiles of raw, cooked, and cooking water from shrimp were examined (n =3). ND, odor threshold non-described; (-), non-detected. aOdor thresholds in water (ng/ml) from van Gemert (2011). bOdor descriptor from: http://www.flavornet.org/flavornet.html. only a total of 42 were considered for both microalgae due to the absence of an odor descriptor for the other compounds. Table 1 shows the sum of the concentration and OAVs of volatiles within a specific group to reveal similarities and differences between the aroma profiles of r, c, and w microalgae. These are composed of several classes of organic compounds: esters and hydrocarbons were the smallest groups with only one compound identified, followed by terpenes and acids (2), ketones (3), alcohols (3), and aldehydes were the largest group (13) (Table 1). The origin of ketones is variable, but generally linear ketones are derived from lipid oxidation, while methyl ketones may result from the β-oxidation of the fatty acids and subsequent decarboxylation (Moran et al., 2022). In general, saturated ketones are related to sweet, floral, and fruity odor notes, while unsaturated ketones are responsible for green odor notes. Ketones of VOCs were low in samples and only one was found with OAV>1 (rNO, cNO, and rTC), namely, 3,5octadien-2-one. Furthermore, 3,5-octadien-2-one concentration was three times higher in NO than in TC. These results agree with Coleman et al. (2022) who refer 3,5-octadien-2one as one of the major compounds that contribute to the overall aroma of NO and TC and confer a mushroom/earthy smell. Moreover, previous studies related the seafood-like odor in microalgae to their high content of diketones, such as 3,5-octadien-2-one (Van Durme et al., 2013;Coleman et al., 2022;Moran et al., 2022). Similarly, our results also suggested that 3,5-octadien-2-one may have a major contribution to the aroma not only in the raw microalgae matrix but also in c and w NO samples, though in a milder way in the processed samples. Most aldehydes have low threshold values and are important VOCs of aquatic products. In this study, only short-chain aldehydes (between 5 and 10 carbons) were found, probably due to the low extraction temperature used. Seven aldehydes were found to be present in rNO with a significant impact on the overall aroma, namely, hexanal, 2-octenal, benzaldehyde, 2,4-decadienal, octanal, trans-2-cis-6-nonadienal, and nonanal. In rTC, the main compounds are 2,4-decadienal, hexanal, octanal, and nonanal, but only the last one was present in the three types of samples (r, c, and w). In microalgae, alcohols are mainly formed as a result of secondary decomposition of hydroperoxides of fatty acids, except for branched alcohols which come from carbohydrates via glycolysis or from amino acids through the Ehrlich pathway (Giri et al., 2010). In contrast to other studies (Coleman et al., 2022;Moran et al., 2022), 1-octen-3-ol does not seem to have an impact on the overall aroma (Van Durme et al., 2013), although the isomer trans-2-octen-1-ol shows an effect. In the present study, 1,5-octadien-3-ol has high OAV for NO and TC, especially in the raw form. Previous studies conducted on lyophilized algae have ascribed the strong fishy/ grassy odors of the NO and TC to the presence of hexanal, 2,6-nonadienal, 2,4-decadienal, and 1-octen-3-ol. As aforementioned, in the current work, the results did not support the hypothesis that 2,6-nonadienal is a key odorant in TC due to the absence of the compound. Furthermore, due to their high odor threshold, 1-octen-3-ol also seems to not justify that strong fishy odor, as its OAV is <1. Furans can provide a burned, savory, sweet, and green aroma to foods (Starowicz and Zieli´ nski, 2019). In rNO, cNO, and rTC samples, 2-penthylfuran, an oxidative derivative of n-6 PUFAs characterized by their green notes, was detected, which may significantly contribute to their aroma in the raw samples, especially in NO. High carotenoid concentrations are normally associated with a high concentration of ionones (terpenes) that are produced due to reactions with the unstable conjugated doublebound structure of carotenes. At the end of the degradation of carotenoids, where their long-chain compounds are mostly oxidized, large amounts of short-chain monooxygenated and Frontiers in Sustainable Food Systems 07 frontiersin.org Moreira et al. 10.3389/fsufs.2022.1011020 FIGURE 1 Chromatogram by GM-MS from (A) raw Nannochloropsis oceanica,(B) cooked Nannochloropsis oceanica, and (C) cooking water from Nannochloropsis oceania. deoxygenated compounds, such as β-cyclocitral, are formed (Van Durme et al., 2013). In NO, it seems that cNO contains a higher concentration of terpenes, particularly β-ionone, when compared with the raw and liquid samples. Yet for TC, the raw sample has an OAV almost three times higher than the cTC. Also, α-ionone is present in both cTC and cNO, although in smaller concentrations. Moreover, for both wNO and wTC samples, α-ionone was not detected. The smell of either rNO, cNO, and rTC can be explained by the mixture of 1,5-octadien-3-ol, hexanal, 2,4-decadienal, Frontiers in Sustainable Food Systems 08 frontiersin.org Moreira et al. 10.3389/fsufs.2022.1011020 2-octenal, octanal, nonanal, 3,5-octadien-2-one, and terpenes. cNO shows the same key compounds as rNO, although in a smaller concentration, except for 2,4-decadienal, nonanal, and terpenes. wNO seems to present a softer smell when compared with the two other NO samples (rNO and cNO) due to the lack or low concentration of some of the key odorants. The odor of cTC and wTC seems to be contributed by nonanal and terpenes, and in cTC also by octanal and 2-octenal. wNO sample seems to retain 1,5-octadien-3-ol, 2-octenal, nonanal, 3,5-octadien-2-one, and terpenes. Aroma-active compounds in macroalgae Seventy-seven VOCs were primarily identified for SL and 79 for UR. Of these, only a total of 27 were considered due to the absence of an odor descriptor for the other compounds, as shown in Table 2. These are composed of several classes of organic compounds: acids were the smallest group (2 compounds), followed by terpenes (2), hydrocarbons (3), alcohols (3), ketones (5), and aldehydes (11). Only one alcohol was found to have an impact on the overall aroma of raw SL, that is, the trans-2-octen-1-ol. Alcohols in macroalgae derive from unsaturated acids by peroxidation reactions, from carbohydrates by glycolysis, or from amino acids (Sánchez-García et al., 2019). A study performed by Pina et al. (2014) found 1-octen-3-ol and other alcohols in dried red macroalgae, but only ethanol and 1-penten-3-ol in samples submitted to different culinary treatments. Branchedchain alcohols like 1-octen-3-ol are often associated with fishy and grassy aromas and seem to contribute significantly to the aromatic fraction in macroalgae if their low odor threshold values are considered, although in this study, only trans-2-octen-1-ol showed an OAV>1 (Sánchez-García et al., 2019). Three aldehydes were present in both UR and SL: octanal, nonanal, and 2-octenal. In general, linear and branched chain aldehydes contribute to herbaceous and grassy aromas, while unsaturated aldehydes provide green and fishy odors (SánchezGarcía et al., 2019). Some authors have reported that, in macroalgae, shortand middle-chain aldehydes are derived from fatty acids (Akakabe and Kajiwara, 2008). Nonanal was found in all studied algae, and in contrast to all other compounds, their concentration was higher in cooking water and in the cooked sample. On the other hand, 2-octenal was found only in raw and cooked matrices in both macroalgae. In contrast to other studies, hexanal was found in both macroalgae, although it does not seem to play a role in overall aroma (Sánchez-García et al., 2019). Regarding hydrocarbons, only three were detected, but none seem to have an impact on the overall aroma. The PCA was performed to explore further similarities of odor among processed and non-processed algal samples based on GC-MS data. The raw matrix data set is presented in Supplementary Table 1.Figure 2 shows the loading plots of VOC concentration with OAV >1 from GC-MS of all processed and non-processed algae. The samples seem to be grouped into five groups: (1) cTC and wTC; (2) rTC and cNO; (3) rNO, wNO, and wSL (4) cSL, rUR, cUR, and wUR; and (5) rSL. Although groups 2, 3, and 4 are in the same quadrant, the distance and the deviation between them are different. The processed macroalgae tend to be more similar to each other due to the already identified VOCs, such as 1,5-octadien-3-ol, 2,4-decadienal, hexanal, nonanal, octanal, α-ionone, β-ionone, and 3,5-octadien-2one. The similarity between rNO, cNO, wNO, rSL, and wSL may be due to the compounds nonanal, β-ionone, and 3-buten-2-one. The key odorants present in microalgae belong essentially to alcohols (1,5-octadien-3-ol and trans-2-octen-1-ol), aldehydes (2,4-decadienal, hexanal, octanal, 2-octenal, trans-2,cis-6-nonadienal, and nonanal), terpenes (α-ionone and β-ionone), and ketone (3,5-octadien-2-one) groups. Macroalgae odor seems to be characterized fundamentally by nonanal, terpenes, and 2-octenal. Compared to microalgae, the amount and concentration of key odorants found in macroalgae are much lower, and it seems that their smell is more green than fishy. As found in microalgae, terpenes play a major role as VOCs, especially β-ionone. Also, between processed and non-processed algae, some compounds are different; for example, hydrocarbon was not found in processed algae. Cooking waters presented a lower number of compounds and, in general, a low concentration of the presented compounds compared with those observed in raw and cooked algae. Comparison of aroma-active compounds in shrimp and algae For shrimp volatile analysis, a total of 21 compounds were primarily identified. Of these, only six were considered due to the absence of an odor descriptor for the others (Table 3). In this work, the major odorant in SH was found to be nonanal. The results obtained for cSH were not similar to those reported in the literature. Zhang et al. (2020) have described that shrimp’s key odorants are mainly pyrazine, such as 3-ethyl-2,5dimethylpyrazine and 2,5-dimethyl pyrazide, trimethylamine, and 3-(methyltio)propionaldehyde. On the other hand, Mall and Schieberle (2017) have reported that 3,5-octadien-2-one and 2-acetyl-1-pyrroline are the main odorants in the studied shrimps. This divergence between the studies could be due to Frontiers in Sustainable Food Systems 09 frontiersin.org