Bioprocessing of macroalgae for bioactive compounds production with food and feed applications

Universidade do Minho Escola de Engenharia Ana Isabel Freitas Baptista Bioprocessing of macroalgae for bioactive compounds production with food and feed applications janeiro de 2021 UMinho | 2021 Ana Isabel Freitas Baptista Bioprocessing of macroalgae for bioactive compounds production with food and feed applications University of Minho School of Engineering Department of Biological Engineering Ana Isabel Freitas Baptista Bioprocessing of macroalgae for bioactive compounds production with food and feed applications Master’s degree in Biotechnology Thesis dissertation Work supervised by: PhD José Manuel Salgado PhD Professor Isabel Belo January 2021 ii DIREITOS DE AUTOR E CONDIÇÕES DE UTILIZAÇÃO DO TRABALHO POR TERCEIROS Este é um trabalho académico que pode ser utilizado por terceiros desde que respeitadas as regras e boas práticas internacionalmente aceites, no que concerne aos direitos de autor e direitos conexos. Assim, o presente trabalho pode ser utilizado nos termos previstos na licença abaixo indicada. Caso o utilizador necessite de permissão para poder fazer um uso do trabalho em condições não previstas no licenciamento indicado, deverá contactar o autor, através do RepositóriUM da Universidade do Minho. Atribuição-NãoComercial-SemDerivações CC BY-NC-ND https://creativecommons.org/licenses/by-nc-nd/4.0/ iii STATEMENT OF INTEGRITY I hereby declare having conducted this academic work with integrity. I confirm that I have not used plagiarism or any form of undue use of information or falsification of results along the process leading to its elaboration. I further declare that I have fully acknowledged the Code of Ethical Conduct of the University of Minho. iv AGRADECIMENTOS Depois deste ano de grande aprendizagem e crescimento a nível pessoal e profissional, tenho de agradecer às pessoas que mais contribuíram para o meu sucesso: Ao meu orientador, Dr. José Salgado, agradeço por toda a ajuda disponibilizada em todos os momentos e disponibilidade para partilhar conhecimento. À minha coorientadora, professora Isabel Belo, agradeço pela oportunidade de integrar o trabalho deste grupo e pela disponibilidade para ajudar e ensinar. À Marta, por toda a ajuda no trabalho prático, pela sua disponibilidade e amizade. A todos os restantes membros deste laboratório, agradeço por me integrarem tão bem e por serem sempre tão prestáveis quando precisei de ajuda. Aos meus pais, um obrigado especial pelo apoio incondicional, por sempre me ajudarem a seguir os meus sonhos e nunca me deixarem desistir. Ao Nuno, pela confiança que tem em mim, por me fazer acreditar em mim e por fazer sempre parte de todos os momentos da minha vida. Ao resto da minha família, à Mariana e aos meus colegas de mestrado, obrigada por serem um apoio sempre que precisei, por ouvirem as minhas queixas e estarem sempre comigo nas minhas vitórias e nas minhas derrotas. v ACKNOWLEDGMENT This work was supported by the R&D&I project “Development of innovative sustainable protein and omega-3 rich feedstuffs for aquafeeds, from local agro-industrial by-products”, reference POCI-010145FEDER-030377, funded by European Regional Development Fund (ERDF) and Portuguese Foundation for Science and Technology (FCT). vi Bioprocessamento de macroalgas para a produção de compostos bioativos com aplicações alimentares RESUMO A indústria de macroalgas atual baseia-se na produção de um único composto, como alginatos, agar ou corantes, sendo que os subprodutos são tratados como resíduo. Assim, esta indústria enfrenta o desafio de desenvolver processos que permitam obter vários produtos com atividade biológica. A fermentação em estado sólido (SSF) é um processo biotecnológico de baixo custo que pode produzir compostos bioativos, como enzimas e compostos antioxidantes. As macroalgas verdes como a Ulva rigida têm potencial para ser usadas como ingrediente em aquacultura. No entanto, estas são difíceis de digerir para muitas espécies de peixe. Neste sentido, a SSF pode alterar a estrutura de polissacarídeos para facilitar a digestão das macroalgas e pode ainda produzir uma grande variedade de produtos com aplicações alimentares como proteínas, enzimas e compostos antioxidantes. O principal objetivo deste projeto é o processamento sequencial da U. rigida por SSF e hidrólise enzimática (EH) para produzir produtos de valor acrescentado e aumentar o valor nutricional das macroalgas, promovendo assim uma economia circular. O passo da hidrólise enzimática foi otimizado pelo desenho experimental Box-Behnken. Durante a SSF da U. rigida produziram-se celulases (40 ± 1 U/g) e xilanases (160 ± 4 U/g). Após a SSF, foi adicionado tampão para iniciar a EH que durou 72h. A variável que teve um maior efeito na libertação de compostos fenólicos, açúcares, atividade antioxidante e aumento da concentração de proteína foi a temperatura. A concentração máxima de compostos fenólicos e atividade antioxidante atingiu-se a uma temperatura intermédia (40°C), a conversão máxima de celulose a glucose e aumento da concentração de proteína atingiram-se à temperatura mais elevada (44°C). As condições ótimas da hidrólise enzimática para maximizar em conjunto as 4 variáveis foram 44°C, carga de sólido 30% w/v e pH 4,1. Nestas condições, atinge-se teoricamente 929 μM de equivalentes de Trolox/g, 1,56 mg de compostos fenólicos totais/g, 231,04 g de proteína/kg e 61% de conversão de celulose para glucose. Em todas as experiências, verificou-se uma diminuição da atividade da xilanase durante a hidrólise enzimática (72h), sendo que a redução foi menor nas experiências realizadas a menor temperatura (35°C). O bioprocessamento da U. rigida por SSF e EH permitiu a obtenção de compostos antioxidantes, açúcares livres que podem ser fermentados noutros produtos de valor acrescentado ou energia e um sólido final enriquecido em proteína. No futuro, devem ser realizadas experiências de modo a aplicar estes produtos na aquacultura. Palavras-chave: Bioprocessos; Fermentação em estado sólido; Hidrólise Enzimática; Macroalgas vii Bioprocessing of macroalgae for bioactive compounds production with food and feed applications ABSTRACT The current seaweed industry is based on a single compound production, as alginates, carrageenan, agars, or colorants, being the remaining seaweed byproduct treated as waste. Thus, macroalgae industry faces the challenge of developing processes allowing to obtain multi-products with biological activities. Solid-state fermentation (SSF) is a low-cost biotechnology process that can produce bioactive compounds as enzymes and antioxidant compounds. Green macroalgae as Ulva rigida have potential to be used as ingredient in aquaculture. However, they are difficult to digest by many species of fish. In this sense, SSF can alter the structure of polysaccharides to facilitate digestion of macroalgae, and it can also produce a wide variety of valuable products for feed applications, such as proteins, enzymes, and antioxidant compounds. The main aim of this project is the sequential bioprocessing of U. rigida by SSF and enzymatic hydrolysis (EH) to produce value-added products and increase the nutritional value of macroalgae, promoting a circular economy. The EH stage was optimized by Box-Behnken experimental design. During SSF were produced cellulases (40 ± 1 U/g) and xylanases (160 ± 4 U/g). After SSF, it was added the buffer to carry out EH during 72h. The variable that had a higher effect on release of phenolic compounds, sugars, antioxidant activity and increase the concentration of protein was the temperature. Maximum concentration of phenolic compounds and antioxidant activity was achieved with intermediate temperature (40°C), the maximum conversion of cellulose to glucose and increase of protein concentration were achieved with the higher temperature (44°C). The optimal conditions of EH to maximize jointly the 4 variables were 44°C, load of solid 30% w/v and pH 4,1. In these conditions, they were predicted an antioxidant activity of 929 μM of Trolox equivalents/g, 1,56 mg of total phenolic compounds U/g, 23,.04 g of crude protein/kg and 61% cellulose conversion to glucose. In all experiments it was observed a decrease of xylanase activity during EH, the reduction was lower in experiments performed with the lowest temperature (35°C). The bioprocessing of U. rigida by SSF and EH allowed to obtain antioxidant compounds, free sugars that can be fermented to other value-added products or energy, and the final solid was enriched in protein. Future works should be performed to apply these products in aquaculture. Key Words: Bioprocess; Enzymatic Hydrolysis; Macroalgae; Solid-state fermentation xiv 1 1. INTRODUCTION 1.1. Macroalgae Macroalgae - also called seaweeds - have great potential, however, only recently have they been attracting more attention from researchers. This is mainly due to a number of favorable characteristics, such as their large biomass yields, fast growth rates and low needs of freshwater and terrestrial land for cultivation (Kostas, White and Cook, 2017; Fernandes et al. , 2019). These are chlorophyll containing organisms, which means that they can photosynthetically convert atmospheric carbon dioxide into a variety of metabolites and organic molecules (Sambusiti et al. , 2015). In contrast with terrestrial plants, macroalgae have, in average, a much higher photosynthetic efficiency (between 6 and 8 % for macroalgae, but only from 1.8 to 2.2% for terrestrial plants) (Chen et al. , 2015). Another advantage of macroalgae when compared to terrestrial plants is the low quantity of lignin they have (most of the times, lignin is absent), which dispenses the need for intensive pre-treatments prior to fermentation, meaning reduced costs and less energy spent (Trivedi et al. , 2015). Furthermore, they can grow in a wide variety of environments, including fresh, salt, temperate and municipal wastewater (Masri et al. , 2018) and they constitute approximately 50% of biomass on Earth (Barbot, Al-Ghaili and Benz, 2016). As such, macroalgae potentially represent a significant source of renewable energy and a primary source of natural products (Ross et al. , 2008; de Almeida et al. , 2011), which makes this an interesting organism to exploit in biorefinery processes. Their metabolism can differ accordingly to certain parameters, such as the water temperature, salinity, light, or available nutrients. This forces macroalgae to quick adaptions to new environmental conditions, which makes them produce a wide variety of secondary metabolites with biological activity (Rodrigues et al. , 2015). Some of the compounds that macroalgae might synthetize are carotenoids, terpenoids, vitamins, saturated and polyunsaturated fatty acids, antioxidants and polysaccharides, such as agar (de Almeida et al. , 2011). Their growth is influenced by the presence of dissolved nutrients in water, specifically, nitrogen, phosphorous and iron and their optimal growth temperature ranges from below 15°C ( Ascophyllum spp. found in Northern hemisphere) to 25°C ( Ulva pertusa , found in Japanese coastline) (Barbot, Al-Ghaili and Benz, 2016). When it comes to structure, macroalgae are simple multicellular organisms, with simple reproductive structures, and they do not have advanced structures such as the ones present in most terrestrial plants like roots, stems, leaves or vascular tissue. Instead, they have a blade that is leaf-like, a stipe that is stem-like and a holdfast that matches roots in terrestrial plants (de Almeida et al. , 2011). 2 Some species reproduce asexually by means of vegetative growth, which means that new individuals will be genetically identical to their parent (Sudhakar et al. , 2018). However, the life cycles of macroalgae are diverse and they have combinations of sexual and asexual reproductive strategies (Roesijadi, Jones and Zhu, 2010). According to the Food and Agriculture Organization of the United Nations (FAO), the annual production, in 2010, of marine macroalgae was over 16 million tons and over 96% of that production is from aquaculture. This production is valued in USD 7 billion, corresponding to a little over 6 billion Euros, however, to obtain the total value of the seaweed industry, it is necessary to consider the added value products obtained after processing of macroalgae ( aquaCase , no date). Five years later, by 2015, total production almost doubled, achieving 30.4 million tons (Ferdouse et al. , 2018). By 2017, the commercial seaweeds market was valued at USD 13.07 billion - that is, 11.6 billion euros - and is expected to be above 18.5 billion euros by 2023 ( Markets and Markets , no date). The Laminaria , Undaria , Porphyra , Eucheuma / Kappaphycus and Gracilaria genera account for approximately 98% of world seaweed production (Pereira and Yarish, 2008). The East Asian countries are world’s greatest contributors on algal biomass - they accounted for 95% of the world’s supply in 2010 (Jung et al. , 2013). The more recent estimates suggest the existence of 72500 species worldwide (Guiry, 2012). 1.1.1. Types of macroalgae There are three major groups of macroalgae, in a classification based on their photosynthetic pigmentation: Chlorophyta (green pigments), Rhodophyta (red pigments) and Phaeophyta (brown pigments) (Chen et al. , 2015). Their distribution depends on environmental factors, with emphasis on the quantity and quality of the sunlight (Sudhakar et al. , 2018). Starting with Chlorophyta , or green algae, there are around 700 to 7000 species and they occur mainly in bays, estuaries, and tide pools. Chlorophyta algae have simple thallus and are characterized by filamentous spongy fingers or paper-thin sheets. This group of macroalgae has, as major photosynthetic pigments, chlorophylls a and b – with the same ratio of chlorophyll a to b as land plants – and carotenoids, such as carotene and xanthophylls (Jung et al. , 2013; Sudhakar et al. , 2018; Leandro, Pereira and Gonçalves, 2020). Some of the most relevant species in this group are Halimeda , Ulva and Codium (Sudhakar et al. , 2018). As for Rhodophyta , the group of red algae, they grow as filaments or sheets of cells. One interesting characteristic of Rhodophyta is that they can be parasites of other algae. This is the most abundant and most widespread group of algae, with records of 4000 species that can live either in deep 3 cold waters or warm shallow waters. The photosynthetic pigments that can be encountered in the Rhodophyta group are chlorophyll a and phycobilins (phycoerythrin and phycocyanin), which is responsible for their red color (Jung et al. , 2013; Sudhakar et al. , 2018; Leandro, Pereira and Gonçalves, 2020). Lastly, the group of brown algae, Phaeophyta , comprises around 1500 species and they occur mainly in temperate and polar locations (North America, Europe, mid-Atlantic and Gulf of Mexico). Also, they can be found in rocky shores and this type of algae have preference for shallow and cold waters. These macroalgae species can grow up to 100 meters and are the ones that have the most complex thallus structure (Sudhakar et al. , 2018). The main photosynthetic pigments present in this type of algae are chlorophyll a and c , β-carotene and other xanthophylls (Jung et al. , 2013). Some of the most important species that belong to the Phaeophyta group are Laminaria and Saccharina (Sudhakar et al. , 2018). The Rhodophyta and the Phaeophyta groups seem to be more present in Portuguese seas than Chlorophyta ( seaExpert , no date; Rodrigues et al. , 2015). 1.1.2. Chemical composition of macroalgae The chemical composition can vary between the different groups of macroalgae and it is influenced by seasonality and geographic locality (Wan et al. , 2019). Also, it can be influenced by the conditions in their habitat such as light, temperature, salinity, nutrients, and pollution. However, in every case, they contain high amounts of carbohydrates (up to 60%) and lower amounts of protein (around 10%) and even lower amounts of lipids (up to 3%). The composition of the three groups of macroalgae regarding water content, carbohydrates, proteins, and lipids is presented in Table 1. 4 Table 1 - Composition of macroalgae (green, red and brown). Compound Green algae Red algae Brown algae Reference Water content (fresh mass) 70-85% 70-89% 79-90% (Barbot, Al-Ghaili and Benz, 2016) Carbohydrates (dry weight) 25-50% 30-60% 30-50% (Jung et al. , 2013) Protein (dry weight) 10-26% 35-47% 7-12% (Miranda, LopezAlonso and Garcia-Vaquero, 2017) Lipids (dry weight) 2-3% 0-3% 0-2% (Jard et al. , 2013) There is a large variety of carbohydrates present in macroalgae: alginate – that provides stability and flexibility - and cellulose are common in all types of macroalgae. Due to its abundance, easiness of processing and variety of applications, cellulose has been getting a lot of attention from researchers and macroalgae have been considered a potential source of cellulose (Siddhanta et al. , 2009). Also, in green algae we can encounter mannan, starch and ulvan; red algae contain agar, carrageenan and lignin and brown algae contain agar, laminarin, cellulose and others (Barbot, Al-Ghaili and Benz, 2016; Miranda, Lopez-Alonso and Garcia-Vaquero, 2017). The quantity of lignin is generally lower in macroalgae in comparison with terrestrial plants and its absence is important to microbial decomposition (Barbot, AlGhaili and Benz, 2016). Regarding lipids, macroalgae contain a significant amount of PUFAs (long-chain polyunsaturated fatty acids). The content of unsaturated fatty acids is proportional to the potential antioxidant activity, as shown in a study with lipophilic extracts from 16 species of seaweeds (Huang and Wang, 2004). However, phenolic compounds are usually claimed to be the major active constituents responsible for the antioxidant activity of macroalgae. These phenolic compounds are highly present in green and brown macroalgae, which could be the reason of lower content in protein comparing to red algae (Kumar et al. , 2008). Another major constituent of macroalgae are pigments, such as chlorophylls, carotenoids, phycobilins and xanthophylls and their abundance depends on the type of macroalgae as discussed in the above section. 5 As for micro-nutrients, macroalgae have high concentrations of minerals, such as calcium, magnesium and potassium, as well as glutamic acid (Barbot, Al-Ghaili and Benz, 2016). They also have high contents in iodine, being Laminaria spp. the best iodine accumulator among all living systems (Miranda, Lopez-Alonso and Garcia-Vaquero, 2017). Some heavy metals can also be found, such as arsenic and mercury, but the amount does not seem to pose any threat to the consumers’ health (GarciaVaquero and Hayes, 2016). Seasonal environmental changes can influence macroalgae’s composition; during summer, they produce higher amounts of volatile solids – amount of organic substance in the solid fraction - and sugars whereas in spring they show higher content in proteins and minerals (Barbot, Al-Ghaili and Benz, 2016). Ulva species, a green type of macroalgae, are listed in FAO as one of the main macroalgae for commercial use. This macroalgae is very common and can be found in marine and brackish waters, being widely distributed across the world. Also, Ulva species can be successfully product in an aquaculture environment (Lopes et al. , 2019). In what concerns its composition, Ulva species are a major source of polysaccharides and oligosaccharides and, despite the lipidic profile not being intensively studied, they also represent an important nutritional role with major importance for PUFAs (Satpati and Pal, 2011; Lopes et al. , 2019). This species has also particular interest as a source of antioxidants and phenolic compounds and the amount of phenolic compounds positively correlates with the radical-scavenging activity, suggesting that phenolic compounds are the major contributor acting as free radical terminators (Mezghani et al. , 2016). 1.1.3. Biorefinery of macroalgae Biorefineries transform renewable biomass into biofuels, food, chemicals, and other bio-based products. On simple terms, biorefineries are the operating units of bio-economies (Zollmann et al. , 2019). The main ideas in which a biorefinery process is based are the sustainable and renewable energy supply, saving foreign exchange reserves, reducing dependency on imported crude and other chemicals and the establishment of a circular economy. The final goal is to generate added value products, with benefits for the economy and the environment. While most researchers are focused on single-feedstock and singleproduct, one more advantageous approach is the co-production of multiple products from the same biomass, because this type of process leads to complete use of the raw material, close to zero-waste and maximum material valorization (Kazir et al. , 2019; Zahra et al. , 2019). The use of all content to produce high-value products makes the biorefinery process more profitable and sustainable, since it is increasing the biorefinery’s global economic performance (Zahra et al. , 2019). 6 Due to limited crop yields and land availability, the future development of the biomass sector is largely uncertain (Zahra et al. , 2019). In this sense, macroalgae are excellent feedstocks [i.e., raw materials that are used in biorefineries (Cherubini, 2010)] since they have, not only high value components, but also compounds that are considered platform chemicals for the bio-based economy (Kostas, White and Cook, 2017). The fact that biorefineries based on terrestrial biomass are not sustainable at present due to environmental as well as economic issues enhances the potential of macroalgae (Jung et al. , 2013). Several seaweed biorefinery processes have already been investigated. Sequential recovery of four fractions with economic interest – a liquid extract, containing nutrients suitable for use as food supplements, a lipid fraction, ulvan and finally a cellulose fraction – was reported by Trivedi et al. (Trivedi et al. , 2016). In a study using Chaetomorpha linum , the authors demonstrated the feasibility of the coproduction of biogas and bioethanol, a process with low production of waste (Ben Yahmed et al. , 2016). Kostas, White and Cook reported the use of Laminaria digitata , a brown seaweed, to successfully produce bioethanol from the residues which remained after the extraction of two valuable polysaccharides (Kostas, White and Cook, 2017). Also, a lot of different products can be obtained from the residual algal biomass including products with application in food/feed, pharmaceutical, nutraceutical and cosmeceutical industries (Suganya et al. , 2016). 1.1.4. Applications of macroalgae Seaweeds have a wide range of applications, being the more traditional ones the commercialization as food and soil fertilizer (van der Wal et al. , 2013). Nowadays, they have a wide range of applications and the components extracted from the macroalgae can potentially be applied in the food, medical and pharmaceutical industries, in the environmental fields, among others. Macroalgae are valuable as a food resource as they are rich in vitamins, minerals, proteins, polysaccharides, and dietary fibers and are low in calories. Phycocolloids such as agar-agar, carrageenan and alginic acids that are present in brown and red algal cell walls are widely used in food industries (de Almeida et al. , 2011). Both agar-agar and carrageenan have gelling, thickening, and stabilizing properties, which allows them to be used as substitutes for gelatin and in dairies, respectively (Pangestuti and Kim, 2015). Also, algae can be applied in aquaculture, ruminant and swine feed industries (Miranda, LopezAlonso and Garcia-Vaquero, 2017). 7 In the medical and pharmaceutical field, macroalgae have also demonstrated to be interesting, as has been proved that various macroalgae have antibacterial, antifungal, and antiviral activities (Smit, 2004; de Souza Barros, Teixeira and Paixão, 2015). Besides that, some components present in macroalgae can be used for their antioxidant and anti-inflammatory potential, which is the case of some polysaccharides (Ananthi et al. , 2010). Some other examples of compounds with potential to be used in the pharmaceutical field are fucoxanthin that could have applications in cancer treatment, because it can induce cell cycle arrest and apoptosis, and also laminarin, because for its nutritional value it can play a role in prevention (Fleurence and Levine, 2016). When it comes to the environmental field, besides being used as a fertilizer since ancient times, it can also be used to control pollution. It has even been developed a device, in 2011, called algal turf scrubber, which absorbs nutrients and is used to help filter aquaria and ponds (HydroMentia, no date). Lastly, macroalgae can also be applied to produce biofuels (Chen et al. , 2015), in bioremediation processes (Sode et al. , 2013) and cosmetics – moisturizing care, photoprotection – and additives for cosmetics – preservatives, essential oils, antioxidants, dyes (Guillerme, Couteau and Coiffard, 2017). 1.1.5. Wastes from macroalgae processing In the many processes, which use macroalgae for the applications mentioned above, there are, often, residues or wastes produced in the course. For example, in the process of phycocolloid extraction there remains the cell wall as residue, in which can be found some impurities such as sand, salts and calcareous deposits, as well as sulfolipids, pigments, nucleic acids, other polysaccharides as cellulose, which can be used in chemical, pharmaceutical and fuel industries (López-Simeon et al. , 2012). Studies have also shown that after the extraction of some polysaccharides, such as agar and alginates, there remains a pulp containing high amounts of carbohydrates, proteins, lipids, and ash (Zahra et al. , 2019). Currently seaweed wastes are used to produce fiber, glycerol, biofertilizers and organic acids but not exclusively. The residues from the alginate industry have also been used for the elimination of toxic heavy metals and for biomethane conversion. Also, make use of seaweed wastes to produce biomethane has already been tested and the results were promising (Barbot, Al-Ghaili and Benz, 2016). In general, the quality of the wastes and, therefore, the application they can further have, depends on the initial composition of the macroalgal biomass and on processing done. For example, residues from phycobilin extraction have a high percentage of volatile solids and a lower percentage of ash (75% and 21%, respectively) while the remains from industrial biomass processing of Laminaria japonica have 8 almost the same percentage of volatile solids and ash (50,9% and 49,1%, respectively) (Barbot, Al-Ghaili and Benz, 2016). However, it is possible to reduce the quantity of waste produced since one can co-extract additional valuable materials in the initial macroalgae processing, rather than treating them as waste (Kazir et al. , 2019; Zahra et al. , 2019). 1.2. Solid-state fermentation Solid-state fermentation (SSF) is a cost-effective bioprocess technology, with potential applications in a variety of areas, such as the feed, food, fuel and chemical industries, but also the production of pharmaceutical products and biologically active secondary metabolites – for example, pigments and antibiotics (Thomas, Larroche and Pandey, 2013; Singhania et al. , 2015). SSF systems seem to be promising to produce value-added products, such as biopharmaceuticals. Also, this technology has been used for the development of bioprocesses for instance bioremediation and biodegradation of hazardous compounds and biological detoxification of agro-industrial residues (Pandey, 2003). SSF is a three-phase heterogeneous process, composed by solid, liquid and gaseous phases (Costa et al. , 2018). It is a fermentation process that occurs in the absence, or near absence, of free water; however, the solid substrate must contain enough moisture to support the microorganism’s growth and metabolic activity (Hölker, Höfer and Lenz, 2004; Thomas, Larroche and Pandey, 2013; Cerda et al. , 2019). The solid matrix where the process occurs can either be the source of carbon or an inert material to support the microorganisms’ growth (Salgado et al. , 2014a; Oliveira et al. , 2016; Costa et al. , 2018). SSF offers many advantages when compared with classic submerged fermentation (SmF), in which the microorganisms grow in liquid medium, with high contents of free water (Soccol et al. , 2017; Wang et al. , 2019). SSF has lower energy requirements and higher productivities, produces lesser wastewater, the products have extended stability, the production costs are lower, and it is less prone to problems with substrate inhibition, hence it allows higher final concentration of product (Pandey, 2003; Hölker, Höfer and Lenz, 2004; Hölker and Lenz, 2005; Barrios-González, 2012; Soccol et al. , 2017; ElMansi et al. , 2019; Wang et al. , 2019). Also, when it comes to environmental related issues, the fact that SSF is conducted in near absence of free water results in minimum water consumption and a low production of effluent water in the process. The fact that SSF is performed at low water activities, reduces 9 the growth of contaminating bacteria and yeasts, thus, in certain cases, semi-sterile conditions may be applied, reducing the energy needed for sterilization (Hölker and Lenz, 2005; Soccol et al. , 2017). Yet another benefit regarding the sustainability of the process is the utilization of low-cost agro-industrial residues as carbon and energy sources (Hölker and Lenz, 2005; Thomas, Larroche and Pandey, 2013). The advantages cited can bring direct economic advantages, therefore the economic efficiency is higher for SSF than SmF (Hölker and Lenz, 2005; Soccol et al. , 2017). Several studies have been conducted comparing both types of fermentation. In a study conducted by Díaz-Godínez et al. , 2001, the production of exopectinases by Aspergillus niger in SSF and SmF was reviewed, and the research concluded that the production of biomass in SSF was higher, independently of the variables tested. Also, the exopectinase production was enhanced by using SSF over SmF. The SSF process may offer advantages in terms of enzyme activity because of reduced proteolysis (Díaz-Godínez et al. , 2001). One other interesting factor of SSF is that it provides the cultivated microorganisms an environment as close as possible to their natural habitat, from where they were isolated (Hölker, Höfer and Lenz, 2004; Thomas, Larroche and Pandey, 2013; Oliveira et al. , 2017). This seems to be the main factor behind the higher productivity yields in SSF when compared with SmF, even if optimal conditions for growth are used (Thomas, Larroche and Pandey, 2013). Nevertheless, the SSF process has also some disadvantages, such as problems with heat buildup, difficulties in controlling process parameters (like pH, temperature, moisture), difficulties on scale-up and higher impurity of the product (Hölker, Höfer and Lenz, 2004; Couto and Sanromán, 2006). 1.2.1. Macroalgae as solid substrate in SSF In the beginning of this sector, it was mentioned that “the substrate for SSF must contain enough moisture to support the microorganism’s growth and metabolic activity”. As we have seen in section 1.1.2, macroalgae’s water content is over 70%, almost in every type of algae, meaning that macroalgae are a suitable substrate for SSF (Barbot, Al-Ghaili and Benz, 2016). Besides that, the substrates commonly contain some macromolecular structure, such as cellulose, starch, lignocellulose, or fibers, so the algae’s composition is adequate as well (El-Mansi et al. , 2019). Also, many other characteristics cited in this report for macroalgae make this an appropriate substrate, for example, fast growth rates, high photosynthetic efficiency and large quantity of carbohydrates, minerals, and amino acids. Macroalgae are a valuable feedstock as they serve as both physical support and nutritional source, allowing for biofuel, biochemical and biometabolites production (General et al. , 2014; Fernandes et al. , 2019). The fact that 16 17 2. MATERIALS AND METHODS 2.1. Raw material During this work, the green algae Ulva rigida was used. The macroalgae was provided dry and powder by Algaplus in 2020, a Portuguese company based in Aveiro. They were stored in a dry place, avoiding exposure to light. 2.2. Reagents The reagents used during this work are described on Table 2, such as the companies that provided those reagents. Table 2 - List of reagents used during this study. Reagents Company 3,5-dinitrosalicylic acid Acros Organics Agar Labkem Xylan (Beechwood) Megazyme CMC Sigma-Aldrich Folin-Ciocalteu reagent Panreac Gallic Acid Acros Organics Glacial acetic acid Fisher Scientific Glucose VWR Methanol Fisher Scientific Peptone Acros Organics Sulfuric acid Fluka Tween-80 Fisher Scientific 2.3. Microorganisms The filamentous fungi Aspergillus ibericus MUM 03.49 was used during this study and it was obtained from Micoteca da Universidade do Minho (MUM) culture collection (Braga, Portugal). It was revived in slants with potato dextrose agar (PDA) medium (4 g/L potato extract, 20 g/L dextrose and 15 18 g/L agar). In order to use the fungus in SSF, it was incubated in PDA medium at 25°C for 7 days. During the experiment period, the species was preserved at 4°C. 2.4. Ulva rigida characterization The macroalgae Ulva rigida used in this study was initially characterized and different analysis were performed, namely, phenolic compounds, protein, lipids, humidity, ashes, salt and cellulase. The methods used in the determination of each parameter are described in sectors further ahead. 2.4.1. Phenolic compounds determination Phenolic compounds were determined by the Folin-Ciocalteu method (Benzie and Devaki, 2017). In order to analyze them in the macroalgae, a previous extraction was performed with distilled water using a ratio of 1:10 w/v. In tubes, 100 μL of sample was added (for the blank, 0,1 mL of distilled water), as well as 2 mL of Na2CO3 at 15%, 500 μL of Folin-Ciocalteu reagent and 7,4 mL of distilled water in duplicate. The tubes were placed in a bath at 50°C for 5 minutes and, after cooling at room temperature, the tubes were vortexed. Absorbance was read at 740 nm. The calibration curve was constructed with gallic acid standard solutions between 0 g/L and 2 g/L. 2.4.2. Protein quantification The Kjeldahl method was used to determine the total nitrogen in the solid sample. This method is used in the measurement of protein content of biological materials, as so in the determination of nitrogen in inorganic materials, solids, or liquids. Kjeldahl method can be divided in two steps - the first one a digestion and the second one a titration. In the first step, a digestion of the sample is performed by heating with concentrated sulfuric acid in the presence of a catalyzer, in this case, selenium (or red Hg, the second one more efficient than the first but also with more environmental implications). This step is responsible for the reduction of organic nitrogen to ammonia, which is recovered in solution in the form of ammonium sulfate: 𝑁𝑜𝑟𝑔𝑎𝑛𝑖𝑐+𝐻2𝑆𝑂4+ 𝑐𝑎𝑡𝑎𝑙𝑖𝑠𝑒𝑟 →𝐶𝑂2+𝐻2𝑂+(𝑁𝐻4)2𝑆𝑂4 19 After digestion, the ammonium is displaced by a strong base in excess, being used aqueous NaOH at 400 g/L. This is, stoichiometrically: (𝑁𝐻4)2𝑆𝑂4+ 2𝑁𝑎𝑂𝐻 →2𝑁𝐻3+ 𝑁𝑎2𝑆𝑂4+2𝐻2𝑂 The resultant solution with NH3 is distillated with vapor that drags with it the NH3, being this recovered in a solution of boric acid: 𝑁𝐻3+𝐻3𝐵𝑂3 → 𝑁𝐻4 ++ 𝐻2𝐵𝑂3 − The borate of the acidic solution is titrated (second step) with sulfuric acid to quantify the quantity of ammonium according to the reaction ( Cole-Parmer Scientific Experts , no date; ExpotechUSA, no date; PanReac AppliChem, no date): 2𝐻++𝑆𝑂42− +2𝐻2𝐵𝑂3−+2𝑁𝐻4 +→2𝐻3𝐵𝑂3+(𝑁𝐻3)2𝑆𝑂3+ 𝐻2O The Kjeldahl method protocol consists in turning on the thermoblock (Tecator system 1007/6) which should achieve the 420°C. In each digestion tube, 0,5 g of the sample are placed and then it is added 10 mL of H2SO4 concentrated and a tablet of catalyzer selenium (Tecator S/3.5). The solution is carefully mixed and placed on the thermoblock previously heated to 420°C. After digestion, the titration is performed with the addition of an alkali solution. 2.4.3. Cellulose, hemicellulose, and lignin quantification To determine the amounts of cellulose, hemicellulose, and lignin present on the fermented solid, a quantitative acid hydrolysis (QAH) was performed, based on Hoebler et al. , 1989, with some modifications. This includes a first stage incubation with 72% wt H2SO4 at 30°C for 1 h and a second stage after dilution to 4% wt H2SO4 at 121°C for 1 h. A sample of about 0,5 g was weighted into a glass cup, 5 mL of 72 % H2SO4 were added to the cup and then they were placed in a water bath at 30°C during 1 hour with periodic agitation (every 10 minutes) - first stage. After this period, the reaction was stopped with the addition of distilled water and 20 the glass cup’s content was transferred for flasks. The waste that stayed attached to the walls was dragged with distilled water, which was added to dilute the solution up at 4% (w/w) H2SO4. The flasks were closed and introduced in the autoclave during 1 hour at 121°C - second stage. Posteriorly, the flasks were cooled and was determined the losses originated during the second stage by weighting the flasks. The entire content of each flask was filtered through a Gooch crucible with known weight. The Gooch crucibles with insoluble fraction were placed on an oven at 105°C. After 24 h the crucibles were cooled on a desiccator containing silica gel and then weighted. This determination was performed in duplicate. The filtrate was analyzed by High Performance Liquid Chromatography (HPLC) system for measure of sugars (glucose, xylose, and arabinose) and acetic acid. Using a Jasco830-IR intelligent refractive-index detector and a Varian MetaCarb 87H column. The column was eluted with 0.005 M H2SO4 and the flux was 0,5 mL/min at 60°C. Calibration curves were constructed with glucose, xylose, arabinose, and acetic acid standard solutions between 0,1 g/L e 10 g/L. With the data of sugars concentrations (glucose, xylose, arabinose, and acetic acid) was calculated the content in polymers (CP). The CP, glucan (CGn), xylan (CXn), arabinan (CArn), and acetyl groups (CGA) were calculated according to Equation A and expressed as grams of polymer per 100 grams of dry waste. 𝐶𝑃 (%)=𝐹∗ 𝑆𝐶𝐹∗ [𝑆] 𝜌∗ 𝑊+𝑊𝐻𝑆∗𝐻 𝑊𝐻𝑆∗(1−𝐻)∗100 Equation A where F is a factor which corrects degradation of sugars (1,04 for CGn, 1,088 for CXn/CArn and 1,00 for CGA), SCF is a stoichiometric correction factor to take in account the increase in molecular weight during hydrolysis (162/180 for CGn, 132/150 for CXn/CArn and 43/60 for CGA), S is the monomer concentration in g/L, ρ is the density of the analyzed dissolution in g/L (as the samples were diluted in water for HPLC analysis the value is about 1000 g/L), W is the weight of added water in grams and corrected to take account the losses during second stage of QHA, WHS is the total weight in grams of humid waste and H is the humidity in grams of water/grams of humid waste. Cellulose (grams of cellulose per 100 grams of dry waste) and hemicellulose (grams of hemicellulose per 100 grams of dry waste) content were determined according to Equation B and Equation C, respectively. 𝐶𝑒𝑙𝑙𝑢𝑙𝑜𝑠𝑒 (%)=𝐶𝐺𝑛 Equation B 21 𝐻𝑒𝑚𝑖𝑐𝑒𝑙𝑙𝑢𝑙𝑜𝑠𝑒 (%)= 𝐶𝑋𝑛+ 𝐶𝐴𝑟𝑛+ 𝐶𝐺𝐴 Equation C The increase weight of the Gooch container matches to Klason lignin, thus the content of lignin (grams of lignin per 100 grams of dry waste) was calculated according to Equation D. 𝐿𝑖𝑔𝑛𝑖𝑛 (%)= (𝑊𝐶𝐷𝑆−𝑊𝐶)∗(1−𝐶𝐸) 𝑊𝐶𝐻𝑆∗(1−𝐻) ∗100 Equation D wherein WCDS is the weight of Gooch container with dry sample in grams, WC is the weight of Gooch container in grams, WCHS is the weight of Gooch container with humid initial sample who underwent to the QHA in grams and H is the humidity in grams of water/grams of humid waste. The parameter CE is added to remove the value of ashes and is calculated by Equation E. 𝐶𝐸 =(𝑊𝐶𝐷𝑆− 𝑊𝐶 )− ((𝑊𝐶𝐷𝑆−𝑊𝐶)∗𝐴𝑠ℎ 100) Equation E 2.4.4. Ashes determination After drying the Gooch crucible at 105°C during 24 h, its content was added to a porcelain container (previously dried at 105°C for 24 hours). The container was weighted before and after the content of the Gooch crucible was added. The porcelain container with the solid was placed in the muffle furnace at 550°C for 2 hours, until constant weight. After cooling in the desiccator containing silica gel for about 15 minutes, it was weighed. Ash percentage (grams of ash per 100 grams of dry solid) is given by Equation F. 𝐴𝑠ℎ (%)= 𝑊𝐶𝐴−𝑊𝐶 (𝑊𝐶𝐻𝑆−𝑊𝐶)×(1−𝐻)×100 Equation F 22 where WCA is the weight of porcelain container with ash in grams, WC is the weight of porcelain container in grams, WCHS is the weight of porcelain container with humid waste in grams and H is the humidity in grams of water/grams of humid waste. 2.4.5. Antioxidant activity determination Directly on an Elisa plate, 200 μL of sample were pipetted and 100 μL of 2,2-Diphenyl-1picrylhydrazyl (DPPH) were added to the same well (Benzie and Devaki, 2017). For the blank, 100 μL of water were added, instead of DPPH. For the control, 200 μL of water were added, instead of the sample. The calibration curve was constructed with Trolox standard solutions with, concentrations between 3,125 and 100 microM. The final result is expressed in millimoles of Trolox equivalent per gram of dry solid substrate. To achieve this result, firstly the scavenging activity (%) is calculated with Equation G. 𝑆𝑐𝑎𝑣𝑒𝑛𝑔𝑖𝑛𝑔 𝑎𝑐𝑡𝑖𝑣𝑖𝑡𝑦 (%)= 1−𝐴𝑏𝑠𝑐 𝑀𝑒𝑎𝑛 𝑜𝑓 𝑎𝑏𝑠𝑜𝑟𝑏𝑎𝑛𝑐𝑒𝑠×100 Equation G where Absc represents absorbance corrected, or, in other words, the absorbance read minus the blank. Finally, with this result and using the calibration curve constructed, the final result is reached. 2.4.6. Salt determination Salt content was determined by adding 100 mL of water to 5 g of algae and it was stirred for 24h. After that time, the mixture was filtered, and the liquid retrieved. The liquid was added to a previously weighed cup and left to dry in an oven at 55°C for 48h. 2.4.7. Lipids quantification Total lipids were determined by Soxhlet extraction, using petroleum ether as a solvent, at 70°C using a FOSS Soxtec 8000 apparatus. 23 2.5. Sequential SSF and enzymatic hydrolysis (EH) SSF and EH were performed sequentially and it was only added a citrate buffer on the beginning of EH, following the methodology described by Fernandes et al . (Fernandes et al. , 2019). SSF by Aspergillus ibericus was conducted in Erlenmeyer flasks of 500 mL, where 10 g of U. rigida were weighted and water was added to reach a humidity of 75% (wet basis). The flasks were sterilized at 121°C during 15 min. After sterilization, in the laminar flow hood, the Erlenmeyer flasks were inoculated with 2 mL of a spore solution prepared by adding peptone (0,1% w/v) and Tween 80 (0,001% w/v), the concentration was adjusted to 1*106 cells/mL. The fermentation process was conducted for 5 days at 25°C. EH with enzymes produced by fungus were performed. Three parameters of EH were optimized using a Box-Behnken experimental design (temperature, load of solid and pH). After SSF, sequentially different quantities of citrate buffer were added to the fermented solid, to adjust the load of solid and pH, according to Table 3. The EH was carried out in orbital shaker at 150 rpm and the temperatures defined for each experiment (Table 3). These values were obtained using experimental Box-Behnken design, which is an incomplete factorial design, combined in blocks, and its main advantage is the reduction of the number of experiments, when comparing to other experimental designs (Czyrski and Sznura, 2019). It was also added thymol, which works as an antifungal to prevent the consumption of released sugars during EH (Salehi et al. , 2018). 24 Table 3 – Matrix of experiments obtained by Box-Behnken experimental design. Runs Temperature (oC) Load of solid (% w/v) pH buffer 1 44 20 4 2 39,5 20 4,6 3 39,5 20 4,6 4 35 20 4 5 44 10 4,6 6 39,5 10 4 7 44 20 5,2 8 35 20 5,2 9 39,5 10 5,2 10 35 10 4,6 11 39,5 20 4,6 12 35 30 4,6 13 39,5 30 5,2 14 44 30 4,6 15 39,5 30 4 Samples were collected at time 0, 4h, 8h, 24h, 32h, 48h and 72h. The samples collected were centrifuged at 8000 rpm for 5 minutes and stored at -20°C until analysis. By the end of the EH time, the resulting mixture was centrifuged at 9000 rpm for 10 minutes and the solid was dried at 60°C during the weekend. Afterwards, the resulting solid and the samples collected were used in different analysis described in the next sectors and the sectors before (phenolic compounds, protein, hemicellulose, cellulose, lignin, ashes, and antioxidant activity). 2.6. Reducing sugars determination Free reducing sugars were measured by the DNS method (Miller, 1959). To each tube 0,1 mL of the sample was added and 0,1 mL of DNS reagent in duplicate (for the blank measurement 0,1 mL of distilled water was used). The tubes were placed in a bath at 100°C for 5 minutes. After cooling, 1 mL of water was added to the mixture and the absorbance was read at 540 nm. 25 The calibration curve for this method was constructed with glucose standard solutions, with concentrations between 0 g/L and 4 g/L. The maximum conversion of cellulose to glucose (CGCmax) during EH was calculated following the equation described in Romani et al. (Romaní et al. , 2011) using the values of glucose analyzed by HPLC: 𝐶𝐺𝐶𝑡 =𝐶𝐺𝐶𝑚𝑎𝑥 𝑡 𝑡 + 𝑡1/2 Equation H where CGCt is the cellulose-to-glucose conversion achieved at time t, calculated as: 𝐶𝐺𝐶𝑡 =100 𝐺𝑡− 𝐺𝑡0 𝐺𝑝𝑜𝑡 Equation I whereas CGCmax is the cellulose-to-glucose conversion predicted for an infinite reaction time, t is the EH time (h), t1/2 (h) is the time needed to achieve CGC= CGCmax/2, Gt is the glucose concentration (g/L) achieved at time t, Gt0 is the glucose concentration at the beginning of the experiments, and Gpot represents the potential glucose concentration (calculated assuming total cellulose conversion into glucose). 2.7. Cellulase and xylanase quantification The quantification of cellulase and xylanase was performed according to the method described by Sousa et al. , 2020.The procedure for determination of cellulases activity was to add 250 µL of cellulase substrate (CMC 1% in 0.1 M sodium acetate buffer, pH 4.6) to test tubes and then, 250 µL of diluted sample in buffer. The test tubes were placed on a bath at 50°C for 30 minutes. After 30 minutes it was added 500 µL of DNS and then the test tubes were placed on a bath at 100°C for 5 minutes. Finally, 5 mL of distilled water was added to each tube and the absorbance was read at 540 nm. The blank was performed with sodium acetate buffer and the addition of the sample after the 30 minutes incubation. A calibration curve was constructed with glucose standard solutions in buffer between 0 g/L and 2 g/L. The procedure to determine the xylanases activity was the same as for the determination of cellulases activity but the duration of the reaction was only 15 minutes instead of 30 minutes and the substrate solution was beechwood xylan (2%). 32 Figure 3 -- Response surface for antioxidant activity as a function of pH and load of solid. Table 6 lists regression coefficients and their statistical significance, as well as the statistical parameters that measure the suitability of the model. The determination coefficient (R2) was 0.995 for antioxidant activity and 0.953 for total phenolic compounds, which demonstrates a satisfactory adjustment of the model. 33 Table 6 - Regression coefficients of model parameters antioxidant activity and phenolic compounds. Coefficients Antioxidant activity TPC Constant 284.13*** 0.475*** X1 21.10*** 0.034*** X1 · X1 -55.94*** -0.180*** X2 60.82*** 0.074*** X2 · X2 -5.44 0.010** X3 -70.84*** -0.004** X3 · X3 -95.79*** -0.005** X1 · X2 6.77 -0.079*** X1 · X3 -59.68*** -0.007** X2 · X3 108.68*** -0.033*** Coefficients of determination R2 0.995 0.953 R2 adj 0.985 0.868 X1: Temperature; X2: Load of solid; X3: pH; ***significant at 99%; **significant at 95%; *significant at 90%; TPC: total phenolic compounds; CGCmax: conversion cellulose to glucose maximum 34 3.3.2. Release of sugars during EH The sugar release during EH was evaluated on samples collected during EH. The results are presented in terms of maximum conversion of glucose to cellulose (Table 7). Table 7 - Results observed and predicted of sugar release during EH. Runs CGCmax (%) Observed Predicted 1 71 74 2 22 19 3 15 19 4 14 16 5 62 61 6 15 14 7 61 59 8 38 35 9 12 16 10 23 22 11 21 19 12 22 23 13 18 20 14 66 67 15 22 18 The values observed were, in all rounds, similar to the values predicted, being that the biggest difference was verified in round 3 (it was observed 14,85% when it was predicted 19%). According to Table 7, the highest value was predicted to appear in round 1 (conditions: 44°C; load of solid 20% w/v; pH 4,6) and the observed value confirmed this tendency. On the other hand, the lowest rate of conversion from cellulose to glucose was expected in round 6 (conditions: 39,5°C; load of solid 10% w/v; pH 4), however it was verified in round 9 (conditions: 39,5°C; load of solid 10% w/v; pH 5,2), that has the same temperature and load of solid as round 6 but was performed with buffer with a different pH. It can be also verified that the rounds performed at 44°C allowed a higher conversion of cellulose to glucose that 35 others performed at lower temperatures. A study performed by Liu et al. also achieved a greater degree of conversion with an increase of the temperature (Liu et al. , 2012). In another study, conducted by Harun and Danquah, the optimal temperature for the conversion of cellulose to glucose on EH was determined to be 40°C so, although in this study the conversion increases with the increase of temperature until 44°C, it could be expected that the conversion would soon decrease if higher temperatures were tested (Harun and Danquah, 2011). Figure 4 shows the effect of temperature and load of solid in the conversion of cellulose to glucose. The load of solid used in EH step did not interfere with the conversion of cellulose to glucose, however, an increase in the temperature at which the EH was performed correlates to an increase on the conversion of cellulose to glucose. Regression coefficients and their statistical significance, as well as the statistical parameters that measure the suitability of the model for the conversion of cellulose to glucose are presented in Annex 1. Figure 4 - Response surface for conversion of cellulose to glucose as a function of temperature and load of solid. 36 3.4. Stability of lignocellulolytic enzymes during EH In all the runs tested, the stability of xylanase decreased during the 72 hours of EH. The results are presented in Figure 5. Figure 5 - Kinetics of xylanase activity during 72 hours of EH. Although all rounds present a reduction in xylanase activity, it is clear that the decrease is less significative in some rounds than others. For instance, round 11 (conditions: 39,5°C; load of solid 20% w/v; pH 4,6) has the lower decay of activity, while round 5 (conditions: 44°C; load of solid 10% w/v; pH 4,6) has the higher decay of activity. Furthermore, all the five rounds that had a higher decay rate (Rounds 1, 4, 5, 9 and 14) were performed under at least one of these conditions: 44°C and pH 4. On the contrary, in the five rounds that had a slower decay of activity (2, 8, 11, 12 and 13) none of these conditions was implemented. In what concerns the load of solid, it was not possible to correlate with the enzyme activity, which means that the enzyme stability was mainly affected by temperature and pH. In what concerns cellulase, its activity did not have a clear behavior, which hinders the analysis of the results, presented in Figure 6. 0,0 2,0 4,0 6,0 8,0 10,0 12,0 14,0 16,0 010 20 30 40 50 60 70 Enzyme activity (U/mL) Time (h) R2 R1 R3 R4 R5 R6 R7 R8 R9 R11 R12 R13 R14 R15 R10 37 Figure 6 - Kinetics of cellulase activity between 0h and 72h of EH. In some rounds, the enzyme activity increases from time 0 to time 24 and then decreases. On the other hand, in round 11 (conditions: 39,5°C; load of solid 20% w/v; pH 4,6) the cellulase activity continuously increased from 0 to 72h. Additionally, in round 10 (conditions: 35°C; load of solid 10% w/v; pH 4,6) the final value is also higher than the value at 24h, however, there is an increase between 24h and 48h and a decrease between 48h and 72h. When comparing the rounds with higher and lower decay of activity during time, the rounds with a bigger decay of activity (1, 7, 12, 14 and 15) were mainly performed at higher temperatures except for round 12. These rounds also did not have a pH in common, but all of them were performed with a load of solid of 20% or 30%. By contrast, in rounds 2,4,5,6 and 9, the decline of activity was slower than in the other rounds. Similarly, these rounds have neither temperature or pH in common, being that these rounds were performed at all three temperatures and pH. In what concerns load of solid, all five rounds that had a slower decline of cellulase activity were performed either with 10 or 20% of load of solid. The main conclusion that can be withdrawn from these results is than cellulase is more stable than xylanase. 0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0 4,5 010 20 30 40 50 60 70 Enzyme activity (U/mL) Time (h) R2 R1 R3 R4 R5 R6 R7 R8 R9 R11 R12 R13 R14 R15 R10 38 3.5. Characterization of solid after sequential SSF and EH The resulting solids were analyzed to determine the amount of lignin, ashes, and protein. The results are presented in Table 8 and Figures 7 and 8, respectively. Table 8 - Results observed and predicted by the model of the content in protein in the solid after SSF+HE. Runs Protein (mg/g) Observed Predicted 1 206.07 ± 0.51 215.4 2 234 ± 11 235 3 230 ± 4 235 4 242 ± 4 274 5 226 ± 15 208 6 249 ± 18 225 7 203 ± 0 218 8 236 ± 3 271 9 267 ± 24 253 10 256 ± 4 288 11 231 ± 0 235 12 235 ± 7 227 13 229 ± 1 188 14 230 ± 1 195 15 239 ± 12 216 The lowest value of protein was in round 7 (conditions: 44°C; load of solid 20% w/v; pH 5,2) and the maximum value was achieved in round 9 (conditions: 39,5°C; load of solid 10% w/v; pH 5,2). The mean of the values is 234 and it can be observed that all the rounds performed at 35°C are above the mean, as well as two of the rounds performed at 39,5°C, with pH 4. In what concerns ashes quantification, bigger differences are presented between rounds performed. The lowest value (142,08±57,19 g/kg) was obtained in round 6 (conditions: 39,5°C; load of solid 10% w/v; pH 4), while the highest values ( 464,07±146,30 and 491,70±77,13 g/kg) were achieved 39 in rounds 3 (conditions: 39,5°C; load of solid 20% w/v; pH 4,6) and 14 (conditions: 44°C; load of solid 30% w/v; pH 4,6), respectively. Regression coefficients and their statistical significance, as well as the statistical parameters that measure the suitability of the model for protein quantification are presented in Annex 1. The quantification of lignin had maximum values (335,98±29,32 and 305,87±33,82 g/kg) in rounds 9 (conditions: 39,5°C; load of solid 10% w/v; pH 5,2) and 10 (conditions: 35°C; load of solid 10% w/v; pH 4,6), respectively. The lowest amount of lignin (205,58±18,18) was determined in round 7 (conditions: 44°C; load of solid 20% w/v; pH 5,2). Figure 7 - Ashes quantification on the resulting solid after sequential SSF and EH. The results represent the average of two independent experiments and error bars represent SD. Letters above each bar indicate the results of Tukey’s test (P < 0.05); values with shared letters in the same graph are not significantly different. 40 Figure 8 - Lignin quantification on the resulting solid after sequential SSF and EH. The results represent the average of two independent experiments and error bars represent SD. Letters above each bar indicate the results of Tukey’s test (P < 0.05); values with shared letters in the same graph are not significantly different. Figure 9 shows the comparison between the maximum value obtained in each determination and the initial amount present in Ulva rigida . In comparison with the results obtained for U. rigida before fermentation (169,08±0,66 g/kg), protein always increased, with a maximum increment of 57,8%, which is a significant difference. The difference between lignin concentration after SSF+HE and initial amount in algae is also significative and it was more than 10 times higher than the initial amount present in Ulva rigida . On the other hand, the quantification of ashes shows that the difference between the amount already present in Ulva rigida is not significantly different from the values presented after SSF+HE. Protein in the final solid increases due to the presence of protein from the fungal biomass. Also, the increase of ashes and lignin, as well as proteins, may be due to an effect of concentration in the final solid after EH. 41 Figure 9 - Comparison between the initial values present in Ulva rigida and the maximum values obtained after SSF+HE. The results represent the average of two independent experiments and error bars represent SD. Letters above each bar indicate the results of Tukey’s test (P < 0.05); values with shared letters in the same graph are not significantly different. 3.6. Optimal conditions of sequential SSF and EH The optimal conditions are presented in Table 9 for each variable separately in specific and also a multiple optimization for all variables at once. Table 9 - Optimum conditions for each variable predicted by the model. Dependent variables Temperature (°C) Load of solid (% w/v) pH Antioxidant activity (μmol TE/g) TPC (mg/mL) Protein (mg/g) CGCmax (%) Single optimization 41 30 4.7 345 39 30 4.0 0.59 35 10 5.2 301 45 27 4.0 75 Multiple optimization 44 30 4.1 279 0.47 231 61 b a a ’ a ’ a ’’ b ’’ 48 55(6), pp. 937–942. doi: 10.1590/S1516-89132012000600018. Leandro, A., Pereira, L. and Gonçalves, A. M. M. (2020) ‘Diverse applications of marine macroalgae’, Marine Drugs , 18(1), pp. 1–15. doi: 10.3390/md18010017. Leite, P. et al. (2016) ‘Ultrasounds pretreatment of olive pomace to improve xylanase and cellulase production by solid-state fermentation’, Bioresource Technology . Elsevier Ltd, 214, pp. 737–746. doi: 10.1016/j.biortech.2016.05.028. Leite, P. et al. (2019) ‘Simultaneous production of lignocellulolytic enzymes and extraction of antioxidant compounds by solid-state fermentation of agro-industrial wastes’, Industrial Crops and Products . Elsevier, 137(May), pp. 315–322. doi: 10.1016/j.indcrop.2019.04.044. Liu, W. et al. (2012) ‘Efficient conversion of cellulose to glucose, levulinic acid, and other products in hot water using SO2 as a recoverable catalyst’, Industrial and Engineering Chemistry Research , 51(47), pp. 15503–15508. doi: 10.1021/ie302317t. Liu, Y. et al. (2014) ‘Solid-supported microorganism of Burkholderia cenocepacia cultured via solid state fermentation for biodiesel production: Optimization and kinetics’, Applied Energy . Elsevier Ltd, 113, pp. 713–721. doi: 10.1016/j.apenergy.2013.08.009. Lopes, D. et al. (2019) ‘Lipidomic signature of the green macroalgae Ulva rigida farmed in a sustainable integrated multi-trophic aquaculture’, Journal of Applied Phycology . Journal of Applied Phycology, 31(2), pp. 1369–1381. doi: 10.1007/s10811-018-1644-6. López-Simeon, R. et al. (2012) ‘Free-lignin cellulose obtained from agar industry residues using a continuous and minimal solvent reaction/extraction methodology’, RSC Advances , 2(32), pp. 12286– 12297. doi: 10.1039/c2ra22185c. Magro, A. E. A. et al. (2019) ‘Solid-state fermentation as an efficient strategy for the biotransformation of lentils: enhancing their antioxidant and antidiabetic potentials’, Bioresources and Bioprocessing . Springer Singapore, 6(1). doi: 10.1186/s40643-019-0273-5. Marín, M., Sánchez, A. and Artola, A. (2019) ‘Production and recovery of cellulases through solid-state fermentation of selected lignocellulosic wastes’, Journal of Cleaner Production . Elsevier Ltd, 209, pp. 937–946. doi: 10.1016/j.jclepro.2018.10.264. Markets and Markets (no date). Available at: https://www.marketsandmarkets.com/MarketReports/commercial-seaweed-market-152763701.html (Accessed: 13 March 2020). Martins, S. et al. (2011) ‘Bioactive phenolic compounds: Production and extraction by solid-state fermentation. A review’, Biotechnology Advances . Elsevier Inc., 29(3), pp. 365–373. doi: 10.1016/j.biotechadv.2011.01.008. Masri, M. A. et al. (2018) ‘A waste-free, microbial oil centered cyclic bio-refinery approach based on flexible macroalgae biomass’, Applied Energy . Elsevier, 224(April), pp. 1–12. doi: 10.1016/j.apenergy.2018.04.089. Mezghani, S. et al. (2016) ‘Characterization of Phenolic Compounds of Ulva rigida (Chlorophycae) and Its Antioxidant Activity’, European Journal of Medicinal Plants , 12(1), pp. 1–9. doi: 10.9734/ejmp/2016/22935. Miller, G. L. (1959) ‘Use of Dinitrosalicylic Acid Reagent for Determination of Reducing Sugar’, Analytical Chemistry , 31(3), pp. 426–428. doi: 10.1021/ac60147a030. 49 Miranda, M., Lopez-Alonso, M. and Garcia-Vaquero, M. (2017) ‘Macroalgae for Functional Feed Development: Applications in Aquaculture, Ruminant and Swine Feed Industries’, p. 6. Mukherjee, A. K., Adhikari, H. and Rai, S. K. (2008) ‘Production of alkaline protease by a thermophilic Bacillus subtilis under solid-state fermentation ( SSF ) condition using Imperata cylindrica grass and potato peel as low-cost medium : Characterization and application of enzyme in detergent formulation’, 39, pp. 353–361. doi: 10.1016/j.bej.2007.09.017. Novelli, P. K., Barros, M. M. and Fleuri, L. F. (2016) ‘Novel inexpensive fungi proteases: Production by solid state fermentation and characterization’, Food Chemistry . Elsevier Ltd, 198, pp. 119–124. doi: 10.1016/j.foodchem.2015.11.089. Oliveira, F. et al. (2016) ‘Olive pomace valorization by Aspergillus species: lipase production using solidstate fermentation’, Journal of the science of food and agriculture , 96(10), pp. 3583–3589. doi: 10.1002/jsfa.7544. Oliveira, F. et al. (2017) ‘Optimization of lipase production by Aspergillus ibericus from oil cakes and its application in esterification reactions’, Food and Bioproducts Processing . Institution of Chemical Engineers, 102, pp. 268–277. doi: 10.1016/j.fbp.2017.01.007. Pal, A. and Khanum, F. (2010) ‘Production and extraction optimization of xylanase from Aspergillus niger DFR-5 through solid-state-fermentation’, Bioresource Technology . Elsevier Ltd, 101(19), pp. 7563–7569. doi: 10.1016/j.biortech.2010.04.033. Pandey, A. (2003) ‘Solid-state fermentation’, 13(November 2001), pp. 81–84. Pangestuti, R. and Kim, S. K. (2015) ‘An Overview of Phycocolloids: The Principal Commercial Seaweed Extracts’, Marine Algae Extracts: Processes, Products, and Applications , 1–2, pp. 319–330. doi: 10.1002/9783527679577.ch19. PanReac AppliChem (no date) ‘Nitrogen Determination by Kjeldahl Method’. Patarra, R. F. et al. (2011) ‘Nutritional value of selected macroalgae’, Journal of Applied Phycology , 23(2), pp. 205–208. doi: 10.1007/s10811-010-9556-0. Pereira, R. and Yarish, C. (2008) ‘Mass Production of Marine Macroalgae’, Encyclopedia of Ecology, Five-Volume Set , (May 2016), pp. 2236–2247. doi: 10.1016/B978-008045405-4.00066-5. Plaza, M., Cifuentes, A. and Ibáñez, E. (2008) ‘In the search of new functional food ingredients from algae’, Trends in Food Science and Technology , 19(1), pp. 31–39. doi: 10.1016/j.tifs.2007.07.012. Rodrigues, D. et al. (2015) ‘Chemical composition of red, brown and green macroalgae from Buarcos bay in Central West Coast of Portugal’, Food Chemistry . Elsevier Ltd, 183, pp. 197–207. doi: 10.1016/j.foodchem.2015.03.057. Roesijadi, G., Jones, S. B. and Zhu, Y. (2010) ‘Macroalgae as a Biomass Feedstock : A Preliminary Analysis’, Analysis , (September), pp. 1–50. doi: 10.2172/1006310. Romaní, A. et al. (2011) ‘Eucalyptus globulus wood fractionation by autohydrolysis and organosolv delignification’, Bioresource Technology , 102(10), pp. 5896–5904. doi: 10.1016/j.biortech.2011.02.070. Ross, A. B. et al. (2008) ‘Classification of macroalgae as fuel and its thermochemical behaviour’, Bioresource Technology , 99(14), pp. 6494–6504. doi: 10.1016/j.biortech.2007.11.036. 50 Sadh, P. K., Duhan, S. and Duhan, J. S. (2018) ‘Agro-industrial wastes and their utilization using solid state fermentation: a review’, Bioresources and Bioprocessing . Springer Berlin Heidelberg, 5(1). doi: 10.1186/s40643-017-0187-z. Salehi, B. et al. (2018) ‘Thymol, thyme, and other plant sources: Health and potential uses’, Phytotherapy Research , 32(9), pp. 1688–1706. doi: 10.1002/ptr.6109. Salgado, J. M. et al. (2014a) ‘Integrated use of residues from olive mill and winery for lipase production by solid state fermentation with Aspergillus sp.’, Applied Biochemistry and Biotechnology , 172(4), pp. 1832–1845. doi: 10.1007/s12010-013-0613-4. Salgado, J. M. et al. (2014b) ‘Screening of winery and olive mill wastes for lignocellulolytic enzyme production from Aspergillus species by solid-state fermentation’, Biomass Conversion and Biorefinery , 4(3), pp. 201–209. doi: 10.1007/s13399-013-0100-8. Sambusiti, C. et al. (2015) ‘Algae as promising feedstocks for fermentative biohydrogen production according to a biorefinery approach: A comprehensive review’, Renewable and Sustainable Energy Reviews . Elsevier, 44, pp. 20–36. doi: 10.1016/j.rser.2014.12.013. Satpati, G. G. and Pal, R. (2011) ‘Biochemical composition and lipid characterization of marine green alga Ulva rigida-a nutritional approach’, J. Algal Biomass Utln. , 2(4), pp. 10–13. seaExpert (no date). Available at: https://seaexpert-azores.com/wp/edible-seaweed-supply/ (Accessed: 15 December 2019). Serra, R. et al. (2006) ‘ Aspergillus ibericus : a new species of section Nigri isolated from grapes ’, Mycologia , 98(2), pp. 295–306. doi: 10.1080/15572536.2006.11832702. Shrestha, P. et al. (2010) ‘Ethanol production via in situ fungal saccharification and fermentation of mild alkali and steam pretreated corn fiber’, Bioresource Technology . Elsevier Ltd, 101(22), pp. 8698– 8705. doi: 10.1016/j.biortech.2010.06.089. Siddhanta, A. K. et al. (2009) ‘Profiling of cellulose content in Indian seaweed species’, Bioresource Technology . Elsevier Ltd, 100(24), pp. 6669–6673. doi: 10.1016/j.biortech.2009.07.047. Singhania, R. R. et al. (2015) ‘An integrative process for bio-ethanol production employing SSF produced cellulase without extraction’, Biochemical Engineering Journal . Elsevier B.V., 102, pp. 45–48. doi: 10.1016/j.bej.2015.01.002. Smit, A. J. (2004) ‘Medicinal and pharmaceutical uses of seaweed natural products: A review’, Journal of Applied Phycology , 16(4), pp. 245–262. doi: 10.1023/B:JAPH.0000047783.36600.ef. Soccol, C. R. et al. (2017) ‘Recent developments and innovations in solid state fermentation’, Biotechnology Research and Innovation . Sociedade Brasileira de Biotecnologia, 1(1), pp. 52–71. doi: 10.1016/j.biori.2017.01.002. Sode, S. et al. (2013) ‘Bioremediation of reject water from anaerobically digested waste water sludge with macroalgae (Ulva lactuca, Chlorophyta)’, Bioresource Technology . Elsevier Ltd, 146, pp. 426–435. doi: 10.1016/j.biortech.2013.06.062. Sousa, D. et al. (2018) ‘Mediterranean agro-industrial wastes as valuable substrates for lignocellulolytic enzymes and protein production by solid-state fermentation’, Journal of the Science of Food and Agriculture , 98(14), pp. 5248–5256. doi: 10.1002/jsfa.9063. Sousa, D. F. et al. (2020) ‘Production of bioactive compounds by solid-state fermentation of oilseed 51 cakes’. de Souza Barros, C., Teixeira, V. L. and Paixão, I. C. N. P. (2015) ‘Seaweeds with anti-herpes simplex virus type 1 activity’, Journal of Applied Phycology , 27(4), pp. 1623–1637. doi: 10.1007/s10811-0140447-7. Sudhakar, K. et al. (2018) ‘An overview of marine macroalgae as bioresource’, Renewable and Sustainable Energy Reviews . Elsevier Ltd, 91(November 2017), pp. 165–179. doi: 10.1016/j.rser.2018.03.100. Suganya, T. et al. (2016) ‘Macroalgae and microalgae as a potential source for commercial applications along with biofuels production: A biorefinery approach’, Renewable and Sustainable Energy Reviews . Elsevier, 55, pp. 909–941. doi: 10.1016/j.rser.2015.11.026. Taboada, C., Millán, R. and Míguez, I. (2010) ‘Composition, nutritional aspects and effect on serum parameters of marine algae Ulva rigida’, Journal of the Science of Food and Agriculture , 90(3), pp. 445–449. doi: 10.1002/jsfa.3836. Tacon, A. G. J., Hasan, M. R. and Metian, M. (2011) Demand and supply of feed ingredients for farmed fish and crustaceans : Trends and prospects , FAO Fisheries and Aquaculture Technical Paper . Available at: http://www.fao.org/docrep/015/ba0002e/ba0002e.pdf. Tewalt, J. and Schilling, J. (2010) ‘Assessment of saccharification efficacy in the cellulase system of the brown rot fungus Gloeophyllum trabeum’, Applied Microbiology and Biotechnology , 86(6), pp. 1785– 1793. doi: 10.1007/s00253-010-2462-1. Thomas, L., Larroche, C. and Pandey, A. (2013) ‘Current developments in solid-state fermentation’. Elsevier B.V., 81, pp. 146–161. Treichel, H. et al. (2020) Utilising Biomass in Biotechnology , Green Energy and Technology . Trivedi, N. et al. (2015) ‘Solid state fermentation (SSF)-derived cellulase for saccharification of the green seaweed Ulva for bioethanol production’, Algal Research . Elsevier B.V., 9, pp. 48–54. doi: 10.1016/j.algal.2015.02.025. Trivedi, N. et al. (2016) ‘An integrated process for the extraction of fuel and chemicals from marine macroalgal biomass’, Scientific Reports . Nature Publishing Group, 6, pp. 1–8. doi: 10.1038/srep30728. Villanueva-Lozano, H. et al. (2017) ‘Successful treatment of Talaromyces amestolkiae pulmonary infection with voriconazole in an acute lymphoblastic leukemia patient’, Journal of Infection and Chemotherapy , 23(6), pp. 400–402. doi: 10.1016/j.jiac.2016.12.017. van der Wal, H. et al. (2013) ‘Production of acetone, butanol, and ethanol from biomass of the green seaweed Ulva lactuca’, Bioresource Technology . Elsevier Ltd, 128, pp. 431–437. doi: 10.1016/j.biortech.2012.10.094. Wan, A. H. L. et al. (2019) ‘Macroalgae as a sustainable aquafeed ingredient’, Reviews in Aquaculture , 11(3), pp. 458–492. doi: 10.1111/raq.12241. Wang, F. et al. (2019) ‘Fungal laccase production from lignocellulosic agricultural wastes by solid-state fermentation: A review’, Microorganisms , 7(12). doi: 10.3390/microorganisms7120665. Wells, M. L. et al. (2017) ‘Algae as nutritional and functional food sources: revisiting our understanding’, Journal of Applied Phycology . Journal of Applied Phycology, 29(2), pp. 949–982. doi: 52 10.1007/s10811-016-0974-5. Ben Yahmed, N. et al. (2016) ‘A biorefinery concept using the green macroalgae Chaetomorpha linum for the coproduction of bioethanol and biogas’, Energy Conversion and Management . Elsevier Ltd, 119, pp. 257–265. doi: 10.1016/j.enconman.2016.04.046. Yoon, L. W. et al. (2014) ‘Fungal solid-state fermentation and various methods of enhancement in cellulase production’, Biomass and Bioenergy . Elsevier Ltd, 67, pp. 319–338. doi: 10.1016/j.biombioe.2014.05.013. Zahra, W. et al. (2019) Economic importance of medicinal plants in Asian countries , Bioeconomy for Sustainable Development . doi: 10.1007/978-981-13-9431-7_19. Zollmann, M. et al. (2019) ‘Green technology in green macroalgal biorefineries’, Phycologia . Taylor & Francis, 58(5), pp. 516–534. doi: 10.1080/00318884.2019.1640516. 53 ANNEXES Annex 1 - Regression coefficients of model parameters of protein and conversion of cellulose to glucose. Coefficients Protein CGCmax Constant 234.65*** 19.00*** X1 -27.96*** 20.53*** X1 · X1 9.66** 26.60*** X2 -18.59*** 1.86 X2 · X2 -14.53*** -2.37 X3 -0.09 0.96 X3 · X3 0.40 0.38 X1 · X2 12.00 1.32 X1 · X3 1.43 -8.38** X2 · X3 -13.75* -0.24 R2 0.954 0.985 R2 adj 0.872 0.959 X1: Temperature; X2: Load of solid; X3: pH; ***significant at 99%; **significant at 95%; *significant at 90%; TPC: total phenolic compounds; CGCmax: conversion cellulose to glucose maximum