A process for extracting proteins from photosynthetic biomass

Abstract

A process for extracting proteins from photosynthetic biomass. The process comprises to mix a photosynthetic biomass sample with an aqueous surfactant-containing solution, to extract proteinaceous material from the photosynthetic biomass sample, to separate the proteinaceous material into a liquid phase comprising solubilized proteins and a solid phase comprising non-soluble material. The formed solid phase is then mixed with a protein extraction solution, proteinaceous material is extracted from the solid phase and the proteinaceous material is separated into a liquid phase comprising solubilized proteins and a solid phase comprising non-soluble material. The two formed liquid phases with solubilized proteins produced in the process are collected.

Description

[0001] A PROCESS FOR EXTRACTING PROTEINS FROM PHOTOSYNTHETIC BIOMASS TECHNICAL FIELD The present disclosure relates to a process for extracting proteins from photosynthetic biomass. BACKGROUND As the global population grows and the demand for protein continues to increase, traditional land-based protein sources may face supply constraints in the future. Alternative terrestrial or aquatic biomasses can help relieve the strain on traditional land-based protein supply, such as livestock and soybean. The development of industrial processes to extract protein from photosynthetic aquatic or terrestrial biomass is still in its infancy due to high complexity of most existing protocols, low yields, and / or low protein purities. Aiming to tackle these challenges, especially the yield issue, laboratory processes often involve a freeze-thawing cycle (Abdollahi, M., et al., (2019). Effect of stabilization method and freeze / thaw-aided precipitation on structural and functional properties of proteins recovered from brown seaweed (Saccharina latissima), Food Hydrocolloids, 96 (March), 140–150), or addition of a physical osmotic shock incubation step at the start of the process to burst seaweed cells, thereby liberating aqueous soluble protein along with e.g. organelles has also been used (Vilg, JV., et al., (2017). pH-driven solubilization and isoelectric precipitation of proteins from the brown seaweed Saccharina Iatissima - effects of osmotic shock, water volume and temperature. Journal of Applied Phycology, 29, 585-593). Other laboratory methods directly recover the solubilized protein via costly freeze- drying (Kazir, M., et al., (2019). Extraction of proteins from two marine macroalgae, Ulva sp. and Gracilaria sp., for food application, and evaluating digestibility, amino acid composition and antioxidant properties of the protein concentrates. Food Hydrocolloids, 87, 194–203.). US20220232877A1 discloses processes for making, extracting, and purifying a protein preparation from plant material. CN115536726A describes a method for extracting plant protein, such as leaf protein from fresh alfalfa. CN114426965A shows a method for simultaneous extraction of DNA, RNA and protein from a plant tissue sample. Existing protocols, as mentioned above, have targeted mainly aqueous proteins and overlooked membrane proteins, constituting a large fraction of the total proteins in photosynthetic biomasses, particularly in chloroplasts and their thylakoids. This contributes to the limited yields and calls for methods targeting both aqueous and membrane-bound protein fractions. Surfactants are known to solubilize the latter protein fraction plus releasing membrane lipids (US9994791 B1; Tamayo, A. T., Boom, R. M., & van der Goot, A. J. (2017). Understanding leaf membrane protein extraction to develop a food-grade process. Food Chemistry, 217, 234–243). A complete extraction protocol encompassing solubilization and recovery steps of aqueous and membrane- bound protein is yet, however, to be disclosed for photosynthetic biomasses. SUMMARY OF THE INVENTION It is an object of the present disclosure to provide an improved or at least an alternative process for extracting proteins from photosynthetic biomass. The invention is defined by the appended independent patent claims. Non-limiting embodiments emerge from the dependent claims, the appended drawings, and the following description. According to a first aspect, there is provided a process for extracting proteins from photosynthetic biomass, the process comprising steps of: a) mixing a photosynthetic biomass sample with an aqueous surfactant-containing solution, b) extracting proteinaceous material from the photosynthetic biomass sample in the surfactant-containing solution, c) separating the proteinaceous material into a liquid phase comprising solubilized proteins and a solid phase comprising non-soluble material, d) collecting the solid phase, e) mixing the solid phase with a protein extraction solution, f) extracting proteinaceous material from the solid phase in the protein extraction solution, g) separating the proteinaceous material into a liquid phase comprising solubilized proteins and a solid phase comprising non-soluble material, h) collecting the liquid phases with solubilized proteins produced in steps c) and g). A photosynthetic biomass sample may comprise only one type of photosynthetic biomass, such as seaweed, duckweed, seagrass, microalgae, cyanobacteria, aquatic leaves, terrestrial leaves or grass, or the sample may comprise a mixture of two or more different types of such photosynthetic biomass. The sample can also consist of residuals from pre-fractionation of the mentioned biomasses, e.g., from polysaccharide recovery processes. The photosynthetic biomass sample may be subject to pre-treatment before being mixed with the surfactant-containing solution. Such pre-treatment may include disintegration, e.g. cutting, grinding or mincing of the biomass sample into smaller pieces, and / or the biomass could be pre-treated with e.g. ethanol. Disintegration may in one embodiment be performed on a fresh, frozen or dried biomass sample. The surfactant-containing solution may be an aqueous surfactant solution. The surfactant-containing solution may essentially comprise surfactant and water or a buffer such as a weak buffer. The buffer may e.g. be Tris-HCl buffer, Tricine buffer, Bicine buffer or HEPES buffer. The pH range of the surfactant solution may be pH 5-12. Enzymes can be included in the surfactant-containing solution and may be proteases (e.g. papain, corolase, pepsin, flavourzyme, alcalase, ^-chymotrypsin, trypsin, bromealin) and polysaccharidases (e.g., ulvan lyases, cellulases, hemicellulases, pectinases, agarases, carrageenases, xylanases, alginate lyases, amylases), and combinations thereof dissolved in appropriate buffer or distillate water. The enzymes can be added for example at concentrations ranging from 1 to 15000 U g-1biomass dry weight. Note that one unit (U) can be defined as the enzymatic activity that liberates 1 μmol of product per minute at a given temperature. No further additives such as a chaotrophic denaturant (e.g. urea, guanidine hydrochloride or guanidine isothiocyanate) are present in the surfactant-containing solution. There is further no adsorbent, reducing agent, or organic solvent in the surfactant-containing solution. In the described process, protein denaturation is not wanted, why such additives are not present in the surfactant-containing solution. Here, reducing agent does not include added oxidation inhibitors that may be reducing, or act as e.g., radical scavengers or chelators. The surfactant-containing solution is a chaotrophic denaturant-free solution or comprises less than 0.01 mol / L of such a denaturant. The surfactant-containing solution is free of added reducing agents or comprises less than 0.1 mmol / L of an added reducing agent. The surfactant-containing solution is free of adsorbents (such as high molecular amphoteric polymers) or comprises less than 0.01% (w / v) of an adsorbent. The surfactant-containing solution is free of organic solvents. The surfactant-containing solution is a chaotrophic denaturant-free solution or comprises less than 0.01 mol / L of such a denaturant, is free of reducing agents or comprises less than 0.1 mmol / L of a reducing agent, is free of adsorbents or comprises less than 0.01% % (w / v) of an adsorbent. The surfactant-containing solution is further free of organic solvents. The surfactant-containing solution comprises essentially or consists of surfactant and water or buffer and possible antioxidants and enzymes, such as proteases or polysaccharidases. The step of extracting proteinaceous material from the photosynthetic biomass sample or from the solid phase (steps b) and f)) comprises solubilization and precipitation of proteins. Solubilization may be performed by subjecting biomass / solid phase and extraction solution to different cell disruptive methods. One example is homogenization. Homogenization may be performed using e.g. a mixer. The mixing and extracting steps (step a) and e) and step b) and f), respectively) may be two sequential steps or may be performed simultaneously. The extraction steps (step b) and f)) may be followed by an optional step of incubation and / or by an ultrasound treatment. During incubation any foam formed during solubilisation may be allowed to dissipate. The incubation time may be 0-180 min, preferably 0-60 min, and the incubation temperature 1-70 °C, preferably 4-25 °C. The process steps may be performed at room temperature or at a lower temperature such as 4-8 ⁰C, when processing temperature sensitive photosynthetic biomass samples. Dark conditions may be applied to avoid photo-oxidation stimulated e.g. by pigments. The proteinaceous material is separated or recovered (step c) and g)) into a liquid phase, the supernatant, comprising solubilized proteinaceous material, and a solid phase, such as a pellet, comprising non-soluble material (comprising non-solubilized proteinaceous material and non-protein material of the biomass). The solid phase, pellet, collected in step d) may by subject to pre-treatment before being mixed with the protein extraction solution. Such pre-treatment may include disintegration, e.g. cutting, grinding or mincing of the solid phase material into smaller pieces, and / or the solid phase material could be pre-treated with e.g. ethanol. Disintegration may in one embodiment be performed on a fresh, frozen or dried solid phase material. In step e) the protein extraction solution may be water, such as distilled water, water or distilled water adjusted to a pH of 6 or higher, such as 6-14 or 8-11, or 8-12 or a surfactant-containing solution as described above and being used in step a). The protein extraction solution may further comprise enzymes, such as proteases or polysaccharidases. The pH of the protein extraction solution may be adjusted by adding sufficient amounts of e.g. NaOH. The above described process is a two-step process, in which solubilized proteinaceous material is collected in liquid phase from the treated photosynthetic biomass sample in a first process cycle (steps a) to c)) and in a second process cycle (steps d) to g)) from the solid phase formed in the first process cycle. The above described process is potentially scalable and may give a total amino acid yield of up to 20%. Using surfactant-containing solution in the first processing cycle (steps a)-c)) results in significantly more proteinaceous material being solubilized than when using traditional alkaline solubilization at e.g. pH 12 in these steps. These findings may be related to the nature of the surfactant, which is able to solubilize membrane proteins by disrupting lipid bilayers such as those found in the envelope membranes of chloroplasts as well as in the thylakoid membranes. Other factors contributing to the enhanced solubilization may be attributed to the improved cell wall permeability due to contact with the surfactant. In the second processing cycle, nitrogen solubility yield was significantly higher for the biomass initially treated with surfactant-containing solution compared to alkali-treated ones, regardless of the protein extraction solution employed in the second processing cycle. This may be due to that cell walls after surfactant treatment become more permeable, as well as showing disrupted lipid bilayer structures, to the extracting solution of the second cycle. By combining the supernatants of the two process cycles in the above discussed process, the nitrogen solubility of proteinaceous material collected in step h) may be as high as 60% versus 48% for traditional alkaline solubilisation. In summary, including a surfactant-containing solution in at least the first process cycle of the two-step process proved to be an effective strategy to maximize the total nitrogen solubility (N-solubility) yield, in step h). The solubilized proteins in the collected liquid phase (step h) can be precipitated. It is well-known that the use of surfactants tends to prevent protein-protein interactions, hindering protein precipitation. Although the steps of the above-described process, comprise the use of surfactants in the first process cycle, protein precipitation of the collected solubilized proteins is possible. The process may comprise steps of: i) concentrating the solubilized proteinaceous material in the collected liquid phase, and j) recovering the concentrated proteinaceous material Concentrating the solubilized proteinaceous material in step i) may for example be performed by lowering the pH of the liquid phase, inducing salting out through adding a salt such as ammonium sulphate, by thermally treating the liquid phase, and / or by adding flocculants to promote proteinaceous material flocculation, aggregation and / or precipitation. Recovering the concentrated proteinaceous material, step j), may for example be performed through filtering, centrifuging, or drying of the liquid phase to recover the concentrated proteinaceous material in a solid phase or as a pellet. A major positive effect from steps a)-c) of the method is on the precipitation yield (step j). The precipitation yield was greatly improved and had higher influence than the N- solubility yield on the total yield, as compared to conventional alkaline solubilization at e.g. pH 12. The mechanism explaining this observation is yet not fully understood, but we hypothesize it is due to charge neutralization of membranes surfaces upon acidification. Hence, this likely promotes non-covalent interactions of the fragmented membranes, which contain embedded lipid-soluble proteins. Liquid phases, supernatants, from step c) and / or g) may be collected and treated separately or be pooled, before the concentration step i). Concentrating the solubilized proteinaceous material in the liquid phase may comprise isoelectric precipitation in which pH of the liquid phase is lowered. Different proteins have different isoelectric points (pI), defined as the pH at which the net charge of a protein molecule is zero. In isoelectric precipitation, a reduced solubility of proteins at acidic pHs close to or at their pI causes the formation of aggregates that precipitate in solution and can then be easily separated from the liquid phase by centrifugation. Here, the pH may be lowered to less than or about 5. The proteinaceous material precipitated may be formed in a pellet after centrifugation. Through the above-described process, a 3-fold increase in total- and precipitation protein extraction yields can be obtained when compared to traditional extraction methods employing alkaline solubilization followed by acid precipitation. In step e) the protein extraction solution may be water, such as distilled water, water or distilled water adjusted to pH of 6 or higher, such as 6-14 or 8-11 or 9-12, or a surfactant-containing solution as described above and being used in step a). The protein extraction solution may further comprise enzymes, such as proteases. These enzymes may be the same enzymes as were discussed above for the surfactant-containing solution, i.e. proteases or polysaccharidases and combinations thereof. The protein extraction solution may be water or distilled water adjusted to a pH of at least 9. In the second processing cycle, nitrogen solubility yield is significantly higher for the biomass initially treated with surfactant-containing solution compared to alkali-treated ones regardless of the protein extraction solution employed in the second processing cycle. If the protein extraction solution used in the second process cycle is water or distilled water adjusted to pH of at least 9, or a pH of 8-12 or 9-11, the total nitrogen solubility yield may be even more maximized. The surfactant-containing solution may comprise enzymes. Enzymes that can be included in the surfactant-containing solution are listed above. The photosynthetic biomass may be selected from seaweed, duckweed, seagrass, microalgae, cyanobacteria, aquatic leaves, terrestrial leaves and / or grass. Below is a non-limiting list of examples of such photosynthetic biomass: ^ Seaweed: green, red, and brown species ^ Other aquatic plants e.g., duckweed, seagrass ^ Microalgae, e.g., Chlorella, Spirulina ^ Cyanobacteria ^ Aquatic leaves, e.g., watercress ^ Terrestrial leaves, e.g., alfalfa, common nettle, beetroot leaves ^ Grass, e.g., ryegrass The biomass can be used as is without any pre-separation or selection of specific parts of the biomass. The surfactant-containing solution may comprise a non-ionic, an anionic, a cationic, and / or a zwitterionic surfactant. Below is a non-limiting list of examples of such surfactants: Non-ionic surfactants: • Ethoxylated sorbitan polysorbates (e.g., Tween 20) • Ethoxylate sorbitol esters (e.g., Span 60) • Alkylphenol ethoxylates (e.g., Triton X-114) • Alkyl polyglycosides • Ethoxylated Alcohols • Ethoxylated Fatty Acids • Ethoxylated Castor Oil • Fatty Amine Ethoxylates • Propoxylated Alcohols • Alkyl Polypentosides • Alkyl Ethanolamines • Alkyl Glucamides • Alkyl Polyethylene Glycol Ethers • Alkyl Glycosides • Alkyl thioglucosides • Alkylphenol ethoxylates • Glycosides • Sugar esters • Alkyl aryl polyether alcohols Anionic surfactants: • Alkyl sulfates • Alkyl sulfonates • Bile salts and salts • Docusate sodium salt • N-Lauroylsarcosine Cationic surfactants: • Quaternary Ammonium Compounds • Amines and Amine oxides • Imidazolinium Compounds • Ethoxylated Amines Zwitterionic surfactants: • Betaines • Sulfobetaines • Phosphobetaines • Non-detergent Sulfobetaines Preferably food-grade surfactants with a relatively high hydrophilic-hydrophobic balance are used. The surfactant-containing solution may comprise two or more surfactants of different classes of surfactants (i.e. ionic, non-ionic, cationic or zwitterionic surfactants) and / or may comprise two or more surfactants from the same class of surfactants (i.e. two or more non-ionic surfactants). Such combinations may enhance the solubilization of membrane proteins. In one embodiment the surfactant may be a non-ionic surfactant. A concentration of surfactant in the surfactant-containing solution may be in a range of 1000 times lower to 100 times higher than the critical micelle concentration (CMC) of the surfactant. In one embodiment the concentration of surfactant in the surfactant-containing solution may be in a range of 100 times lower to 100 times higher than the critical micelle concentration (CMC) of the surfactant. In one embodiment the concentration of surfactant in the surfactant-containing solution may be in a range of 1 times higher to 100 times higher, or 5 times higher to 75 times higher than the critical micelle concentration (CMC) of the surfactant. The use of these low concentrations of surfactant in the first cycle of the process may contribute to enabling protein precipitation of the collected solubilized proteins, such as through isoelectric precipitation. In step a) of the process, the photosynthetic biomass sample may be mixed with the non-surfactant solution in a ratio of 1:1 to 1:20 (w / v). In one embodiment the ratio of biomass:non-surfactant solution may be 1:2 to 1:6. In step e) of the process, the solid phase may be mixed with the protein extraction solution in a ratio of 1:1 to 1:20 (w / v). In one embodiment the ratio of solid phase:protein extraction solution may be 1:2 to 1:6. In step i) of the process, pH of the liquid phase may be lowered to a pH of 5.5 or less to concentrate the solubilized proteinaceous material. In one embodiment the pH is between 1.5 and 5.5. BRIEF DESCRIPTION OF THE DRAWINGS In Fig.1 is schematically illustrated a process for extracting proteins from photosynthetic biomass. Fig.2 shows the nitrogen solubility yield (%) after processing seaweed U. fenestrata using the process scheme shown in Fig.1. An alkaline solution with pH 12 or a surfactant solution comprising 2% Triton X-114 were used as the protein extraction solution in a first processing cycle. The resulting pellet was mixed with one of the mentioned protein extraction solutions, (N=2), in a second processing cycle. The standard deviations correspond to the yields of these first and second processing cycles. Fig.3 shows nitrogen solubility yield (%) after processing U. fenestrata using the process scheme shown in Fig.1. An alkaline solution with pH 12 or a surfactant solution comprising 0.1-2% Triton X-114 were used as the protein extraction solution in a first processing cycle. The resulting pellet was mixed with an alkaline solution at pH 12, (N=2), in a second processing cycle. The standard deviations correspond to the yields of these first and second processing cycles. Fig.4 shows polypeptide patterns of U. fenestrata biomass, liquid phases / supernatants, and solid phase / pellets analysed through SDS-PAGE under reducing conditions. Twenty micrograms of protein were loaded in each lane. Fig.5 shows nitrogen precipitation yield (%) of liquid phases / supernatants obtained after the first and second processing cycles using the process scheme of Fig.1, as well as the pooled liquid phase / supernatant (N=2). An alkaline solution with pH 12 or a solution containing 0.1 or 0.5% Triton X-114 was used as protein extraction solution to process U. fenestrata in a first processing cycle. The resulting seaweed pellet was mixed with an alkaline solution at pH 12 in a second processing cycle. Fig.6 shows total nitrogen yield (%) of supernatants obtained using the process scheme of Fig.1 after the first and second processing cycles, as well as for the pooled supernatant (N=2). Proteins were solubilized by processing U. fenestrata with an alkaline solution at pH 12 (A) or with surfactant solutions containing 0.1 (F) or 0.5% (G) Triton X-114 and reprocessing the resulting seaweed pellet with an alkaline solution at pH 12. Fig.7 shows the relationship between amino acid up-concentration relative to crude U. fenestrata (N=3) and total amino acid yield (N=2) as well as a comparison to literature values. Note that pH-shift refers to alkaline protein solubilization, followed by isoelectric precipitation at low pH values. Fig.8a and Fig.8b show confocal laser scanning micrographs of freeze-dried protein extracts from U. fenestrata treated with alkaline solution (pH12) in the first cycle and alkaline solution (pH 12) in the second cycle (Fig.8a). For Fig.8b the seaweed was treated with 0.1% Triton X-114 in the first cycle and alkaline solution (pH 12) in the second cycle. Before analysis, the freeze-dried protein extracts were resuspended in distillate water adjusted to pH 2. All samples were stained for protein (shown in grey in the pictures) and lipid (shown in white in the pictures) with Texas Red and BODIPYTM, respectively. All scale bars correspond to 50 µm. Fig.9 shows N total yields (%) of preliminary trials conducted on U. fenestrata biomass with different surfactants in the first cycle of the process and using alkaline solution in the second process cycle, compared to using alkaline solution both in the first and second process cycles. The concentration of surfactant was 10 times above the critical micelle concentration (CMC) of each surfactant. DETAILED DESCRIPTION A process for extracting proteins from photosynthetic biomass, such as seaweed, duckweed, seagrass, microalgae, cyanobacteria, aquatic leaves, terrestrial leaves and / or grass, is schematically shown in Fig.1. It comprises steps of: a) mixing a photosynthetic biomass sample (as e.g., dried, fresh, frozen, freeze-thawed) with a protein extraction surfactant-containing solution, which may comprise one or more non-ionic, anionic, cationic, and / or zwitterionic surfactants. The concentration of surfactant in the surfactant-containing solution may be in a range of 1000 times lower to 100 times higher than the critical micelle concentration (CMC) of the surfactant. The photosynthetic biomass sample may be mixed with the protein extraction solution in a ratio of 1:1 to 1:20 (w / v). Proteinaceous material is extracted from the photosynthetic biomass sample in the surfactant-containing solution in step b) and in step c) the proteinaceous material is separated into a liquid phase comprising solubilized proteins and a solid phase comprising non-soluble material. In step d) the solid phase is collected and in step e) mixed with a protein extraction solution. The protein extraction solution used may be the surfactant-containing solution used in step a) or the protein extraction solution may be water or distilled water or water or distilled water adjusted to a pH of 6 or more after mixing with the solid phase. In the examples shown below an alkaline solution with pH 12 is used. Is is, however known from previous works on seaweed protein extraction (Vilg et al.2017; Harrysson 2018), that adjusting the pH of the protein extraction solution to a value equal to or above pH 6 increases protein solubility. It can also be a buffer adjusted to a pH of 6 or more. The solid phase may be mixed with the protein extraction solution in a ratio of 1:1 to 1:20 (w / v). Proteinaceous material is extracted from the solid phase in the protein extraction solution, step f), and separated into a liquid phase comprising solubilized proteins and a solid phase comprising non-soluble material, step g). The liquid phase with solubilized proteins produced in steps c) and g) above are collected in step h). Thereafter, the process may further comprise a step i) of concentrating the solubilized proteinaceous material in the collected liquid phase, and a step j) of recovering the concentrated proteinaceous material In the section below are specific examples of using the above described process, wherein the photosynthetic biomass used was the seaweed U. fenestrata and the protein extraction solution used was distilled water adjusted to a pH of 12 or a surfactant-containing solution comprising a non-ionic surfactant: Triton X-114, Triton X- 45, Tween 20, Tween 40, Tween 60, in concentrations ranging from 0.1 to 2.0% (w / v). The process is suitable for efficient protein extraction of photosynthetic biomasses beyond seaweed. This is because all these biomasses share a similar cell architecture where proteins are highly compartmentalized within the cell, more specifically inside the chloroplast (Tenorio, A. T., Kyriakopoulou, K. E., Suarez-Garcia, E., van den Berg, C., & van der Goot, A. J. (2018). Understanding differences in protein fractionation from conventional crops, and herbaceous and aquatic biomass - Consequences for industrial use. Trends in Food Science and Technology, 71 (November 2017), 235– 245). Moreover, unlike soybean proteins which are mainly storage and salt-soluble proteins (Day, L. (2013). Proteins from land plants – Potential resources for human nutrition and food security. Trends in Food Science & Technology, 32(1), 25–42), proteins from photosynthetic biomasses are highly heterogeneous in terms of charge, solubility, and hydrophobicity. This heterogeneity arises from the unique characteristics of photosynthetic tissue, where proteins are instead classified depending on their function as enzymatic or structural or according to their solubility as aqueous-soluble (e.g. Rubisco) and lipophilic (e.g. thylakoid membrane proteins) (Tenorio et al., 2018). The process will likely work with other non-ionic surfactants as well as with cationic, anionic, and zwitterionic surfactants since the use of surfactants is widely adopted within the field of molecular biology to solubilize lipophilic proteins for further structural characterization (Arachea, B. T., Sun, Z., Potente, N., Malik, R., Isailovic, D., & Viola, R. E. (2012). Detergent selection for enhanced extraction of membrane proteins. Protein Expression and Purification, 86(1), 12–20). The process will likely retain its advantage in terms of total protein extraction yields for a range of 1000 times below to 100 times above the surfactant’s CMC. The experimental data in the section below suggested that concentrations equal or above to 200 times the CMC do not allow protein precipitation to occur, while protein precipitation occurred at concentrations that were 10 and 50 times higher than the CMC. For surfactant concentrations below 1000 times the CMC, it is expected that it is inert, i.e., does not interfere with precipitation (non-ionic ones) or act as buffering agents (anionic ones). The use of polysaccharidases in combination with the protein extraction solution is expected to further increase the solubility yield of the proteinaceous material as demonstrated elsewhere, particularly for seaweed biomasses (Fleurence et al., 1995). Materials and Methods Collection of seaweed biomass Harvesting of sea-cultivated U. fenestrata took place in February 2022 near Tjärnö Marine Laboratory, Sweden. Then the seaweed was promptly transferred to a 500 L tank maintained at a temperature of 14 °C, under a 16:8 h (L:D) light cycle with an irradiance of 100 µmol m-2s-1. To ensure optimal nutrient levels for seaweed protein accumulation, herring-derived food process water was added to the tank to reach a concentration of 25 µM NH4+; this was repeated every third day to avoid nutrient depletion. Following a 15-day post-harvest cultivation, the protein-enriched U. fenestrata was collected and transported to Chalmers University of Technology on the same day. Protein extraction The fresh U. fenestrata was frozen at -80 °C, and then minced in the frozen state using a meat grinder with a 2.0 mm hole plate. Then, the biomass was divided into portions of 100 g, placed in polystyrene bags, and stored again at -80 °C until further use. The protein extraction process, comprising a solubilization step and a precipitation step, was repeated in duplicate for each version tested. All processing steps were performed at 4±1 °C, to avoid phase separation of the tested surfactant and to assure that proteins did not denature. Protein solubilization To solubilize proteins from U. fenestrata, around 80 g of minced biomass was mixed in a 1:4 ratio (seaweed:solution; w / v) with (i) aqueous Triton X-114, with concentrations ranging from 0.1 to 2% (w / v) or (ii) distilled water. For the mixture containing only seaweed and distilled water, the pH was brought to pH 12 with 1 M NaOH. The mixtures were homogenized (LM5, Silverson) for 30 s at 6500 rpm and then increased to 8000 rpm for an additional 60 s. Then, the mixtures were incubated for 1 h with stirring, except for in cases where ultrasound was applied. After incubation, all mixtures were centrifuged at 8500×g for 15 min at 4°C (Sorvall LYNX 6000, Thermo Scientific). The recovered supernatants containing the solubilized proteins were kept in an ice-water bath, whereas the pellets were subjected to a second processing cycle. The reprocessing consisted of mixing the pellet with the abovementioned surfactant solutions or distilled water in a 1:2 ratio (pellet:solution; w / v). Then, the mixtures were homogenized and centrifuged as described earlier. All pellets were freeze-dried and ground to a fine powder with a mortar and pestle. Ultrasound-assisted protein solubilisation Instead of the 1h incubation, one mixture was incubated for 15 min to allow foam dissipation, followed by an ultrasound treatment (UIP1000hdT, Hielscher) for 45 min at an amplitude of 60% (2.0 cm flat tip probe) in a pulsed on-time for 10 s and off-time of 10 s. The ultrasound-treated mixture was then centrifuged as described in the previous section. Isoelectric protein precipitation To precipitate the solubilized protein, the pH of the following samples was adjusted to pH 2.0 with 1M HCl: (i) supernatant from the first processing cycle; (ii) supernatant from the second processing cycle; (iii) mixture of the first and second supernatants, preserving the original ratio (v / v) between them. The pH value of 2 has been reported in several studies as the average isoelectric point of soluble Ulva sp. proteins (Angell, A. R., Paul, N. A., & de Nys, R. (2017). A comparison of protocols for isolating and concentrating protein from the green seaweed Ulva ohnoi. Journal of Applied Phycology, 29(2), 1011–1026.). After acidification, solutions were incubated under stirring for 15 min and were then centrifuged at 8500×g for 15 min at 4°C. The resulting supernatant was discarded, whereas the pellet contained the recovered proteins and here is referred to as protein extract. All obtained protein extracts were freeze-dried and ground to a fine powder with a mortar and pestle. Total nitrogen analysis The freeze-dried U. fenestrata, pellets and protein extracts were analyzed for nitrogen content through combustion elemental analysis (Elementar vario MICRO cube, Elementar Analysensysteme). The nitrogen solubility yield, nitrogen precipitation yield, and total nitrogen yield were calculated according to Equations 1-3. Microstructure analysis Confocal laser scanning microscopy (CLSM) analysis was performed on freeze-dried protein extracts, which were analyzed for protein and lipids. After mixing two drops of suspensions of the freeze-dried protein extracts with 2 µL each of Texas Red (Invitrogen) and BODIPYTM FL C16 (Invitrogen), followed by incubation for 30 min. For protein and lipid analysis, a 488 nm argon laser and a 594 nm HeNe laser were used for excitation; emissions were collected at 500–536 nm (lipids shown in white in the pictures) and 608–639 nm (protein shown in grey in the pictures). Micrographs were captured with a Leica TCS SP5 (Heidelberg, Germany) with a HCX PL APO CS 63x1.20 W objective. The image format was 1024x1024 pixels, eight lines average Other analytical analyses Table 1 - Other analytical analyses conducted and respective references. Analyses Reference Notes Total amino Trigo, J. P., Engström, N., Steinhagen, S., Juul, L., Harrysson, H., Toth, G. See text below on how total acids B., Pavia, H., Scheers, 710 N., & Undeland, I. (2021). In vitro digestibility and amino acid yield was Caco-2 cell bioavailability of sea lettuce (Ulva fenestrata) proteins extracted calculated using pH-shift processing. Food Chemistry, 712356 (October 2020) Ash content Harrysson, H., Hayes, M., Eimer, F., Carlsson, N. G., Toth, G. B., & - and total fatty Undeland, I. (2018). Production of protein extracts from Swedish red, green, acids and brown seaweeds, Porphyra umbilicalis Kützing, Ulva lactuca Linnaeus, and Saccharina latissima (Linnaeus) J. V. Lamouroux using three different methods. Journal of Applied Phycology, 30(6), 3565–3580. Triton X-114 Garewal, H. S. (1973). A procedure for the estimation of microgram - content quantities of Triton X-100. Analytical Biochemistry, 54(2), 319–324. Polypeptide Abdollahi et. al (2019) All samples, including the pattern surfactant-free ones, were treated with ice-cold acetone to remove the Triton X-114 Ionic strength - Sodium chloride was used as a standard and measurements were done at 20±1°C. The total amino acid yield was calculated following Equation 4. Statistical analysis For N solubility, N precipitation, and total N yields, the t-test was employed. For the compositional data, the one-way ANOVA was used together with the Tukey’s post hoc test for pairwise comparisons. If data was not normally distributed, then the non- parametric Kruskal-Wallis test was used. All tests were conducted using SPSS Statistics software and differences were considered significant at p<0.05. Results and discussion Protein solubilization Effect of sequential extraction with surfactant and alkaline aqueous solutions Fig.2 shows the nitrogen solubility yield (%) after processing U. fenestrata with 2% Triton X-114 and reprocessing the resulting pellet with 2% Triton X-114 or distilled water at pH 12 i.e. alkaline solution. The highest total N solubility yield was recorded for processing U. fenestrata with surfactant, followed by reprocessing the pellet with alkaline solution, pH12 (treatment D). This value was found to be significantly greater (p=0.017) than the control treatment (treatment A), which only utilized alkaline solutions, pH 12, i.e. alkaline solution in both process cycles. The improved total N solubility yield in treatment D can be attributed to the slightly higher solubility yields of the first and second processing cycles, although each of them was not statistically different (p values of 0.064 and 0.052 for each cycle, respectively). By shifting the order of employing the surfactant and alkaline solutions, i.e. alkaline treatment, pH 12, followed by treatment with 2% Triton X-114 (treatment C), the total N solubility yield was similar to the control. Furthermore, performing both processing cycles with surfactant solutions, i.e.25 Triton X-114, (treatment B), resulted in a statistically higher total N solubility yield (p=0.045) compared to the control treatment (treatment A). Treatment E, which involved a single processing cycle with surfactant solution, 25 Triton X-114, and application of ultrasound during the incubation step was hypothesized to increase N solubility, however, the results revealed no statistically significant advantage of this treatment over the first processing cycle only using surfactant solution (treatments B and D). Using surfactant solution in the first processing cycle (treatment B) resulted in more N being solubilized than when using traditional alkaline solubilization at pH 12 (treatment A) (p=0.024). This finding may be related to the amphipathic nature of Triton X-114, which is able to solubilize membrane proteins by disrupting lipid bilayers such as those found in the envelope membranes of chloroplasts as well as in the thylakoid membranes (Buchanan, B., Gruissem, W., & Jones, R. (2015). Biochemistry and molecular biology of plants). Other factors contributing to the enhanced solubilization may be attributed to the improved cell wall permeability since surfactants are known to partially dissolve hemicellulose – a structural polysaccharide in Ulva sp. (Holdt, S. L., & Kraan, S. (2011). Bioactive compounds in seaweed: Functional food applications and legislation. Journal of Applied Phycology, 23(3), 543–597; Muñoz, S. S., Rocha Balbino, T., Mier Alba, E., Gonçalves Barbosa, F., Tonet de Pier, F., Lazuroz Moura de Almeida, A., Helena Balan Zilla, A., Antonio Fernandes Antunes, F., Terán Hilares, R., Balagurusamy, N., César dos Santos, J., & Silvério da Silva, S. (2022). Surfactants in biorefineries: Role, challenges & perspectives. Bioresource Technology, 345(October 2021)). Subsequent analysis of the total ash content of U. fenestrata and the seaweed pellets supports this hypothesis.75.3±2.3% of the total ash in the original biomass was released into the supernatant after treatment with surfactant (2% Triton X-114) (cycle 1) and subsequently alkaline treatment (cycle 2) (treatment D in Fig.2), compared to only 39.3±0.1% after alkaline treatment (cycle 1) followed by alkaline treatment (cycle 2) (p<0.001) (treatment A in Fig.2). Hence, seaweed cell walls treated with Triton X- 114 were probably more permeable to the extracting solvents, enabling higher N solubility yields. In the second processing cycle, the N solubility yield was significantly higher (p=0.045) for the biomasses initially treated with surfactant (treatment B and D) compared to the alkali-treated ones (treatment A and C), regardless of the solution employed in the second processing cycle. This was likely because chloroplast membranes were disrupted (light microscopic pictures not shown) leaving integral proteins more accessible for extraction and seaweed cell walls after surfactant treatment were more permeable to the extracting solution of the second cycle. In summary, including a surfactant containing solution such as Triton X-114 in at least the first process cycle of a two-step protein solubilization process proved to be an effective strategy to maximize the total N solubility yield. This effect was mainly seen when the surfactant was used in the first solubilization step, followed by an alkaline solubilization in the subsequent step. The improved total N-solubility yield is likely due to the surfactant´s ability to disrupt lipid bilayers and to induce structural cell wall changes. Effect of surfactant concentration Fig.3 depicts the effect of different surfactant concentrations on the N solubility yield. Surfactant concentrations of 0.1%, 0.5% and 2% (w / v) Triton X-114 were tested following the sequence of treatment D, as it achieved the highest total N solubility yield (as discussed above). The underlying motivations of this trial were to: (i) minimize the presence of surfactant in the final protein extracts; (ii) decrease costs associated with surfactant usage; and (iii) enable protein precipitation. Overall, reducing surfactant levels from 2% (treatment D) to 0.1% (treatment F) and 0.5% (treatment G) from 200- to 10- and 50 times higher than Triton’s X-114 CMC, respectively, retained the total N solubility yields, which were significantly higher (p-values of 0.031 and 0.025, respectively) compared to the control treatment with alkali in both steps (treatment A). Therefore, the surfactant usage can be significantly reduced without compromising the total N solubility yield or individual yields from the first and second cycles. Polypeptide pattern in supernatants and pellets Fig.4 illustrates the polypeptide patterns of U. fenestrata, as well as the supernatants and pellets obtained from treatments A and D, under reducing and non-reducing conditions. In the first processing cycle, surfactant extraction (treatment D) yielded bands appearing at the top of the stacked gel and at >250 kDa, which were absent in the control treatment (treatment A) and the U. fenestrata biomass. Both extractions (treatment A and D) exhibited common bands at around 10, 50, and 105 kDa. A potential match for the band at ~50 kDa, as suggested by proteome analysis of the brown seaweed Saccharina japonica, is the ATP synthase subunit beta, which is located in the thylakoid membranes (Buchanan et al., 2015; Kim, E., Kim, D., & Kim, Y. (2011). An improved method of protein isolation and proteome analysis with Saccharina japonica (Laminariales) incubated under different pH conditions.123–130). The band at ~105 kDa displayed a higher intensity following surfactant extraction (treatment D) compared to after alkaline extraction (treatment A). Regarding this band, an inverse relation was observed for the pellets, with the pellet obtained from the surfactant extraction showing a less intense band at ~105 kDa, which indicates its extraction (Fig.4). Moreover, under non-reducing conditions, the only high molecular weight band observed was around 129 kDa (data not shown), thus suggesting that the band ~105 kDa is likely a protein subunit. Studies on the plant Nicotiana tabacum have associated a molecular weight of ~105 kDa to the ATP-binding subunit of the ATP- dependent caseinolytic protease, which is an integral protein found in the chloroplast (Wu, W., & Yan, Y. (2018). Chloroplast proteome analysis of Nicotiana tabacum overexpressing TERF1 under drought stress condition. Botanical Studies, 12). In the second processing cycle, the band at >250 kDa was not observed after using alkaline solution at pH 12 (treatment D, Fig.4), whereas the band at the top of the stacked gel had a lower intensity compared to the respective first cycle. Regarding the comparison between supernatants of treatment D and treatment A obtained after reprocessing, no major visible differences were found, although treatment D resulted in a more intense band at 10 kDa. The first processing cycle (treatment D) resulted in a supernatant displaying a dark green colour (not shown). In contrast to the nearly colorless supernatant (not shown) obtained after extraction at pH 12 (treatment A). Upon chloroplast disruption, it is plausible that aqueous-soluble or alkali-soluble proteins located in the stroma could have been solubilized with Triton X-114. After the second cycle, both supernatants (treatments A and D) showed green coloration although the intensity was higher after treatment D. The presence of green coloration indicates co-extraction of chlorophyll, which has previously been utilized as an indicator for the concentration of thylakoid membranes (Emek, S. C., Szilagyi, A., Åkerlund, H. E., Albertsson, P. Å., Köhnke, R., Holm, A., & Erlanson-Albertsson, C. (2010). A large-scale method for preparation of plant thylakoids for use in body weight regulation. Preparative Biochemistry and Biotechnology, 40(1), 13–27). This is due to the association of chlorophyll with light harvesting complexes (LHCs), such as the trimer LHCII. This particular complex represents about half of the total protein in the thylakoid membranes and consists of ~25 kDa monomers (Buchanan et al., 2015). While the presence of LHCII or its monomers were not detected in the SDS-PAGE (Fig.4), it is reported that the trimer has a strong tendency to aggregate outside its in vivo conditions (Schaller, S., Latowski, D., Dawood, A., Wilhelm, C., Strza, K., & Goss, R. (2011). Regulation of LHCII aggregation by different thylakoid membrane lipids. Biochimica et Biophysica Acta (BBA) - Bioenergetics, 1807(3), 326–335). Moreover, membrane lipids can remain attached to LHCII after Triton-based extraction (Simidjiev, I., & Barzda, V. (1997). Isolation of Lamellar Aggregates of the Light-Harvesting Chlorophyll a / b Protein Complex of Photosystem II with Long-Range Chiral Order and Structural Flexibility. Analytical Biochemistry, 175(250), 169–175).). Therefore, the bands at the top of the stacked gel and at >250 kDa in Fig.4 likely correspond to aggregated LHCs associated with membrane lipids. Overall, surfactant- and alkali-based extractions yielded different polypeptide patterns of proteins extracted from U. fenestrata. More specifically, Triton X-114 based extraction seems more selective towards integral proteins although this would need to be further confirmed. Isoelectric protein precipitation Protein precipitation was carried out by acidifying the obtained supernatants to pH 2. Using a concentration of 2% Triton X-114 (treatment D), the proteins remained soluble at pH 2, while protein precipitation occurred at 0.1% and 0.5% Triton X-114 (treatments F and G). While non-ionic surfactants bind weakly to proteins through hydrophobic interactions (Dickinson, E. (1993). Proteins in solutions and at interfaces. In E. D. Goddard & K. P.621 Ananthapadmanabhan (Eds.), Interactions of surfactants with polymers and proteins. CRC 622 Press)), at high concentrations they can shield the hydrophobic regions of proteins. Therefore, it is likely that this shielding effect may prevent protein-protein hydrophobic interactions, which are known to be the primary forces responsible for protein aggregation and precipitation. Fig.5 depicts the N yield after isoelectric protein precipitation of the supernatants obtained from the first and second processing cycles, as well as the pooled mixture of these two supernatants. The reason to test the mixture was that in an industrial setting, conducting isoelectric protein precipitation using a single pooled supernatant would be more cost-effective and time-saving compared to performing two parallel acidifications. Furthermore, we hypothesized that the isoelectric point (IS) and Triton-concentration of the first supernatant would be further diluted into the second supernatant, leading to higher N precipitation yield and lower amounts of Triton X-114 in the final protein extract, respectively. Overall, N precipitation yields of the pooled supernatants containing Triton X-114 (treatments F and G) were more than 3-times higher (p<0.05) compared to the pooled supernatants from the control treatment (treatment A), with no statistical differences observed between 0.1% and 0.5% Triton X-114 (treatments F and G). Moreover, pooling supernatants from treatments F and G led to slightly higher precipitation yields compared to when using only the supernatant obtained from the first cycle; albeit the differences were not statistically significant. When comparing the precipitation of the pooled supernatant with the second one alone, it was similar in treatments A and F (p=0.179 and p=0.885, respectively) and significantly higher in G (p=0.033). Ionic strength after acidification in treatment A was 198, 129, and 172 mM NaCl eq. for the first, second, and pooled supernatants, respectively. The corresponding data for treatment F were 166, 123, and 156 mM, respectively, while for treatment G they were 173, 146, and 172 mM, respectively. The contribution of the second supernatant to the volume of pooled mixture was 41±1% for the control treatment (treatment A) and between 22-25% for treatments containing Triton X-114 (treatments F and G). Based on all these data, it is plausible that low IS values contributed to the two highest N precipitation yields in the control treatment, particularly in case of the second supernatant alone. However, in presence of Triton X-114 (treatments F and G), no significant correlation was seen between N-precipitation yield and IS (Pearson correlation, R=0.16, p>0.05, N=6), thus the hypothesis formulated between IS and precipitation yield is rejected. Total N yield and relationship with protein purity of the protein extracts Fig.6 depicts the total N yield of treatments A, F, and G. Extraction with 0.1 or 0.5% Triton X-114, followed by reprocessing the pellet at pH 12 (treatments F and G) resulted in significantly higher total N-yields (p<0.001) when compared to the control treatment (treatment A). The statistical significance remained consistent, whether protein precipitation was carried out on the individual supernatant samples or on the pooled supernatant. Fig.7 presents the total protein yield based on the total amino acid content plotted against the amino acid up-concentration relative to the original biomass. The amino acid analysis was carried out in the protein extracts obtained from the pooled supernatants given the higher scalability potential of this strategy. The total amino acid yield of a treatment with 0.1% Triton-X in the first process cycle and alkaline treatment in the second cycle (treatment F) was 22.6±1.5%, which represented a 3.4-fold increase when compared to using an alkaline treatment in both process cycles (treatment A) (p=0.01). Furthermore, similarly to the total N yield, no statistical differences in total amino acid yield were found between 0.1 and 0.5% Triton X-114 (treatments F and G), Fig.6. In our earlier works that extracted proteins from U. fenestrata, we reported a total amino acid yield ranging from 5 to 11% (Harrysson et al., 2018; Trigo et al., 2021), thus placing our new extraction method around 2-fold above the maximum range of this interval. Other works on protein extraction using the same Ulva species reported total protein yields based on the Lowry analysis method, thus hindering direct comparisons with our present data (Harrysson, H., Konasani, V. R., Toth, G. B., Pavia, H., Albers, E., & Undeland, I. (2019). Strategies for Improving the Protein Yield in pH-Shift Processing of Ulva lactuca Linnaeus: Effects of Ulvan Lyases, pH-Exposure Time, and Temperature. ACS Sustainable Chemistry & Engineering, 7(15), 12688–12691.; Juul, L., Danielsen, M., Nebel, C., Steinhagen, S., Bruhn, A., Jensen, S. K., & Undeland, I. (2021). Ulva fenestrata protein – Comparison of three extraction methods with respect to protein yield and protein quality. Algal Research, 60(August)). It is however important to note that those studies maximized protein precipitation through a freeze- thawing cycle as part of the pH-shift method (Harrysson et al., 2018, 2019; Juul et al., 2021; Trigo et al., 2021) or ammonium-sulphate induced precipitation as part of a traditional protein extraction using sonication in water (Harrysson et al., 2018). Although relevant for academic purposes, both strategies are likely not scalable due to the elevated time and energy consumption as well as the additional steps required to remove ammonium sulphate. Above is described the use of Triton X-114 as the surfactant used in the first process cycle. Preliminary data with surfactants from using the Tween family (Tween 20, Tween 40, Tween 60) and Triton X-45 in the first process cycle and alkaline solution, pH 12, in the second cycle (Tween20-Alk, Tween40-Alk, Tween60-Alk, TritonX45-Alk) revealed that at a concentration 10 times higher than their CMC, total N yield was 1.4 to 2.6 times higher than when using alkaline treatment, pH 12, in the first and second process cycles (treatment A, Alk+Alk), see Fig.9. Furthermore, different types of non- ionic surfactants have been reported to be selective towards different types of membrane proteins (Arachea et al., 2012), so if combined could potentially improve extraction yields. Surfactants are also characterized by their hydrophilic-lipophilic balance (HLB) value which can dictate their extraction efficiency as earlier reported for e.g., mitochondrial protein and phospholipids (Egan, R. W. (1976). Hydrophile-lipophile balance and critical micelle concentration as key factors influencing surfactant disruption of mitochondrial membranes. Journal of Biological Chemistry, 251(14), 4442–4447). In the latter study, optimum extraction was found between HLB values of 12.5 and 13.5; Triton X-114 has a HLB value of 12.4. Microstructure of protein extracts Confocal laser scanning micrographs were captured for the protein extracts resuspended at pH 2 from protein extracts from seaweed treated with alkaline solution (pH 12) in the first cycle and treated with alkaline solution (pH 12) in the second cycle (treatment A), Fig.8a. Seaweed treated with 0.1% Triton X-114 in the first cycle and alkaline solution (pH 12) in the second cycle (treatment F), resuspended at pH 2, is shown in Fig.8b. Samples were unstained for chlorophyll autofluorescence and stained for protein (shown in grey in the pictures) and lipid (shown in white in the pictures) with Texas Red and BODIPYTM, respectively. The objective was to understand potential reasons for the higher total yields observed with 0.1% Triton-X 114 and alkaline treatment compared to alkaline treatment followed by alkaline treatment. Texas Red staining revealed that the protein aggregates from treatment with treatment F were primarily composed of protein (Fig.8b). It also showed more compact aggregates than from treatment with treatment A (Fig.8a). Staining with BODIPYTMindicated the presence of lipids inside the aggregates from treatment with treatment F, Fig.8b, but not in the case from treatment with treatment A (Fig.8a). It is reported that membrane lipids in leaves remain bound to chlorophyll with light harvesting complexes (LHCs) even after Triton-based solubilization (Simidjiev & Barzda, 1997). Moreover, membranes in U. fenestrata, particularly those found in the thylakoids, contain lipids with sulfonic groups (Khotimchenko, S. V., & Yakovleva, I. M. (2004). Effect of solar irradiance on lipids of the green alga Ulva fenestrata Postels et Ruprecht. Botanica Marina, 47(5)). Due to the very low pKa of these sulfated groups (pH≤2), it is probable that acidification to pH 2 triggered charge neutralization of the membrane surfaces, followed by solubility loss, promotion of hydrophobic interactions between the membrane lipids and their subsequent precipitation together with the bounded membrane proteins. Relationship between the fatty acid profile of the protein extracts and co-extraction of membrane lipids Glycolipids constitute one of the main lipid classes in seaweed and terrestrial plants (Buchanan et al., 2015; Khotimchenko & Yakovleva, 2004). These lipids are the building blocks for photosynthetic membranes and are characterized by high levels of n-3 fatty acids, specifically 16:4n3, 18:3n3, and 18:4n3 in the case of U. fenestrata (Khotimchenko & Yakovleva, 2004). Thus, the content of n-3 fatty acids and the mentioned individual fatty acids may serve as a marker for co-extracted membrane lipids bound to the LHCs. According to Table 1, the protein extract produced in 0.1Trit+Alk (0.1 % Triton X-114 in the first process cycle and alkaline solution in the second cycle) had 22% more n-3 fatty acids than the control Alk+Alk (alkaline solution in first and second cycle) (p<0.05). This difference was mainly attributed to a higher content in 18:3n3 and 18:4n3 (Table 2). As additional evidence of co-extraction of membrane lipids, the total n-3 fatty acids to protein ratio was similar between the extract produced with 0.1Trit+Alk and the biomass (~0.020), while in the Alk+Alk protein extract it was lower (0.013). A similar trend was observed for the ratios of 18:3n3 and 18:4n3, while the one for 16:4n3 remained consistent across all protein extracts and biomass. Composition of protein extracts Table 2 presents the chemical composition of the protein extracts in terms of amino acids (total amino acids - TAA, total essential amino acids - TEAA), total fatty acids, total ash content, and Triton X-114 content. These parameters contribute to evaluating the nutritional value of the protein extracts, while also disclosing potential benefits and limitations in comparison to protein extracts obtained through other extraction methods. Total amino acid content differed significantly (p<0.05) between the crude U. fenestrata and protein extracts obtained from the pooled supernatants of treatments A (Alk+Alk), treatment F (0.1Trit+Alk), and treatment G (0.5Trit+Alk). Moreover, total amino acid content ranked the protein extracts in the following order: treatment G (0.5Trit+Alk) > treatment F (0.1Trit+Alk) > treatment A (Alk+Alk), which indicates that selecting lower surfactant concentration levels led to higher protein purities. Only the control treatment (Alk+Alk) and treatment F (0.1Trit+Alk) were within or above the range of 51.2-58.4% TAA per dw reported for protein extracts from U. fenestrata (Harrysson et al., 2018; Trigo et al., 2021) Regarding TEAA, it increased from biomass to protein extracts regardless of how proteins were extracted, which is consistent to previously described protein extraction methods (Harrysson et al., 2018; Trigo et al., 2021). Compared to the crude biomass, total fatty acids were significantly up-concentrated 1.7- and 2.2-fold in Alk+Alk and 0.1Trit+Alk (Table 2), respectively, with the difference between the two treatments being statistically significant (p<0.05). The result between treatments aligns with the increased presence of lipids in the micrographs of the protein extract from 0.1Trit+Alk (Fig.8b). Protein extraction resulted in a decrease in total ash content, with the treatments using Triton X-114 (treatment F and G (0.1Trit+Alk and 0.5Trit+Alk)) exhibiting the lowest values. This reduction in ash levels is an important factor associated with improved N digestibility in protein extracts derived from Ulva sp., as ascribed by Juul et al. (2022). As for surfactant levels, our results indicate that lower Triton X-114 concentrations during extraction are more advantageous due to its decreased presence in the protein extract. In extracts of material treated with treatment F (0.1Trit+Alk) and treatment G (0.5Trit+Alk), 32.4±2.8% and 45.5±8.2% of the initial Triton X-114 amount, respectively, was present in the final product, most likely due to high levels of soluble phase also in the precipitated pellet. Moreover, we rejected the hypothesis that Triton- concentration of the first supernatant would be further diluted into the second supernatant, leading to lower amounts of Triton X-114 in the final protein extract. This is because in treatment F (0.1Trit+Alk), protein extracts from individual precipitation of the first and second supernatants had a Triton X-114 content of 15.4±2.9% and 15.3±2.6%, which were statistically similar (p>0.05) to the protein extract produced when pooling the supernatants (10.0±0.4%) (Table 2). Similar observations were found for treatment G (0.5Trit+Alk). Overall, further optimization studies should focus on finding a balance between maximizing total protein yield and protein purity, while minimizing surfactant presence in the final product. Conclusion The above-described process is a new, more efficient and potentially scalable extraction method that targets aqueous soluble and aqueous-insoluble membrane proteins from photosynthetic biomass, here exemplified for U. fenestrata. The process consists of a two-step protein solubilisation: aqueous surfactant-containing solution, exemplified by e.g. Triton X-114, in a first process cycle, and a protein extraction solution, exemplified by an alkaline aqueous solution, in a second process cycle. After each step centrifugation may be performed and recovery of the solubilized proteins via e.g. isoelectric precipitation and centrifugation. Among all tested combinations to maximize N solubility yield, the combination comprising a first processing cycle with 2% Triton X-114 followed by reprocessing the pellet with water at pH 12 achieved the best results for U. fenestrata. This was likely because in the first processing cycle, the surfactant allowed a broader range of polypeptides to dissolve compared to with alkali alone, likely membrane-bound. Extraction of chlorophyll-binding proteins was likely confirmed by the presence of a band >250 kDa and at the top of the stacked gel during SDS-PAGE, both likely aggregates. Other reason for improved N solubility yield may be related to the capacity of Triton X-114 to promote permeability of cell walls and thereby easing access of the extracting solvent. In our case, this phenomenon was supported by the significantly higher ash removal when compared to alkaline-based extraction. It was also found that Triton X-114 levels can be reduced at least 20 times from 2% without compromising N solubility yields. This reduction was also necessary to enable protein isoelectric precipitation. Once enabled, N precipitation yields were up to 3.2-3.6 times higher when testing 0.1 and 0.5% Triton X-114, compared to the control comprising only alkaline extractions. Moreover, pooling the supernatants before protein precipitation resulted in similar N precipitation yields compared to when not pooling them, making the process more scalable. Total protein yields based on N and TAA analyses were up to 4.4- and 3.4-times higher when using Triton X-114, respectively, as compared to the control treatment. Among all tested treatments and comparable data from literature, the extraction process comprising 0.1% Triton X-114 was the closest to an ideal scenario delivering high TAA yield and high TAA up-concentrations relative to the original biomass. Nonetheless, further optimization studies shall focus on finding a compromise between maximizing total protein yield and protein purity, while minimizing surfactant consumption. This is particularly important since surfactant levels were directly related to the purity of the protein extracts. In summary, this invention encompasses a potentially scalable extraction method that can contribute to the rising global need for more sustainable food protein. Future studies shall continue prioritizing viable food-grade alternatives to Triton X-114.

[0002] t,)3≥ N(s dic ao ni )3 m= afN(ots n mretn eto nci4 st1 c1- artXn x eotini retTd orna p)d 3 e = cuN(dtorn petfn o o ncoih t c i a slopato mto,)C:3= 2 N el(s badiT c a

Claims

CLAIMS 1. A process for extracting proteins from photosynthetic biomass, the process comprising steps of: a) mixing a photosynthetic biomass sample with an aqueous surfactant-containing solution, b) extracting proteinaceous material from the photosynthetic biomass sample in the surfactant-containing solution, c) separating the proteinaceous material into a liquid phase comprising solubilized proteins and a solid phase comprising non-soluble material, d) collecting the solid phase, e) mixing the solid phase with a protein extraction solution, f) extracting proteinaceous material from the solid phase in the protein extraction solution, g) separating the proteinaceous material into a liquid phase comprising solubilized proteins and a solid phase comprising non-soluble material, h) collecting the liquid phases with solubilized proteins produced in steps c) and g).

2. The process of claim 1, further comprising steps of: i) concentrating the solubilized proteinaceous material in the collected liquid phase, and j) recovering the concentrated proteinaceous material 3. The process of claim 1 or 2, wherein the protein extraction solution is selected from a surfactant-containing solution or water, optionally with added enzymes and / or antioxidants, wherein pH of the protein extraction solution is at least 6.

4. The process of any of claims 1-3, wherein the surfactant-containing solution comprises enzymes.

5. The process of any of claims 1-4, wherein the photosynthetic biomass is selected from seaweed, duckweed, seagrass, microalgae, cyanobacteria, aquatic leaves, terrestrial leaves and / or grass.

6. The process of any of claims 1-5, wherein the surfactant-containing solution comprises a non-ionic, an anionic, a cationic, and / or a zwitterionic surfactant.

7. The process of claim 6, wherein the surfactant is a non-ionic surfactant.

8. The process of any of the preceding claims, wherein a concentration of surfactant in the surfactant-containing solution is in a range of 1000 times lower to 100 times higher than the critical micelle concentration (CMC) of the surfactant.

9. The process of any of claims 1-8, wherein in step a) the photosynthetic biomass sample is mixed with the non-surfactant solution in a ratio of 1:1 to 1:20 (w / v).

10. The process of any of claims 1-9, wherein in step e) the solid phase is mixed with the protein extraction solution in a ratio of 1:1 to 1:20 (w / v).

11. The process of any of claims 2-10, wherein in step i) pH of the liquid phase is lowered to a pH of 5.5 or less to concentrate the solubilized proteinaceous material.

Images