Patent Application
20220192221
The invention relates to a process to extract and/or isolate albumin proteins from sunflower seeds. The invention further relates to the products thus obtained.
The global demand for proteins is expected to rise at a fast pace in the coming years due to the world population growth. Plant-based proteins can help to satisfy this demand.
Oil seeds, such as sunflower seeds, are an important source of proteins having a good nutritional value. In particular, proteins contained in sunflower seeds are now widely used in the food industry as, for example, food additives or stabilisers, or as major nutritious components.
Sunflower (Helianthus annuus L.) is the fourth most important oilseed with production reaching the value over 47 million tons in 2016 (FAO Eurostat). The solid residue (meal) remaining after oil extraction process is a valuable source of proteins (30-50% on dry matter basis).
Sunflower meals are generally prepared from a sunflower seed in which the sunflower seed undergoes the following steps: cleaning, drying, dehulling (shelling), crushing, flaking, cooking (optional) and mechanical pressing—usually through screw-presses (expellers)—to form, what is known in the art as, a «press(ed) cake» or «press(ed) meal», containing 15-20% of oil. The press cake can then be extracted with a non-polar (hydrophobic) solvent—usually hexane—to remove or reduce residual oil from the sunflower press cake to form, what is known in the art as, a «defatted meal» or «solvent extracted meal». When such an organic solvent is used the oil (or lipid) content remaining in the defatted meal is residual (e.g. ranging from 0.1 to 4 wt % by weight of the total defatted meal) (see for review Laisney et al., 1996). Dehulling sunflower seeds has shown to be particularly effective to obtain a cake or meal with a higher level of proteins than non-dehulling seeds.
Sunflower proteins are extracted and purified as concentrates or isolates depending upon their degree of purity. Isolates must further meet a number of varied demands from the food industry in terms of solubility, exclusion of components seen as undesirable, such as phytic acid and chlorogenic acid, and organoleptic properties and, in particular, colour. These characteristics are linked, at least partially, to the processes used for their extraction. The major classes of proteins present in sunflower seeds are globular helianthinins (50-80%) and sunflower albumins (SFAs) (25-35%) (Gonzalez-Perez et al., 2007; Wildermouth et al., 2016; Kortt et al., 1990; Mazhar et al., 1998; Raymond et al., 1995).
Helianthinins have an oligomeric structure of 300-350 kDa. The predominant hexameric structure (11S) consists of six subunits composed of a basic polypeptide and β acidic polypeptide with a molecular weight of 21-27 kDa and 32-44 kDa, respectively. The isoelectric point (IP) of helianthinins is about pH 4-6 (Durante et al., 1989; Gonzalez-Perez et al., 2002). Sastry et al. “Binding of Chlorogenic Acid by the Isolated Polyphenol—free 11S protein of Sunflower (Helianthus annuus)” Seed JOURNAL OF AGRICULTURAL AND FOOD CHEMISTRY, 1 Dec. 1990″ discloses a method to minimise the binding of polyphenols to the 11S sunflower protein (helianthinin) in order to obtain a polyphenol-free 11S proteins. It does not relate to sunflower albumins.
Sunflower albumins (2S) are a polymorphic group of proteins constituted of a single basic polypeptide chain (10-18 kDa). Several individual forms of SFAs (8-13) have been identified including 2S albumin storage proteins (16 kDa) and methionine-rich 2S proteins (12 kDa). SFAs are basic proteins with an average isoelectric point (IP) of pH 8.8 (Gonzalez-Perez et al., 2007).
Many reports describe sunflower protein isolate production processes. As for many other plant source proteins, the highest protein extraction yields are observed under alkaline conditions. According to preliminary study on sunflower proteins undertaken by Sosulski et al. (1969) their extractability at basic pH (in 0.2% NaOH w/w) could reach 90%. Later works of Ordonez et al. (2001) and Gonzalez-Perez et al. (2002), demonstrate maximal protein extraction yields ranging 50-70% at pH close to 10. In these conditions, both SFA and helianthinins are extracted. Nonetheless, at high pH values phenolic compounds, notably chlorogenic acids (1-4% of the meal on dry matter basis) are largely co-solubilized. The basic pH is in favour of chlorogenic acid oxidation leading to phenol-protein covalent bonding (Wildermuth et al., 2016; Ozdal et al., 2013; Bongartz et al., 2016). This yields protein isolates with a characteristic greenish colour that are unsuitable for food applications.
Different process approaches have been attempted in order to limit the protein-phenol interactions, while keeping a satisfying protein extraction yield. The first one was to use high ionic strength extraction phase to increase the protein recovery at pH around neutrality. Optimization of solid/liquid extraction process performed by Pickardt et al. (2009) and Ivanova et al. (2012) showed that pH around 6 and 1.6-2.1 mol·L−1 NaCl allowed a satisfying extraction yield (50% vs 20% without salt) and resulted in light-coloured protein product. In PICKARDT ET AL: “Isoelectric protein precipitation from mild-acidic extracts of de-oiled sunflower (L.) press cake”, EUROPEAN FOOD RESEARCH AND TECHNOLOGY, 25 May 2011; optimum conditions to carry out precipitation of sunflower seed protein are proposed. The extraction is carried out at mild pH around 6 with a concentration of NaCl ranging from 1.6 to 2.1 mol·L−1. However, this huge amount of salt seems to be unreasonable for the food industry due to the cost of effluent treatments and the corrosive risk toward stainless steel. Some authors have tried to inhibit formation of green pigment by adding of reducing agent as Na2SO3, NaHSO3 or ascorbic acid (Salgado et al., 2012; Friedman et al., 1996; Richard-Forget et al., 1997; Kuijpers et al., 2012). Nevertheless, this method leads to further greening upon alkaline pH or storage of protein products because of the presence of residual polyphenols (Wildermouth et al., 2016; Yoruk et al., 2003). Another approach was to carry out a polyphenol extraction with organic or organic-aqueous mixtures such as butanol (Saeed et al., 1988), 60% (v/v) methanol (Malik et al., 2017), 80% (v/v) methanol (Gonzalez-Perez et al., 2002) or 95% (v/v) ethanol (Cater et al., 1972). Although, this treatment following to alkaline extraction and purification conducts to obtain low-coloured protein isolates, important lost in low-molecular proteins has been previously reported (Rahma et al., 1979). Furthermore, the use of organic solvents for food processing is more often questioned.
Consequently, the conventional strategy that consists in trying to maximise the overall protein extraction yield and minimize the green colour of protein isolates comes to a dead end. Moreover, it yields large amounts of solid residue having poor value because of low nitrogen content. Hence, an alternative process aimed at a more sustainable extraction of sunflower proteins should be considered.
Among the two principal protein fractions, only SFAs present completely-balanced amino acid profile and usually high amount of sulphur-containing amino acids (Kortt et al., 1990 and 1991; Gonzalez-Perez et al., 2007) in accordance to the requirement pattern of the WHO/FAO/UN. It is also well known, that SFAs exhibit interesting foaming and emulsifying properties (Gueguen et al., 2016; Burnett et al., 2002). The excellent nutritional value and technofunctional capacities make SFAs attractive for various food applications.
Selective extraction of albumins from oleaginous meals has been previously denoted upon acidic conditions (Gonzalez-Perez et al., 2005; Nioi et al., 2012). Prior reports (Nissar et al., 2017; Kumar et al., 2010; Cheryan et al., 1980) agree about the efficient elimination of phytic acid from oleaginous meals under lower pH values.
It is therefore highly desirable to provide a high yield process to obtain a sunflower albumin isolate which process also leads to a value-added residual solid with high protein content.
Alternatively, or additionally it is also highly desirable to provide a sunflower albumin isolate with negligible or at least small amounts of chlorogenic acid and/or phytic acid.
Alternatively, or additionally it is also highly desirable to provide a sunflower albumin isolate having high solubilisation properties in water within a broad pH range and/or improved organoleptic properties.
Within the framework of their research the Inventors have performed a study on the impact of experimental conditions on solid/liquid extraction to find a suitable process for obtaining a sunflower albumin isolate from sunflower seed press cake.
The Inventors have now found solid/liquid extraction conditions which result in selective production of colorless SFA isolates with low phenolic contamination and good SFA extraction yield (≥60%). In particular the Inventors have found that the combination of method steps which include, inter alia, an acidic protein extraction step with the use of small amount of NaCl, and a diafiltration step is useful to achieve such a result. Simultaneously, the optimal extraction condition allowed obtaining a helianthinin-rich solid residue with reduced amount of antinutritional phytate and other non-protein compounds.
Accordingly, the invention provides a process for producing a sunflower protein isolate, said process comprising the following steps:
The sunflower protein isolate obtainable or obtained by the process described therein is rich in sunflower albumins (SFAs); in particular it contains at least 70 wt %; preferably at least 85 wt %, more preferably at least 90 wt %, by weight of the total proteins.
Advantageously, the SFA extraction yield (=SFA content in the protein isolate/SFA content in sunflower seed press cake) according to the process of the present invention is at least 55%, preferably at least 65%, more preferably at least 70%.
Advantageously, said process does not contain a step of precipitation of said protein after step (b) and prior to step (c).
Advantageously, said process does not contain a step of contacting an exogenous phytase enzyme with the solubilised protein solution.
As used herein the term “sunflower seed” refers to oil seed obtained from a plant of the genus Helianthus and particularly from the species Helianthus annuus L. and from any particular sub-species or variety of said species, including wild perennial, hybrids thereof together with mutant and genetically modified varieties.
According to a specific embodiment of step b), NaCl can be replaced by KCl or CaCl2) or a mixture thereof, i.e., a mixture of NaCl and KCl, or a mixture of NaCl and CaCl2), or a mixture of KCl and CaCl2), or a mixture of NaCl, KCl and CaCl2).
Protein content is measured on dry matter by determining the nitrogen content using the Kjeldahl method (see the Examples, infra) and multiplying it by a conversion factor of 5.6 (i.e., Nx5.6 conversion factor) determined for sunflower proteins as described by Defaix et al. (Defaix et al. 2019) and used by other authors; (Pickardt et al., 2009, Ivanova et al., 2012; Gonzalez et al., 2005).
Step (a)
In the context of the invention the term “sunflower seed press cake” relates to sunflower seeds from which sunflower oil is partially extracted by mechanical pressing to form what is known in the art as a press(ed) cake or press(ed) meal, also known as “partially defatted/deoiled cake/meal”. Processes for obtaining sunflower seed press cake are well known in the art (Laisney et al., 1996). The sunflower seed press cake is not extracted by a non-polar solvent (such as hexane, pentane or a mixture thereof) to further remove oil from said sunflower seed press cake.
Advantageously, the sunflower seeds are first dehulled (decorticated), at least partially (e.g., 80 wt % measured as raw material depletion), before being transformed into a sunflower seed press cake. The use of dehulled seeds has shown to be particularly effective to extract a high level of proteins.
Although, any sunflower seed press cake may be used, it is preferred to use a sunflower seed cold press cake. By “cold press” it is particularly meant that the sunflower seeds are not cooked prior to its passing through the press and the temperature of the sunflower seed during the pressing step is of 85° C. or less, preferably 75° C. or less, more preferably 70° C. or less, most preferably 60° C. or less, to form a sunflower seed cold press cake. Usually, proteins contained in a cold press cake are less denatured than those contained in a press cake which has undergone an oil extraction process wherein the temperature of the sunflower seeds has reached 86° C. or more.
The press cake can be grounded into particulates and sieved so that only the fraction of particulates smaller than 500 μm is used. Press cakes made of fractions smaller than 700 μm, or than 800 μm or less, and even smaller than 1 mm may also be considered in order to carry out the process of the invention.
The protein content of the sunflower seed press cake can be ranging from about 15 wt % to about 58 wt % of proteins, preferably from 28 wt % to 45 wt % and more preferably from 34 wt % to 45 wt % on dry matter basis the press cake.
The oil, or lipids, content of the sunflower seed press cake can be ranging from about 12 wt % to about 22 wt % of lipids (e.g., about 15 wt %), on dry matter basis of the press cake.
The phytic acid content of the sunflower seed press cake can be superior to 4 wt % on dry matter basis of the press cake, usually it may range from 5 to 10 wt % on dry matter basis of the press cake.
The process of the present invention also encompasses a sunflower seed press cake which has been processed in order to extract other substances than its oil/lipids. For example, a sunflower seed press cake from which some proteins have already been extracted can be used according to a process of the invention.
Step (b)
The sunflower seed press cake, preferably the sunflower seed cold press cake, is mixed with an aqueous NaCl solution having a NaCl concentration ranging from 0.2 to 0.6 mol·L−1 (M), at a pH ranging from 3 to 4.5, in order to solubilize proteins present in said sunflower seed press cake and to thus obtain a solubilised protein solution. The aqueous NaCl solution can be prepared by adding to an aqueous solution an aqueous solution of NaCl at 0.1-1.0 mol·L−1.
This aqueous solution is a liquid able to extract water-soluble proteins and which is mainly or essentially constituted of water. As used herein the term “water” refers to any type of available water, such as tap water or drinking water. It may include a small proportion (e.g., less than 15 wt %, 10 wt %, 5 wt %, 2 wt % or 1 wt % by weight of the total liquid) of at least one another component. Such another component can be naturally occurring in the water (such as various types of salts, metallic or otherwise such as KCl, CaCl2)) or added on purpose, in particular to adjust the pH and/or the ionic strength of the solution.
The ionic strength of the solubilised protein solution should be controlled and kept at a level ranging from 0.2 to 0.6 mol·L−1, preferably from 0.2 to 0.5 mol·L−1, even more preferably from 0.2 to 0.4 mol·L−1. In a preferred embodiment, the NaCl ionic strength adjustment is carried out by the addition of an aqueous solution of NaCl at 0.1-1.0 mol·L−1.
The pH of the solution which contains the solubilised protein is adjusted to be acidic, that is from 3 to 4.5. In order to adjust the pH of the solubilised protein solution of step b) a component such as an acid is added. For example, this component can be a strong acid, such as hydrochloric acid, or a weak acid, such as citric acid, lactic acid or phosphoric acid. In a preferred embodiment, the pH adjustment is carried out by the addition of an aqueous solution of HCl at 1.0 mol·L−1.
Advantageously, no other salts than NaCl are added and/or only NaCl is added to the water used to solubilise the proteins.
It is preferred not to use a solvent such as methanol, propanol, iso-propanol and/or tetrahydrofuran. In particular it is preferred that no organic solvent be used in the aqueous solvent used in step (b).
According to a more preferred embodiment of the invention said aqueous NaCl solution of step (b) has a NaCl concentration ranging from 0.2 to 0.3 mol·L−1 and said pH is from 4.0 to 4.2, preferably said aqueous NaCl solution of step (b) has a NaCl concentration of 0.25 mol·L−1 and said pH is of from 4.05 to 4.15, most preferably a pH of 4.1.
According to a preferred embodiment of the invention, said aqueous NaCl solution of step (b) has a NaCl concentration ranging from 0.3 to 0.6 mol·L−1 and said pH is from 3.1 to 3.5, preferably said aqueous NaCl solution of step (b) has a NaCl concentration ranging from 0.4 to 0.5 mol·L−1 and said pH is of about 3.2 to 3.4.
The sunflower seed press cake and the aqueous NaCl solution are mixed together using conventional method to form a slurry which contains dissolved proteins in solution, and may further contained a suspension of protein, oil and optionally fibers as well as anti-nutritional and phenolic compounds. The weight solid/liquid ratio (w/w) of the sunflower seed press cake/aqueous NaCl solution usually ranges from 1:5 to 1:20 (wt %), preferably 1:6 to 1:15 (wt %) and more preferably about 1:8 (wt %) or 1:10 (wt %).
The extraction or solubilisation of the proteins is usually carried out by stirring or agitating the slurry formed by the sunflower seed press cake and the aqueous NaCl solution for a time period ranging from 10 min to 120 min, preferably 30 min to 90 min (e.g., around 60 min). The stirring speed can be ranging from 100 rpm to 800 rpm, for example from 150 rpm to 900 rpm, e.g., 600±20%.
The pH adjustment can be done either before and/or during stirring.
The temperature of the slurry is preferably room temperature (i.e., 20° C.) or higher. In particular it may range from 40 to 70° C., preferably from 50 to 60° C. (e.g., around 55° C.).
According to a preferred embodiment, the extraction (i.e. solubilisation of the proteins) step is not carried out using a blanket of inert gases. More preferably no inert gases are used in the process of the invention.
Step (c)
Once step (b) is carried out, the liquid phase comprising the solubilised protein solution and the solid phase contained in the mixture (slurry) are separated. The means to carry out this separation are well known in the art and include centrifugation means, such as a decanter centrifuge, filtration means, pressing means, such a screw press, a filter press, a belt press, a French press, decantation means, and/or any other means that separates the slurry into a solid phase and a liquid phase. This separation may be performed using a centrifuge, for example at g-force ranging from 1 000 to 20 000 g, preferably from 12 000 to 18 000×g, for example about 15 000×g. As the skilled person will directly understand the solid phase contains a small proportion of liquid and conversely the liquid phase will comprise a small proportion of solids or solid particles.
In a particular embodiment of the invention the liquid phase containing residual solids is further subjected to another separation step using for example at least one disk stack centrifuge. The g-force of this centrifugation may be ranging from 6 000 to 20 000×g, for example 17 000×g.
The spent solids, containing helianthinin proteins, can either be disregarded or recovered for further use, optionally after a drying step.
In an optional particular embodiment of the invention, the spent solids obtained from step (c) described above is further subjected to a mixing with an aqueous NaCl solution at a pH ranging from 3 to 4.5, in order to further solubilize proteins present in said solids and to thus obtain a further solubilised protein solution, wherein said aqueous NaCl solution has a NaCl concentration ranging from 0.2 to 0.6 mol·L−1, as described above. The liquid phase containing the solubilised protein solution is then separated from the slurry in suspension as described above. The second solubilised protein solution enriched in protein can be pooled with the first solubilised protein solution enriched in protein. The pooled solution enriched in protein is then subjected to one or several membrane filtration(s).
Step (d)
The recovered solubilised protein solution, enriched in protein, is advantageously subjected to diafiltration steps, and preferably some preliminary purification steps such as filtration, microfiltration, or ultrafiltration, preferably ultrafiltration, to recover a purified protein solution.
Means to carry out membrane filtration are well known in the art.
According to an embodiment, the solubilised protein solution can be subjected to one or more microfiltration step(s). Such a microfiltration may be performed by using filtration membrane having a nominal pore size ranging from 0.1 μm to 2 μm, preferably from 0.1 μm to 1 μm (e.g. 0.1 μm). Microfiltration may, as it is usual, comprise one or more diafiltration step with non-salt water (e.g., tap water) or an aqueous solution of salt, for example NaCl or CaCl2), preferably NaCl at a suitable concentration. Such salt concentration may be ranging from 0.05 mol·L−1 to 1 mol·L−1, preferably from 0.4 mol·L−1 to 0.6 mol·L−1. Furthermore, the pH may advantageously be controlled and/or adjusted, for some or all the filtration and/or the diafiltration steps. In order to adjust the pH a pH modifier, such as an acid or a base, can be added, e.g., phosphoric acid, to the water.
The collected permeate from the microfiltration step may be concentrated prior or after diafiltration step using, for example, an ultrafiltration (UF) membrane system. The level of concentration chosen can be achieved applying a VRF of 1 to 20 of the solubilised protein solution but is advantageously ranging from 2 to 8. Hence the concentration can be carried out by a VRF of 2, 3, 4, 5, 6, 7, 8, 9 or 10.
According to another embodiment, the solubilised protein solution can be subjected to one or more diafiltration step(s). Advantageously, the solubilised protein solution can be concentrated prior or after the diafiltration step(s). The level of concentration of the solubilised protein solution can achieve a volumetric reduction factor (VRF=[volume of the feed]/[volume of the retentate]) of 2 to 20 of the solubilised protein solution, but is advantageously ranging from 5 to 10, e.g., 8. The collected permeate from this microfiltration step is then subjected to the ultrafiltration step described subsequently.
According to another embodiment, the solubilised protein solution can be subjected to one or more ultrafiltration step(s). The ultrafiltration is preferably carried out using a filtering device made of a suitable material such as regenerated cellulose, a polysulfone (PS) or a polyethersulfone (PES) which has low protein retention. The molecular weight cut-off (MWCO) of the filter material may ranges from 1 kDa to 20 kDa, preferably from 1 to 10, most preferably from 1 to 5 kDa. Ultrafiltration may, as it is usual, comprise one or more subsequent concentration step or diafiltration step or combination of both.
This retentate is then diafiltrated. At least one and preferably more than one (e.g., 2 or 3) dialfiltration step can be carried out. This dialfiltration step can be carried out using non-salt water or an aqueous solution of salt, for example NaCl or CaCl2) at a suitable concentration. Such concentration may be ranging from 0.05 mol·L−1 to 1 mol·L−1, preferably from 0.4 mol·L−1 to 0.6 mol·L−1. The diafiltration step with salt water can be carried out after an ultrafiltration step which is carried out with water (and not salt water), possibly at an elevated temperature (e.g. 55° C.).
Once the NaCl diafiltration, and eventually the pH adjustment, is carried out, a further diafiltration step(s) can be carried out on the retentate obtained from the described NaCl diafiltration step with water. The water used can be tap water but is advantageously purified by removing organic/inorganic particles and contaminants and eventually dissolved gases. The retentate can be washed using preferably 1 to 10 DV of cold or hot water, preferably more than 1 DV, such as 2, 3, 4, 5, or 6 DV, or more.
In a particular embodiment of the process of the invention, step d) comprises the following step: subjecting the permeate of step c) to at least one ultrafiltration step, optionally followed by at least one diafiltration step, and harvesting the sunflower protein isolate.
It is also possible to precipitate the solubilised protein by adding ammonium sulfate as a kosmotropic agent, for example at a concentration of 5 mol·L−1 or 6 mol·L−1. Doing so, the amount of water available for protein solvation dramatically decreases resulting in selective protein precipitations. Therefore, in another particular embodiment, of the process of the invention, step d) comprises the following steps:
(d1) subjecting the solubilised protein solution obtained in step c) to at least one microfiltration step and harvesting a permeate, preferably wherein the solubilised protein solution obtained in step c) is concentrated by a volumetric factor at least 3, preferably from about 4 to 9, for example 5 or 8 and then the retentate may be washed with 2 diafiltration volumes of 0.5 mol·L−1 NaCl,
(d2) adding ammonium sulfate to the permeate of step d1) to form a mixture, for example up to up to 65% of saturation,
(d3) centrifugation of the mixture (e.g., 15 000×g for 30 min at 20° C.) and collecting the pellet,
(d4) dissolving the pellet of step d3 in water, preferably deionized water, to obtained an protein solution,
(d5) subjecting the protein solution obtained in step d4) to a low pressure chromatography system and eluting the protein with deionized water, and
(d6) collecting the protein solution peak recorded at 280 nm and less than 1% of conductivity.
Additionally, or alternatively, a pasteurising step can further be carried out before that the dia-ultrafiltration with water or salt water takes place. Such a pasteurising step can take place at 75° C. for 15 minutes. In a preferred embodiment the temperature can also be elevated either slightly (e.g., around 30° C.) or more positively (e.g., around 55° C.).
According to another embodiment of the invention the steps from step b) to step d) are carried out at ambient temperature or at an elevated temperature such as 55° C. but less than 85° C.
Step (e)
In order to obtain an isolate and to preserve the structure of the sunflower protein thus isolated by the process of the invention, it is advantageous to freeze dry, lyophilized or to spray dry the purified protein solution in order to obtain a dry powder. These techniques are well known in the art. For instance, to freeze dry, the purified protein isolate is frozen at temperature from −80° C. to −20° C. until complete freezing. Then freeze-drying is carried out by the use of a standard freeze-dried apparatus at a sublimation temperature around −20° C. To spray dry, it is customary to use a standard vertical spray dryer equipped with nozzle, with an inlet temperature ranging from 150° C. to 200° C. and an outlet temperature ranging from 70 to 90° C. These methods permit to obtain a powder having less than 7 wt % of water, and preferentially less than 4 to 6 wt %.
The process of the invention further encompasses a process wherein any one of steps above described may be repeated, eventually more than once.
Sunflower Seed Protein Isolate
The invention also provides a sunflower seed protein isolate obtainable or obtained by the process of the invention.
The sunflower seed protein isolate of the invention is preferably a native protein isolate having a protein content of at least 88 wt % by weight of the total dry matter (/dm), usually at least 90 wt %/dm and more advantageously at least 93 wt %/dm. Protein content is measured on dry matter by determining the nitrogen content using the Kjeldahl method (see the Examples, infra) and multiplying it by a conversion factor of 5.6 (i.e., Nx5.6 conversion factor) determined for sunflower proteins as described by Defaix et al. (Defaix et al., 2019; Pickardt et al., 2009).
Albumin proteins are identified by a molecular weight of 10-18 kDa measured by SDS-PAGE method (see the Examples, infra). According to a particular embodiment of the invention, the sunflower seed protein isolate obtainable or obtained by the process of the invention has an albumin protein content of at least 85 wt %, preferably at least 90 wt %, by weight of the total proteins.
Phytic acid content is determined by a method adapted from that described by Garcia-Estepa et al. (1999) (see the Examples, infra). According to another particular embodiment of the invention, the sunflower seed protein isolate obtainable or obtained by the process of the invention has less 4 wt %/dm, and by order of preference less than 3 wt %/dm, 2 wt %/dm and 1 wt %/dm of phytic acid.
Chlorogenic acid compounds are a major cause of the dark colour and undesirable taste of sunflower seed protein isolates. In sunflower seed, these compounds are mainly three isomeric forms of chlorogenic acid 3-CQA, 4-CQA and 5-CQA. According to another particular embodiment of the invention, the sunflower seed protein isolate contains no or negligible amount of any or all of such compounds and a method of obtaining it. A negligible amount of a chlorogenic acid compound can be an amount equal or inferior to 1%, preferably equal or inferior to 0.5%, advantageously equal or inferior to 0.2%, using the measuring method described herein below. Hence, a further object of the invention is a sunflower seed protein isolate obtainable or obtained by the process of the invention having a content of at least one chlorogenic acid isomer of at most 0.2 wt % by weight of the total proteins in said isolate. Preferably, the isolate has less than 0.2 wt % of all chlorogenic acids isomers, aka ‘chlorogenic acid’ by weight of the total proteins in said isolate.
The invention also provides a sunflower seed protein isolate obtainable or obtained by the process of the invention, wherein in step (b) describe above, said aqueous NaCl solution has a NaCl concentration ranging from 0.2 to 0.3 mol·L−1 and said pH is from 4.0 to 4.2, preferably said aqueous NaCl solution of step (b) has a NaCl concentration of 0.25 mol·L−1 and said pH is of from 4.05 to 4.15. According to this embodiment, the sunflower seed protein isolate has at least one, preferably all the following features:
In a preferred embodiment, the sunflower seed protein isolate has the features i, ii, iii, iv and v as defined above, more preferably the features i, ii, iii, iv, v and vi as defined above, and most preferably the features i, ii, iii, iv, v, vi, vii and viii as defined above.
Sunflower Seed Protein Composition
The invention also provide a sunflower seed protein composition (solids) obtainable or obtained by the process after step (c) and before step d) described therein. This sunflower seed protein composition is rich in proteins.
The sunflower seed protein composition (solids) of the invention is preferably a native protein composition having a protein content of at least 35 wt % by weight of the total dry matter (/dm), advantageously at least 40 wt %/dm. It preferably comprises a protein content ranging from 35 wt % to 55 wt %/dm, more preferably from 40 wt % to 55 wt %/dm.
The sunflower seed protein composition (solids) of the invention is preferably a native protein composition having a protein content of at least 35 wt % by weight of the total dry matter (/dm), advantageously at least 40 wt %/dm. Protein content is measured on dry matter by determining the nitrogen content using the Kjeldahl method (see the Examples, infra) and multiplying it by a conversion factor of 5.6 (i.e., Nx5.6 conversion factor) determined for sunflower proteins as described by Defaix et al. (Defaix et al., 2019; Pickardt et al., 2009).
According to a particular embodiment of the invention, the sunflower seed protein composition (solids) of the invention has less 4 wt %/dm, and preferably less than 3.8 wt %/dm, of phytic acid.
According to another particular embodiment of the invention, the sunflower seed protein composition (solids) of the invention has a phenolic compound content of less than 3.5 wt %/dm, and preferably less than 2.7 wt %/dm. The phenolic compounds content is advantageously comprised between 0.5 wt %/dm to 3.5 wt %/dm, preferably 1 wt %/dm to 2.7 wt %/dm. The phenolic compounds content is determined by the method described in the Example 2.1, infra.
Use
The invention also provides the use of the sunflower seed protein isolate as described therein in the food industry, for example as a main component, a supplement or an additive. Hence the protein isolate of the invention can be used according to the invention in food product or food ingredient, preferably for beverages, such as acidic beverage with a pH value less than 6, preferably less than 3.5, neutral beverage with a pH value comprised between 6 and 8 or basic beverage, with a pH value more than 8.
The invention also provides a food or beverage comprising a sunflower seed protein isolate of the invention, a method of making a foodstuff, or a food supplement, by adding and/or mixing said protein isolate to other ingredients. In a particular embodiment, the invention provides a drink (including a soft drink), particularly a coffee or a chocolate preparation including a whitener or not, comprising at least 1 wt % of the sunflower seed protein isolate according to the present invention. In another particular embodiment, the invention provides a nutritional composition, such as a milk product (for example a yogurt) comprising at least 1 wt % of the sunflower seed protein isolate of the invention.
The invention also provides the use of the sunflower seed protein composition (solids) obtainable or obtained by the process after step (c) and before step d) described therein as feed, dietary supplement or additive, for animal feeding.
Foregoing and other objects and advantages of the invention will become more apparent from the following detailed description and accompanying drawing, which refers to non-limiting examples illustrating the process according to the invention.
1.1. Chemicals
Sodium chloride (NaCl, CAS 7647-14-201), sodium hydroxide pellets (NaOH, CAS 1310-73-2) ethylenediaminetetraacetic acid (EDTA, CAS 6381-92-6) were obtained from VWR (Darmstadt, Germany). Hydrochloric acid (HCl, CAS 7647-01-0) was from Carlo Erba (Milan, Italy). Tris(hydroxymethyl)aminomethane (Tris, CAS 77-86-1), glycine (CAS 56-40-6), iron (III) chloride (FeCl3, CAS 7705-08-0), sodium sulfate (Na2SO4, CAS 7757-82-6), 5-sulfosalicylic acid hydrate (CAS 304851-84-1) was from Fisher Scientific (Hampton, USA).
1.2. Solid/Liquid Extraction
Sunflower cold press meal (or cake) was provided by Olead (Pessac, France). The protein, fat and phytic acid content in the meal were 42.8, 14.6, and 6.6% on dry matter basis, respectively. The meal was extracted with 1:9 solid/liquid ratio (w/w). The pH (3-6) and various concentrations of NaCl solutions (0-0.5 mol·L−1) were used according to the DoE matrix. The mixture was stirred at 600 rpm during 60 min at 20° C. If necessary, the pH was readjusted. Then, the slurry was centrifuged (15 000×g, 30 min, 20° C.) and partly clarified on a Whatman filter paper. About 1600 mL of aqueous extract was obtained. The remaining solid residue was collected, stored at −80° C. and freeze dried.
1.3. Experimental Design and Process Optimization
The design of experiment (DoE) and statistical analysis were based on response surface methodology (RSM). The influence of two factors: pH (3-6) and NaCl concentration (0-0.5 mol·L−1) on three responses: extraction yield (SFAYIELD%), content (CSFA%) and phenolic contamination (SFAPHEN mg·g−1) of albumins in liquid extract, as well as acid phytic (CPHYT%) and protein (CPROT%) content in remaining solid residue were taking into consideration. For design of experiments Modde 9.1.1.0 software form Sartorius Stedim Biotech (Gottingen, Germany) was used. According to generated experimental matrix, 11 experiments were performed containing three replications at the central point. The response surface curve, polynomial equation and observed/predicted plots were achieved. The mathematical relationship between factors and response was described by second-degree polynomial equation as follow:
y=β
0+β1x1+β2x2+β11x12+β22x22+β12x1x2
where:
y—response,
β0—constant,
β1—coefficient of linear effect of pH,
β2—coefficient of linear effect of NaCl concentration,
β11—coefficient of cubic effect of pH,
β22—coefficient of cubic effect of NaCl concentration,
β12—coefficient of interactive effect of pH and NaCl concentration,
x1—uncoded pH value,
x2—uncoded NaCl concentration value.
The obtained equations were statistically verified by evaluation of regression coefficient (R2), residual standard deviation (RSD) and analyse of variance test (p-value, lack of fit). The significance level of p=0.05 was claimed.
1.4. Multi-Objective Optimization
The multi-objective optimization of extraction process was carried out with using the predictive equations of RSM. The genetic-evolutionary algorithms were employed to identify the optimum of experimental parameters in term of pH and NaCl concentration. The non-dominated solutions were calculated including the set of constraints: SFAYIELD>70%, CSFA>90%, SFAPHEN<1.6 mg·g−1%, CPHYT<4% and CPROT>40%. The selected optimal conditions were validated on additional experimental batch extraction by comparing the observed responses with prediction intervals of models (PI) and calculation of relative error (RE). The optimization runs (n=2 000) and all data exploitation were carried out using MATLAB software from MathWorks (Natick, USA).
1.5. Protein Purification
Protein purification was performed in three principal stages: extract clarification by microfiltration, protein precipitation from extract using ammonium sulfate and protein desalting by size exclusion chromatography. The microfiltration step was carried out on Akta system from GE Healthcare (Illinois, USA) using Hydrosart membrane system (0.2 μm 200 cm2) from Sartorius (Gottingen, Germany). The 4 L of collected liquid extract was concentrated by a volumetric factor of 8 and then the retentate was washed with 2 diafiltration volumes of 0.5 mol·L−1 NaCl. The total microfiltration permeates were pooled for next step. Ammonium sulfate was added to microfiltration permeates up to 65% of saturation and stirred for 30 min at a room temperature. After protein precipitation and centrifugation step (15 000×g for 30 min at 20° C.) obtained pellet was dissolved in 750 mL of deionized water. The desalting of proteins was carried out on low pressure chromatography system of Akta Pure from GE Healthcare (Illinois, USA). The sample volume of 20% bed volume of column (10 cm of height, 5 cm of diameter) was injected into G-25 Fine silica gel (GE Healthcare, Illinois, USA). The elution was performed at 10 mL·min−1 with deionized water and the peak corresponding to protein recorded at 280 nm and less than 1% of conductivity was collected and freeze-dried.
1.6. Analytical Methods
1.6.1. Protein Characterization by Gel Electrophoresis
Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was performed according to Leammli method (Laemmli et al., 1970). Sunflower aqueous extract was diluted in 0.1 mol·L−1 of sodium phosphate buffer at pH 7 to obtain a final concentration of 2 g·L−1. Then, the sample was solubilized in 50 μL of Laemmli buffer containing 2% β-mercaptoethanol (v/v) and heated at 95° C. for 5 min. Molecular weight markers ranging 250-10 kDa (Precision Plus Protein Standards) and 26.6-6.5 kDa (Polypeptide SDS-PAGE Standard) were used (Bio-Rad, Hercules, USA). 10 μL of sample and markers were deposed to a 5% of polyacrylamide stacking gel and separated in a 17% polyacrylamide resolving gels. The migration step was carried out at 20 mA per gel. The gels were subsequently stained with Coomassie Brilliant Blue and destained overnight in 10% acetic acid solution (v/v).
1.6.2. Determination of Albumin and Globulin Content by SE-HPLC
SE-HPLC analysis was performed according to the method of Defaix et al. (Defaix 2019). The analyses were carried out on HPLC Shimadzu LC30 system coupled with photodiode array (PDA) detector and operated by LabSolutions software, all from Shimadzu Corporation (Kyoto, Japan). 5-20 μL of sample was injected onto a Biosep SEC s-2000 column (300×7.8 mm; 5 μm) from Phenomenex (Torrance, USA). The exclusion rang of molecular weight was comprise from 1 to 300 kDa. During analysis the autosampler and column compartment were maintained at 20 and at 35° C., respectively. The mobile phase consisted in acetonitrile/water/trifluoracetic acid (45:54.9:0.1 v/v). The elution flow rate was set at 0.6 mL·min−1. All solvents were HPLC grade and were supplied from Fisher Scientific (Hampton, USA). The ultrapure water (H2O) with resistivity≥18.2 MΩ.cm−1 was used. The PDA signal was recorded between 190 and 400 nm with maximal absorption at 214 or 280 nm for protein or 325 nm for phenolic compound detection. To determine globulin and albumin proportion in sunflower aqueous extract the meal globulin/albumin ratio (70:30) was considered. This ratio corresponds to the mean value denoted in several articles (Mazhar et al., 1998; Baudet et al., 1977; Raymond et al., 1995). All measurements were performed in triplicate and the average value was calculated. All measurements were performed in triplicate and the average value was calculated.
1.6.3. Determination of Soluble Proteins, Free Chlorogenic Acid Isomers and Chlorogenic Acid Bound
Sunflower protein and free chlorogenic acid isomer (3-CQA, 5-CQA et 4-CQA) content were determined using the SE-HPLC method of Albe Slabi et al. (Albe Slabi, 2019). The same HPLC system and chromatographic as presented in section 1.6.2 was used except for mobile phase composition (acetonitrile/water/formic acid (10:89.9:0.1 v/v)). The amount of covalently fixed CQAs (milligram of CQAs bound per one gram of SFAs) were quantified considering the surface of peak at 325 nm eluted at the retention time of SFAs and the average calibration slop of 3-CQA, 5-CQA et 4-CQA. The mass of soluble SFAs in liquid extract was determined basing on the concentration of total sunflower proteins and the proportion of SFA fraction in aqueous extract (section 1.6.2.). In the calculations, the same value of molar attenuation coefficient for helianthinins and SFAs was assumed, as it has been confirmed by Defaix et al. (Defaix, 2019). All analysis was carried out in triplicate and an average value was calculated.
1.6.4. Kjeldahl Method
The measure of total nitrogen content in sample was carried out in accordance to Kjeldahl method procedures described in AOAC method 991.20 (AOAC). 0.5-2 mL of sample was mineralized in a digestion flask with 4 mL of 96% H2SO4 (v/v) and approximately 10 mg of catalyst. The mineralization step was achieved at 450° C. during 150 min. After this time, the solution was distilled with 32% NaOH (w/v) and the mixture was titrated against 0.01 mol·L−1 HCl. A blank consisted of non-protein containing sample. The non-protein nitrogen in sample was determined in supernatant after protein precipitation using 50% trichloroacetic acid (w/v). A nitrogen to protein conversion Nx5.6 was used. All analyzes were repeated in triplicate. Average values of concentration and standard deviation were calculated.
1.6.5. Quantification of Phytic Acid
Phytic acid determination was adapted from the method described by Garcia-Estepa et al. (Garcia-Estepa et al., 1999): 250-300 mg of solid sample was stirred with 20 mL of 0.4 mol·L−1 HCl/10% Na2SO4 (w/v) for about 120 min at room temperature. The blank consisted of deionized water. Then, the mixture was centrifuged (10 000×g, 30 min, 20° C.) and supernatant was additionally clarified using 0.22 μm filter. In centrifuge tube 2.5 mL of 20 mmol·L−1 FeCl3, 0.4 mol·L−1 HCl/10% Na2SO4 (w/v), 20% sulfosalicylic acid (w/v) and filtered supernatant were mixed and heated at 100° C. for 20 min. After cooling at room temperature, the sample was centrifuged (10 000×g, 30 min, 20° C.) filtered (0.22 μm) and transferred into 50 mL volumetric flask. The pellet was washed with 4 mL of deionized water, centrifuged (10 000×g, 10 min, 20° C.) and supernatant was added after filtration (0.22 μm) into 50 mL volumetric flask. This step was repeated in triplicate. The volumetric flask was completed with deionized water. The pH of 20 mL of obtained solution was adjusted to 2.5±0.5 using glycine. After heating to 70-80° C. the sample was tittered against 2 mmol·L−1 EDTA solution. The equivalent volume was reached when the solution changes color from burgundy to yellow-green. The results were expressed as phytic acid content on dry matter basis of sunflower meal (Ac Phyt/dm %). All measurements were performed in triplicate and average value was calculated.
1.6.6. SFA Solubility
The SFA isolate was suspended in deionized water at 5.0 g·L−1 (room temperature). The pH was adjusted to a given value by adding either 0.1 mol·L−1 NaOH or 0.1 mol·L−1 HCl and kept constant during 30 min. Then, the slurries were centrifuged (15 000×g, 20 min, 20° C.). The protein concentration in supernatant was measured by SE-HPLC according to Albe Slabi et al. (Albe Slabi et al., 2019 cf. ref 1). All analyzes were repeated in triplicate and average value of concentration and standard deviation were calculated.
1.6.7. Colour Measurement
The solution of protein powder in deionized water was prepared at a concentration of 1% (w/v) and clarified thought 0.22 μm membrane filter. The color was recorded in CieL*a*b* scale using Lovibond PFX195 Tintometer at room temperature. The measure was performed in ten repetitions and average value of L*, a*, b* parameters with standard deviation were calculated.
1.6.8. Amino Acid Composition
The determination of amino acid content (except tryptophan) was performed according ISO 13903:2005 procedures. Generally, protein sample (10 mg) was first hydrolysed using hydrolysis mixture solution and the hydrolysate was injected into C18 column of HPLC system. Composition in amino acids was determined by reaction with ninhydrin using photometric detection at 570 nm or 470 nm (for proline). For tryptophan quantification (EU 152/2009) sample was hydrolysed with barium hydroxide solution at 110° C. for 20 h and then injected into C18 column of HPLC system coupled with fluorescence detector (excitation 280 nm, emission 356 nm).
1.6.9. SFA Structure Study
SFA proteins were analysed by Dynamic Light Scattering (DLS), Circular Dichroism (CD) and Differential Scanning calorimetry (DSC). DLS measurement was recorded on Zeta Sizer Nano-S from Malvern Instruments (Worcestershire, UK). 20 μL of filtered (0.22 μm) SFA solution at a concentration of 1 g·L−1 at pH 4 (10 mM sodium phosphate buffer), pH 7 (10 mM sodium phosphate buffer) and pH 9 (10 mM borate buffer) were used. Samples were maintained at 25° C. during measurement. The number of acquisitions varied between 13 and 17 scans. Volumetric distribution of particle size was determined.
Circular dichroism (CD) was carried out using a Chirascan Plus device from Applied Photophysics (Leatherhead, UK). The far-UV spectra were obtained in the UV region of 180-280 nm. CD spectrophotometer was kept under constant flow of nitrogen gas and the temperature was maintained at 20° C. Samples were prepared in the same way as for DLS analysis and blank assay corresponding to appropriate buffer solution was subtracted. All spectra were repeated at least in triplicate and mean measurement was calculated. The far-UV spectra were converted into mean residue ellipticity [θMRE] using the number of amino acids set at 130. The content of α-helix, β-sheet structures and random coils were obtained after spectrum deconvolution using CDNN software version 2.1.
DSC determination was carried out by Microcal VP-DSC from Malvern Panalytical (Worcestershire, UK) using the SFA solution at a concentration of 2 g·L−1 in 10 mM sodium phosphate buffer, pH 7. The solution was filtered (0.22 μm) prior analysis. The blank assay consisted on protein-free phosphate buffer. Thermogram was acquired over a temperature range from 20 to 130° C. at the rate of 1° C./min. Thermal properties of SFA were expressed as the temperature of denaturation (Tm) and the enthalpy calorimetry ΔHcal, that represents the total amount of energy emitted during the denaturation process.
1.6.10. Functional Properties of SFA Isolate
For foaming study, 20 mL of 1% (w/v) SFA isolate solution in 10 mM sodium phosphate buffer pH 7 was mixed at 10 000 rpm during 5 min at room temperature using Ultra-Turrax T25 digital homogenizer from IKA (Staufen im Breisgau, Germany). Foaming capacity was calculated as ratio (in %) foam volume (mL) in to initial volume of SFA solution (mL). The foaming stability was expressed as the percentage of remaining foam volume (mL) after 5, 15, 30, 60 and 120 min.
The emulsifying capacity were evaluated using 5% (w/v) SFA isolate solution in 10 mM sodium phosphate buffer pH 7. For this purpose, 5 mL of above mentioned SFA solution was mixed with 2.5 mL of sunflower oil and mixed by Ultra-turax Homogenizer at 10 000 rpm at room temperature during 30 s. After this time, 2.5 mL of sunflower oil was additionally mixed at the same speed during 90 s. The mixture was then centrifuged at 1 100×g during 5 min. The ratio (in %) of emulsion volume (mL) after centrifugation and initial volume of mixture (mL) represents the emulsifying capacity. The stability of emulsion was determined after heating at 85° C. during 15 min and additional centrifugation (the same parameters). It represents the ratio (in %) of emulsion volume (mL) after and before heating. All measurement of technofunctional properties were carried out in triplicate and average value with standard deviation were calculated. The technofunctional properties of soy proteins (Solae, St. Louis, USA) measured in the same way as those of SFAs were used as a reference.
1.6.11. Other Method for Determining Protein Solubility at Room Temperature
The solubility of the protein isolate of the invention in aqueous solution is measured as follows:
About 250 mg of protein powder is weighted, the exact mass recorded, and then dissolved in 25 mL of a distilled water in a beaker. The solution is transferred into a 50 mL volumetric flask at room temperature. The beaker is washed three times with 5 mL of distilled and the washing solutions are transferred into a 50 mL volumetric flask. Finally, the volumetric flask is completed with distilled water. A volume of 15 mL of solution is placed in a 25 mL beaker and stirred on a stirring plate at approximatively 300 rpm at room temperature during 10 min. Then, the pH of the solution is adjusted to the required pH using an aqueous solution of 0.1 mol·L−1 NaOH or 0.1 mol·L−1 HCl and the stirring maintained during 30 min. After this time, the solid precipitate was separated by centrifugation at 11 000×g during 20 min at 20° C. Subsequently, the concentration of protein in the initial solution and in the supernatant was measured according the Kjeldahl method (Nx5.6). The protein solubility at the given pH was calculated as follows (Equation 1).
2.1. Effect of pH on Sunflower Proteins Extraction
The trial study aimed at selecting the pH range that allows selective extraction of SFAs. For this purpose, the influence of pH (3-7) on extraction of helianthinins and SFAs was investigated (
The graph in
Besides, albumins (SFAs) extraction was high (around 50%) under strong acidic conditions (pH 3-5). It started to decrease at pH 6 to reach a minimum at pH 7 (25.87±2.38%).
The graph in
The SFAs extraction yield under acidic condition was only around 50%. The effect of NaCl was also investigated. To assess a synergetic effect between NaCl and pH, a design of experiments (DoE) study was implemented. Beside SFAs extraction yield, SFAs composition and polyphenol content in the liquid phase was investigated.
2.2. Effect of pH and NaCl on SFA Yield, Content and Phenolic Contamination During Acidic Extraction
The effect of the pH (x1) (3-6) and mild NaCl concentration (x2) (0-0.5 mol·L−1) on SFA extraction was studied by DoE. Responses related to the liquid extract and the solid residue co-produced after protein extraction were considered. Regarding the liquid phase, albumin extraction yield (SFAYIELD%), albumin content (CSFA%) and phenolic contamination of albumins (SFAPHEN mg·g−1) were measured. Concerning the solid reside, the phytic acid (CPHYT%) and protein (CPROT%) content were considered. A full factorial design with experimental matrix consisting on 11 runs with the central point repeated in triplicate was applied. The obtained coefficients of the models (constant terms, cubic effects, interactions) and statistical model parameters (R2, RSD, p-value, lack of fit) are summarized in Table 1.
The albumin extraction yield (SFAYIELD%) ranges between 32.06 and 75.25%. According to statistical analysis, pH (x1), NaCl concentration (x2), and both cubic terms of pH (x11) and NaCl concentration (x22) significantly impacted the studied response. Regression coefficient (R2) was high and demonstrates that 95.3% of data fitted model. The variation between predicted/observed plot was 3.61. The model p-value (0.000) and the lack of fit (0.056) were not significant. From these results antagonist effect of pH and ionic strength could be deduced. As observed above, DoE shows that SFA extraction yield is negatively impacted at low NaCl concentration by increasing pH (minimum around 6). This also demonstrates that SFA extraction can be considerably increased by NaCl in the whole studied pH range. The maximal extractability of albumins (75%) was found in the area of pH from 3.25 to 4.35 and salt addition above 0.33 mol·L−1.
Relationship of pH and salt concentration on the albumin proportion in the liquid phase (CSFA%) was also investigated. The obtained response function varied from 37.6 to 90.8%. The results of ANOVA analysis show that the significant terms are pH (x1), NaCl concentration (x2), quadratic terms of pH (x11) and interaction of pH and NaCl concentration (x12) at the 95% confidence level. The regression parameters of the obtained model (R2=0.981, RSD=3.10) confirm well-fitting predicted and actual values. Furthermore, p-value of the model (0.000) and lack of fit (0.051) confirm good model performance. The impact of both pH and NaCl was low between pH 3 and 5. Under these conditions the SFAs content was relatively constant (80-90%) whatever the ionic strength. The constant albumin proportion in relation to above-mentioned conclusion about increased SFA extraction yield in this pH region suggests that upon salt addition extraction yield of helianthinins is also improved. This finding indicates a protective effect of ionic strength against helianthinins denaturation under acid conditions. A negative effect of pH and NaCl was observed between pH 5 and 6. Obviously, this effect is reinforced by high NaCl concentration since the lowest SFA content (<40%) was observed at pH 6 and 0.5 mol·L−1. Thus, selective extraction of SFAs could be only achieved in lower pH values regardless of used salt concertation.
The third response studied was the amount of CQA covalently bound to SFAs (SFAPHEN mg·g−1). The response surface shows a significant effect of pH (x1), NaCl concentration (x2), cubic term of pH (x11). The responses varied in the range of phenolic contamination between 0.6481 and 2.1201 mg·g−1. The plot of predicted vs actual values showing R2=0.937, RSD=0.15, p-value=0.000 and lack of fit=0.311 confirm reliable model for predictive purposes. From these results, the antagonistic effect of pH and NaCl concentration on the irreversible phenolic association with SFAs can be observed. At low NaCl concentration, the higher pH value leads to phenolic complexation. The maximum of phenol bonding to proteins was observed at pH 5-6 without salt addition (>2 mg of CQA per gram of SFAs). This effect is most pronounced under strong acidic pH. CQA (the main phenolic compounds in sunflower) has a negative net charge (pKa=3-3.5) in this pH range, while SFAs are positively charged. Hence, there is probably a high concentration of CQAs at SFA surroundings due to electrostatic interactions that might potentiate the covalent binding between these two compounds.
2.3. Effect of pH and NaCl on Phytic Acid and Protein Content in Solid Residue after Acidic Extraction
The nitrogen content and the level of phytic acid are known to make a large part of oilseed meal value for feed applications. Phytic acid (3-10% on dry matter base of sunflower meals) is particularly pointed out due to decrease of biodisponibility of some minerals (Ca2+, Mg2+, Zn2+, Fe2+, Mn2+, Cu2+) and proteins in digestive tract by forming unabsorbable complexes (Nissar et al., 2017; Kumar et al., 2010; Cheryan et al., 1980).
Based on RSM results the content of phytic acid content in solid residue after extraction (CPHYT%) was found in the range from 3.14 to 6.69%. Among all variables, pH (x1), NaCl concentration (x2) quadratic term of pH (x11) and interaction of both (x12) were kept in model equation. The interaction between factors (x11) has no noticeable impact on the response. The obtained model fits within acceptable limits (R2=0.802, RSD=0.74) in relation to actual data. Good p-value (0.027) of the model and no significant lack of fit (0.154) were also observed. The results indicate that the content of phytic acid in the solid increases considerably near mild acidic pH and reaches the higher values around pH 3 and 6. On the other hand, the negative effect of ionic strength was also observed in the region of moderate acidic pH. The highest phytate reduction (50%) in relation to initial phytic acid content in sunflower meal (6.6%/dm) was observed in the zone of pH 4-4.5 and 0.4 mol·L−1 NaCl. Surprisingly, these extraction behaviours of phytic acid coincides well with the extraction profile of SFAs.
Regression coefficients and ANOVA test were also applied to study the influence of pH and NaCl concentration on protein content in sunflower solid residue (CPROT%) ranging from 29.74-45.68%. The analysis of variance reveals the negative effect of pH (x1), NaCl concentration (x2), quadratic term of pH (x11) and pH/NaCl concentration interaction (x12). The excellent value of R2 (0.942), RSD (1.31), p-value of model (0.001) and lack of fit (0.154) indicate good accuracy of the model for simulation of protein content in solid residue. Minimum residual protein content in solid was achieved with high value of pH (about 6) and high NaCl concentration (about 0.5 mol·L−1). In contrast, protein rich solid is produced using pH ranging from 4 to 5.5 without salt addition.
2.4. Multi-Objective Optimization
The regression equations of models were used to identify the most suitable process for solid/liquid extraction. The objective of the optimization was to maximize extraction yield and content of SFAs, while minimize phenolic contamination of albumins. Simultaneously, a value-added residual solid characterized by high protein level and reduced content of in phytic acid was desired. To reach the targets, the following constraints were selected: SFAYIELD>70%, CSFA>90%, SFAPHEN<1.6 mg·g−1, CPHYT<4% and CPROT>40%. The set of non-dominated solutions from multi-objective optimization was presented in
The accuracy of models to predict the process performance was verified by comparing of predicted and actual values from the extraction in the optimum of parameters. The results of model validation including prediction interval (P1), experimental value and relative error (RE %) were summarized in Table 2.
The observed responses for CSFA (88.78%), SFAPHEN (1.55 mg·g−1), CPHYT (3.74%) and CPROT (40.97%) are included in the 95% probably prediction intervals. Only SFAYIELD (61.62%) was lower (SE=−14.20%) in relation to estimated values. This is probably due to extremely good replicates in the central point of DoE (reproducibility=99.3%) resulting in narrow prediction interval (±2.83%). However, experimental data of SFA yield fit to predicted values within a satisfying limit of RE≤15% that is commonly used for process prediction. Therefore, the selected condition guarantees the most sustainable albumin extraction from sunflower meal and the increase of the value of residual solid.
2.5. Characterisation of SFA Obtained from Acid Extraction
Based on the results of multi-objective optimization, SFAs were extracted at pH 4.1 and 0.25 mol·L−1 NaCl and purified by Size Exclusion chromatography. The chemical composition of the obtained SFAs were displayed in Table 3.
The proteins (93.3% proteins/dm) revealed rich in sunflower albumins (90.3±0.1%). They contained virtually no free chlorogenic acid isomers (not detected), low in phytic acid (0.88%/dm) and its colour was close to white (L*=92.2±2.9, a*=1.3±0.7, b*=11.0±1.4) (
The structural and functional properties of isolated SFAs were also studied. The graph in
In order to assess information about size distribution of SFAs under various pH values (4, 7, 9) DLS analysis were performed. A single peak with size between 3.8-4.0 d. nm (99.5-100% of total sample volume) observed whatever the pH. The peak was attributed to low molecular weight proteins of 18-19 kDa. This confirms that the sample consisted mainly on SFAs (with theoretical molecular weight of 10-18 kDa) (Gonzalez et al. 2005; Gonzalez et al., 2007; Kortt et al., 1990; Berecz et al. 2010; Salgado et al., 2012; Gueguen et al., 2016) and no protein aggregation was detected.
On the other hand, the stability of SFA structure against various pH and thermal treatment was investigated. The far-UV CD spectra indicated that there is no change in SFA conformation state according variable pH values (under acidic (pH 4), neutral (pH 7) and alkaline (pH 9) conditions). The deconvolution of CD spectra revealed similar proportion of α-helix (36-38%) and β-sheet (36-37%) elements, while random coil forms were found at slightly lower level (26-27%). Also, DSC thermogram of SFA solution at pH 7 recording the peak of denaturation at the temperature of 120.8° C. with a ΔHcal of 17.6 kcal·mol−1 indicated extremely high thermoresistance of produced SFAs.
The foaming (a) and emulsifying proprieties (b) of SFAs were compared with commercially available soy proteins used as a reference. Generally, SFAs present higher foaming capacity (359±14% of initial solution volume) in relation to lower value obtained for soy proteins (297±19% of initial solution volume). Also, the foam formed by SFAs (about 50% over 120 min from mixing) turned out to be more stable comparing to the more labile foam of soybean proteins (24±14% over 120 min from mixing). On the other hand, two proteins evaluated exhibited comparable ability to form emulsion (about 40-45% of initial solution volume) and the thermal stability at 85° C. of emulsions obtained by SFAs and soy isolate were similar and very high (about 95-100% of initial emulsion volume). Altogether, the obtained results met the principal conclusions that SFAs have an important potential as an alternative plant-based ingredient.
2.6. Solubility of Sunflower Albumin Isolate at pH 4, 7 and 9 Determined by the Kjeldahl Method.
The solubility of sunflower albumin isolate at pH 4, 7 and 9 determined by the Kjeldahl method was 104.2%, 101.8% and 100.8%, respectively.
The results show that solid/liquid extraction at pH 4.1 and 0.25 mol·L−1 NaCl resulted in selective production of SFAs (88.78%) with low phenolic contamination (1.55 mg·g−1) and good SFA extraction yield (61.62%). Simultaneously, the optimal extraction condition allowed obtaining the protein-rich solid residue (40.97%/dm) with reduced amount of antinutritional phytate (3.74%/dm) and other non-protein compounds. The colorless SFAs isolated in the optimum of extraction parameters were perfectly soluble in water (about 100% in all pH range 2-11) and highly resistant on both pH and temperature treatments. Also, comparable functional properties (foaming and emulsifying) of SFAs to the routinely used protein products from soybean were shown. SFAs are therefore very promising as valuable protein-based ingredient which could be successfully used in various food applications. Therefore, proposed alternative process for preparation of SFAs and residual solid is an answer for sustainable valorization of sunflower meal in food and feed industry.
A sunflower albumin isolate (SFA) was obtained according to the process of the invention. The process and the analytical methods were performed as described in Example 1, unless otherwise specified. For this purpose, a cold press meal from (fully) dehulled sunflower kernels was used (see Table 5 below).
2.1 Determination of Total Phenolic Content
Total phenolic content was measured according to ISO 14502-1: 2005 procedures (Determination of substances characteristic of green and black tea—Part 1: Content of total polyphenols in tea-colorimetric method using Folin-Ciocalteu reagent. In ISO 14502-1 International Standardization (p. 10). International Organization for Standardization Switzerland). For this purpose, solid sample was first extracted with 70% methanol (v/v) respecting the solid/liquid of 1:25 (w/v) at 70° C. during 10 min. After this time, the mixture was cooled to room temperature and centrifugated at 3 500 rpm at 20° C. for 10 min. The obtained pellet was subjected to the second extraction using the same parameters. The supernatant from both extraction steps was pooled and then analysed within 24 h.
Total phenolic content in supernatant was measured colorimetrically using Folin-Ciocalteu reagent. The calibration curve was performed using gallic acid stock solutions prepared in the concentration range from 10 to 50 g/L. 200 μL of supernatant, calibration stock solution or water (blank essay) was mixed with 1 mL of Folin-Ciocalteu reagent (previously diluted ten times with ultrapure water) and stirred energetically during 1 min. Between 1 and 8 min from stirring, 0.8 mL of sodium carbonate (7.5% w/w) was added. The mixture was left for 1 h at 20° C. After this time, the absorbance was recorded at 765 nm at 23° C. The concentration of total phenolics in supernatant was expressed in gallic acid equivalent.
Starting Material
Solid/Liquid Extraction
500 g of the cold press meal from dehulled sunflower kernels was mixed with a solution of NaCl (0.25 mol·L−1) in a solid/liquid ratio of 1:9 (wt %). The pH was adjusted to 4.1 using a solution of HCl (1.0 mol·L−1). The mixture was stirred at 300 rpm at room temperature during 60 min. After centrifugation conducted at 15 000×g during 30 min at 20° C., the supernatant was filtered using a Whatman filter paper (Fisherbrand, cellulose, diameter 190 mm, thickness 0.17 mm, particles retention 17-30 μm). The liquid phase was collected to be purified. The wet residual solid was freeze-dried and then analysed. The wet residual solid was freeze-dried and then analysed.
Purification
Protein purification was performed in three principal stages: extract clarification by microfiltration, protein precipitation from extract using ammonium sulfate and protein desalting by size exclusion chromatography. The microfiltration step was carried out on Akta system from GE Healthcare (Illinois, USA) using Hydrosart membrane system (0.2 μm 200 cm2) from Sartorius (Gottingen, Germany). The 3.8 L of collected liquid extract was concentrated by a volumetric factor of 5. The microfiltration permeate was used for next step. Ammonium sulfate was added to microfiltration permeates up to 65% of saturation and stirred for 30 min at a room temperature. After protein precipitation and centrifugation step (15 000×g for 30 min at 20° C.) obtained pellet was dissolved in 190 mL of deionized water. The desalting of proteins was carried out on low pressure chromatography system of Akta Pure from GE Healthcare (Illinois, USA). The sample volume of 20% bed volume of column (10 cm of height, 5 cm of diameter) was injected into G-25 Fine silica gel (GE Healthcare, Illinois, USA). The elution was performed at 10 mL·min-1 with deionized water and the peak corresponding to protein recorded at 280 nm and less than 1% of conductivity was collected and freeze-dried.
Results
The powder had a high purity (104.7% on dry matter basis) and was rich in sunflower albumins (89.0%) and had a low phytic acid content (0.7% on dry matter basis) (Table 6).
The protein solubility of sunflower albumin isolate at pH 4, 7 and 9, determined by the SE-HPLC method, was 100%, 97.2% and 95.9%, respectively (Table 7).
The protein solubility at pH 4, 7 and 9, determined by the Kjeldahl method, was 94.8%, 94.2% and 96.7%, respectively (see Table 8 below).
The composition of sunflower residual solid (Table 9) was 53.9% and 3.8% of protein and phytic acid content, respectively.
A sunflower wet residual solid was obtained according to the process of the invention using a cold press meal from (fully) dehulled sunflower kernels was used (see Table 5). The process and the analytical methods were performed as described in Examples 1. The determination of total phenolic content is described in section 2.1.
Solid/Liquid Extraction: 200 g of the cold press meal from dehulled sunflower kernels was mixed with a solution of NaCl at different concentrations—0.6 mol·L−1, 0.25 mol·L−1 and 0.2 mol·L−1—in a solid/liquid ratio of 1:9 (wt %). The pH was adjusted to different values—respectively 3.0, 4.1 and 4.5—using a solution of HCl (1.0 mol·L−1). The mixture was stirred at 300 rpm at room temperature during 60 min. After this time, the mixture was centrifuged at 15 000×g during 30 min at 20° C. and the pellet that represented the wet residual meal was collected to be analysed (Table 10). the
Results
| Number | Date | Country | Kind |
|---|---|---|---|
| 19305608.2 | May 2019 | EP | regional |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2020/063097 | 5/11/2020 | WO | 00 |
