This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Mustard is a series of plants of the genera Brassica and botanical family Cruciferae. Evidence of human use of mustard seeds has been traced back to 4000 B.C. in China and Pakistan (Fenwick, Heaney and Mullin 1982), carbonized seeds dated to 3000 BC have also been found in Iraq, evidencing the use of mustard by the Mesopotamian civilization (Zohary and Hopf 2000). The cultivation of mustard is believed to have been introduced to Europe by the moors. Its spread around Europe during the middle ages can be explained by factors such as the Crusades and the development of commerce around the Mediterranean (Fenwick, Heaney and Mullin 1982). The English word mustard has its origin in the French term "moustarde", from Latin "mustum".
2.1.1 Types of mustard
There are three different kinds of mustard seeds: black mustard (brassica nigra), popular in the Middle East and parts of Asia; brown and oriental mustard (brassica juncea), whose origin is uncertain, with proposed sources between Eastern Europe, the Middle East or China (Labana and Gupta 1993); and yellow mustard (sinapis alba), which originated in the Mediterranean region and is broadly consumed around the world.
Black mustard seeds are roughly globular with a diameter of 1 to 1.5 mm and a dark brown colour; the seed coat is pitted and when soaked in water the seeds produce a strong pungent odour. Brown mustard seeds are similar to black mustard seeds, their diameter is less than 2 mm and have a reddish brown to dark brown colour, it is primarily grown for the European market and has also become popular in North America as a replacement of yellow mustard. Oriental
mustard seeds vary in colour from yellow to dark yellow and brown. It is mostly used in the Asian and Japanese markets as a condiment. These varieties have a pungent taste and contain about 28% of oil and 30% of protein (Heath 1981). Yellow mustard seeds, on the other hand, vary in colour from light creamy yellow to yellow and in some cases yellowish brown, have a roughly globular shape and have a diameter of 2 to 3 mm; the seed coat is minutely pitted, and seeds turn mucilaginous when soaked in water. Yellow mustard has a pungent taste and is low in starch, contains about 30% of oil and 25% of protein (Heath 1981) (Figure 1).
Yellow MustardC:\Users\Ben\Desktop\U OF T\Tesis\mustards.jpg
Figure : Types of mustard seeds
Mustard is a broad-leaved, yellow-flower plant that requires a short growing season, between 85 to 95 days for yellow mustard seeds to reach maturity and between 95 to 105 days for the oriental and brown varieties to reach maturity (McKenzie 2010). Crops require an annual precipitation of between 350 and 450 mm and give higher yields in temperate zones with a cool and dry weather. Mustard is capable of growing in a variety of soils from sandy loam to clay loam (Agroecommerce Network Private Ltd. 2002). Mustard seeds are considered more tolerant to frost, drought and heat than other crops like canola or flax, which makes the dry brown and dark soils, warm dry summers and cold dry winters in the southern Canadian prairies an ideal place for mustard growth.
The sowing of mustard in Canada began in the 1930s with a modest 40 hectares, but in the next 30 years, it quickly grew to 60,000 hectares (Agriculture and Agri-Food Canada 2011). As of 2007, mustard crop occupied 176,000 hectares of harvested area with an annual production of 114,000 tonnes, representing a farm gate value of around 100 million dollars (Agriculture and Agri-Food Canada 2007). Canada is considered the world largest exporter of mustard seed and the second largest producer (Canadian Special Crops Association 2007) surpassed only by India. Mustard seed production for the year 2011 is presented in Table 1.
Table : Seeded area and production of Canadian mustard
1 Thousand hectares. November Estimates of Production of Principal Field Crops, Catalogue no. 22-002-X, vol. 90 no. 8 Released December 6, 2011; Statistics Canada
2 Thousand tonnes. Small Area Data 1976-2010 Statistics Canada, Agriculture Division, Crop Section
2.1.2 Mustard uses
Pythagoras mentioned the use of mustard seeds for scorpion stings and Hippocrates used it for the preparation of medicines. The medicinal properties of mustard were known to the Greeks and Romans, and ancient documents written by Cato, Columella and Pliny (Fenwick, Heaney and Mullin 1982) suggest that mustard seeds were cultivated and used as a condiment, mixing the ground seeds with wine must to make a paste, hence the name "mustard". The use of mustard seeds to prepare food condiments is still their main use, and has a wide range of applications in the food industry. Dried seeds are milled for flour production and wet milling is used to manufacture mustard paste. Whole ground seeds are also used for spice mix preparations and meat processing. Traditional or hot-dog mustard is prepared using the whole ground seed.
Mustard is also used as a protein source, flavour enhancer and as a binder in the manufacturing of processed meats. The different mucilage contents in the three varieties of mustard allow the manufacturing of products with different viscosities. Seed hulls are also used as a thickening agent and stabilizers in prepared foods. Heat inactivated whole ground seed is used in a variety of food products to enhance their flavour, colour, texture and viscosity and it can also be used as an emulsifier agent. The presence of sinigrin in the brown and oriental varieties makes them suitable for the manufacture of hot mustard for the European market and the production of mayonnaise. High oil content oriental mustard is used to cover the oilseed demand in the Indian subcontinent where one of its main uses is cooking oil production (Jimmerson 2005).
2.2 Mustard seed components
Mustard seeds contain a hull that represents between 15 and 20% of the seed weight and is composed of a hygroscopic integument containing lignin, cellulose, hemicellulose and mucilage, while the kernel makes up between 80 and 85% and contains most of the oil, proteins and soluble sugars. They have a thin endosperm membrane and occur in seed pods varying in quantity from 10 to 40 seeds (Appelqvist 1971). Mustard seeds contain 28-32% protein by weight and 30-35% of oil, although these values can vary slightly between varieties, growing regions and crop years as shown by Figures 1 and 2 (Canadian Grain Commission 2012).
Figure : Average fix oil content of mustard seeds in Canada
Survey data from the Grain Research Laboratory shows that cool and moist weather tends to increase the fixed oil content in the seed as well as the iodine values, on the other hand protein content tends to be lower (Siemens 2011).
Figure : Average crude protein content for mustard seeds in Canada
Between 95 and 98% of the oil in brassica seeds is composed of triglycerides, and only a small amount, in the range of 0.3 to 0.5% are free fatty acids (Appelqvist 1971), although the quantity may increase due to incorrect seed handling after harvest. The content of mono and diglycerides is usually low. The typical brassica juncea and sinapis alba varieties have a high erucic acid content (Table 2).
Nonsaponifiable material in mustard seeds is low and in the order of 0.5% of the oil. Mustard seeds also contain polar lipids apart from nonpolar triglycerides, mainly phospholipids and galactolipids which are comparable to soybean phospholipids.
Table : Fatty acid composition of Yellow and Brown mustard
Erucic acid (%)
Around 28-32% of the mustard seed total weight is composed of proteins. Proteins are polymers of amino acids. Proteins form the structural elements of cells and tissue in the human body and are considered as the basis of life, but they are also essential components in different food systems. Proteins are complex bio-molecules formed by amino acid aggregates and are essentially composed of carbon (50-55%), hydrogen (6-7%), oxygen (20-23%), nitrogen (12-29%) and sulfur (0.2-3%) but also may contain phosphorous, iron, magnesium and copper among other elements.
The building blocks of proteins are L-¡-amino acids, organic compounds containing a central carbon atom connected to a basic amino group (-NH2), an acid carboxyl group (-COOH) and one of the 20 possible organic substituents (R) as shown in Figure 4. These substituents differ in their physical and chemical properties and hence are the basis of the physicochemical differences in proteins such as polarity, acidity, basicity, conformational flexibility, reactivity and functionality. Amino acids can be classified according to the chemical characteristics of the substituent chain in: nonpolar, polar uncharged, polar positively charged and polar negatively charged (Ludescher 1996).
Regardless of the side chain, amino acids are zwitterions at neutral pH, which means that are molecules with both a positive and a negative electrical charge (Figure 4).
Amino acids can polymerize through the formation of a peptide bond into polypeptides, which are the basic constituents of proteins.
Figure : Amino acid structure
The peptide bond is a kind of covalent bonding between the amine group of an amino acid and the carboxyl group of another, producing an amide and a water molecule. The electronic structure of the peptide bond gives proteins and peptides their conformational properties (Ludescher 1996).
The shape and functionality of proteins are determined by their secondary, tertiary and quaternary structures, while their composition or sequence of amino acids along their backbone determines their primary structure. The secondary structure of a protein is the spatial configuration of the amino acid sequence. Secondary structures can be periodic, where there is a repetition in the values of the dihedral angles generating a helix, such as ¡-helices and ¢-sheets structure types; and aperiodic, where there is no repetition of the dihedral angles, like in ¢-turns structures (Ludescher 1996). In proteins, the secondary structure is defined by non-covalent interactions and patterns of hydrogen bonds between the backbone amide and carboxyl groups. The tertiary structure describes the atomic coordinates of each atom in a protein molecule. It is the folded and complete tri-dimensional structure of the polypeptide chain and is consequence of all non-covalent interactions between the amino acids in the molecule and between the molecule and the solution. Quaternary structure is the result of the association through weak non-covalent bonds of several polypeptide chains with a tertiary structure to form a larger protein complex. Each of the polypeptide chains is a subunit, and the quaternary structure is their assembling arrangement; it is the consequence of the non-covalent interactions between the subunits in the molecule and between the molecule and the solution (Ludescher 1996).
Most of the 20 amino acids can be synthetized by the human body, but there are nine essential amino acids that cannot be made by the organism and must be supplied in the diet. Amino acid requirement values for essential amino acids, as well as the amino acid composition of yellow mustard protein are shown in Table 3. The amino acid composition is well-balanced and comparable to other vegetal protein sources such as soybeans. It has been proposed that mustard proteins, along with other brassica proteins have a lower digestibility value than casein, due their structural rigidity and lower nitrogen release in early digestion phases (Wanasundara 2011), particularly napin proteins show more resistance to degradation.
Table : Values of oriental and yellow mustard amino acid composition compared to FAO indispensable amino acid requirements
Indispensable amino acid
requirements (WHO/FAO 2007)
Soybeans (Rackis, et al. 1961)
(VanEtten, et al. 1967)
(Cserhalmi, et al. 2001)
(Sarwar, et al. 1981)
Composition (mg/g protein)
* Value for phenylalanine + tyrosine
Of the proteins in mustard seed, around 70% is composed of storage proteins, cruciferin and napin, which are found inside the protein bodies and have no catalytic functions. Up to 10% is considered to be oleosin, a main structural component of the membrane surrounding the oil bodies (Bell, Rakow and Downey 1999), the rest of the protein in the seed is part of other cellular organelles while some of have catalytic functions, such as myrosinase (Appelqvist 1971). There are two main types of storage proteins present in mustard seeds: legumin type globulins (11S, cruciferins), and napin-type proteins (2S, napins), which are water soluble and have an isoelectric point around a pH value of 7 (Wanasundara 2011). Proteins are found in special organelles called protein bodies, which are generated by the storage protein vacuoles inside the seed. The relative content of cruciferin and napin proteins in mustard is variable and depends on the seed variety.
Although allergic reactions to 2S napins in mustard seed have been reported, including celiac disease and asthma (Monsalve, Villalba and Rodriguez 2001), the incidence of mustard allergies in animals seems to be low, since canola and mustard meals have long been used as a proteinaceous feed. Because of the occurrence of 2S napins in mustard seed and others from the brassica family, the European Union has listed mustard as an allergenic food ingredient (EU Directive 2003/89/EC).
Mustard, as well as other brassica oilseeds can be considered an important source of protein, but is most currently used for livestock feeding due to its content of anti-nutritional components and due to protein denaturation during industrial oil extraction, limiting its uses in the food industry. The utilization of friendly processing conditions and proper separation processes such as the ones used in this study would open the opportunity for value increase of mustard seeds. These techniques will be discussed in the following sections.
2.2.3 Protein-oil interactions in mustard seeds
Different binding forces are present in lipid protein interactions, such as covalent binding, electrostatic binding, polarization interaction, dispersion interaction and hydrophobic binding. Evidence has shown that electrostatic and hydrophobic binding and metal ion participation are particularly important in lipid protein structures (Chapman 1969).
The main lipid-protein interaction within mustard seeds occurs in cellular organelles. Mustard seeds, like most oil bearing seeds, store oil reserves in oil bodies. Oil bodies have a spherical shape, with a diameter that ranges from 0.2 to 2.5 m (Huang 1992) depending on the seed species and consist in a triacylglycerol core surrounded by a phospholipid monolayer and an outer surface layer composed of proteins (Figure 5). The average size is also affected by nutritional and environmental factors. The main component, triacylglycerols comprise about 92 - 98% of the total organelle weight. Phospholipids represent 0.6 - 4% and proteins around 0.6 - 3% (Gitte, Mundy and Jason 2001). The phospholipid monolayer in oil bodies is composed of phosphatidylcholine, and lesser quantities of phosphatidylserine, phosphatidylethanolamine, and phosphatidylinositol are also present (Huang 1992). The outer layer of oil bodies is formed by a special type of proteins called oleosins.
Oleosins are alkaline proteins with a molecular weight varying from 15 to 30 kDa. Recent studies have found that these proteins are not only present in the oil bodies, and as much as 5% can be found on endoplasmatic reticulum segments inside the cells (Gitte, Mundy and Jason 2001). Oleosin structure consists of three different regions according to its amino acid sequence: A hydrophilic N-terminal portion which contains between 50 - 70 amino acid residues, a central portion which is a hydrophobic chain made of around 70 amino acid residues and a C-terminal
Figure . Structure of oil bodies (Huang 1992)
amphipatic portion of variable length that interacts with the phospholipid layer, with the positively charged residues facing the phospholipid monolayer and the negatively charged residues facing the oil body surface (Hsieh and Huang 2004). It has been proposed that the center of the hydrophobic portion is formed by two antiparallel ¢-strands connected by three proline and one serine residues, interacting to form a "proline knot" that is inserted into the triacylglycerol matrix (Hsieh and Huang 2004).
Due to the presence of the oleosin and phospholipid monolayer, oil bodies present a negative electrical charge at neutral pH and a hydrophilic surface, preventing coalescence with one another and are able to retain their shape even through seed desiccation. The main function of these discrete and small organelles is to provide a large surface area per triacylglycerol unit in order to enable lipase binding during seed germination (Hsieh and Huang 2004). Oil body size in oilseeds is related to the particular seed species and is also determined by the relationship between oil and oleosin contents. As the triacylglycerol content in the oil bodies increase, the phospholipid and protein content decreases and the diameter of the oil bodies grows larger. It has been found that oil bodies in mustard seeds have an average diameter of 0.73 m and a composition of around 95% lipids, 3% protein and 1.5% phospholipids (Tzen, et al. 1993).
The presence of oil bodies in mustard seeds may play an important role in the efficiency of the extraction process, particularly in an aqueous extraction process. The extent of the disruption of the cell oil bodies prior to extraction has a direct impact in oil yields as they may remain intact after flaking or grinding, although coalescence can be induced by the use of enzymes (Campbell, Glatz and Johnson, et al. 2011).
Glucosinolates are considered anti-nutritional compounds and their presence is important for the food applications of brassica seed meals and derived products. In vivo models in rats show that high levels of glucosinolates and their breakdown products have an adverse thyrotoxic effect, but are not seen when protein isolates with low glucosinolate levels are used (Wanasundara 2011). They are responsible for the bitter taste of mustard, and their breakdown products, isothyocianates, for the pungency and hot flavour. Glucosinolates in brassica seeds are digested by the endogenous enzyme myrosinase to isothiocyanates, glucose and sulfates. The glucosinolate content in brown/oriental mustard is about 5-7% (Mustakas, et al. 1965) and in yellow mustard around 9% (Josefsson 1970). Glucosinolates are thioglucosides with a cyano and a sulfate group (Zrybko, Fuduka and Rosen 1997). There is a considerable variation in the glucosinolate content of mustard seeds due to factors such as genetic origin, age, and environmental conditions in which the plant is grown (Fenwick, Heaney and Mullin 1982). The predominant thioglucoside in yellow mustard (sinapis alba) is sinalbin and its reaction with myrosinase is shown in Figure 6.
Sinapine acid sulfate
Figure : Isothyocianate release reaction for sinapis alba
Similarly, in the brown/oriental mustard (brassica juncea), the main thioglucoside sinigrin reacts in the presence of myrosinase to produce allyl isothyocianate (Figure 7), which is a volatile pungent liquid and gives brown/black mustard its pungent flavour and odour. The main function of these substances in the plant is self defense mechanisms against pests and other diseases (Zrybko, Fuduka and Rosen 1997). Several studies have found that isothyocianates can inhibit the neoplastic effects of different carcinogens in different organs (Stoewsand 1995 and Spitz, et al. 2000). On the other hand, isothyocianates have also been shown to have goitrogenic properties, interfering with iodine uptake and affecting the function of the thyroid glands, inhibiting hormone production (Zukalová and Vasák 2002). Heat treatment for the inactivation
Figure : Isothyocianate release reaction for brassica junceaMyrosinase
of myrosinase has been shown to be an effective method to avoid the breakdown of glucosinolates from brassica seeds (Fenwick and Heaney 1983) but has an adverse effect due to protein denaturation during the thermal process and glucosinolates may undergo an enzyme mediated reaction to produce isothyocianates after ingestion. Alternatively membrane processing has also been shown effective for the reduction of glucosinolates from mustard protein isolates (Lui 1998).
Phytates, salts of calcium, magnesium and potassium from phytic acid (Figure 8) are other of the components in mustard seeds. About 3% of the yellow mustard seed is composed by phytates, on an oil free basis (Luo 1998). These compounds accumulate in the protein storage vacuoles as crystals and show strong electrostatic interactions with proteins, particularly at pH values lower than their isoelectric point, above which both dissociate. Phytic acid is capable of forming insoluble protein complexes and attention should be kept in the pH extraction values of the protein (Okubo, Myers and Iacobucci 1976). Because of the nature of phytic acid, there have been a series of studies that show contrasting consequences of phytate ingestion. While beneficial effects related to its natural antioxidant activity have been reported, suppressing iron-mediated oxidation reaction in the colon (Graf and Empson 1987), phytic acid is a strong chelating agent and can decrease the bioavailability of minerals such as calcium, zinc and iron and lead to mineral deficiencies in mammals. Studies have shown that rats fed with yellow mustard protein concentrate show symptoms of zinc deficiency (Wanasundara 2011).
Alkaline extraction of grounded yellow mustard seed, followed by ultrafiltration and diafiltration of the protein extract has been considered an effective method in the reduction of phytic acid levels in protein isolates, where the excess of basic cations prevents the formation of protein-phytate complexes and free phytates are effectively removed by membrane processing (Luo 1998).
Figure : Chemical structure of phytic acid
2.2.6 Phenolic compounds
There is a wide variety of phenolic compounds in mustard seeds which includes esterified and free forms of phenolic acids. These compounds are usually found as methoxylated derivatives of benzoic and cinnamic acids. The most abundant phenolic compounds present in yellow mustard are p-hydroxybenzoic acid and sinapic acid (Figure 9), present also as sinapine, its choline ester form (Kozlowska, Zadernowski and Sosulski 1983). Phenolic compounds are known to have a strong antioxidant effect, but are also responsible for a bitter and astringent taste in the mustard seed meal as well as a dark colour (Shahidi and Naczk 1989), both of them un-wanted characteristics in a food additive or a food ingredient. Four types of interactions exist between these compounds and proteins: hydrogen bonding, covalent bonding, ionic bonding and hydrophobic interactions (Xu and Diosady 2002). It has been shown that alkaline extraction, followed by treatment with 0.05 M sodium chloride and membrane processing can reduce the unbound phenolic fraction and the ionic protein-bonded fraction, while treatment with sodium lauryl sulphate is able to reduce the hydrophobic protein-bonded fraction (Xu and Diosady 2002).
Figure : Chemical structure of phenolic compounds
2.3 Protein extraction
Two main problems arise when considering mustard seed and other oilseeds for the production of food grade protein isolates; current oil extraction methods increase protein denaturation by the use of organic solvents and high temperatures (Pedroche, et al. 2004), and the presence of anti-nutritional components such as phytates, glucosinolates and phenolic compounds (Naczk, et al. 1998).
Protein denaturation is a physical-chemical process in which the configuration, conformation and state of folding of the polypeptide chains within the molecule is changed to a different arrangement by an energy input that can consist in heat, light, pressure, etc. Depending on the type of protein, denaturation can hinder or induce desirable functional properties. Proteins can be denatured by different types of processes such as thermal effects, presence and concentration of a denaturant like urea, guanidine hydrochloride and various salts that induce conformational changes of proteins (Kilara and Harwalkar 1996), high pressures related to extrusion processes and changes in pH that can lead to an unstable protein molecule. The effects of any of these factors depend on the nature of the protein; not all will suffer denaturation at the same conditions of temperature, pH, pressure or salt ion concentration.
Currently, most of the oilseed processing plants are focused in the production of edible oil and little attention has been given to the production of food grade protein from the meal fraction. But the need for additional sources of high quality protein for human nutrition has pushed forward the development of alternative processes, such as aqueous extraction systems.
2.3.1 Solvent extraction process
In the traditional solvent extraction process, the time-temperature-moisture relationship is essential (Becker 1970). As the value of each of these variable increases, the protein denaturation will also increase, affecting the quality and functional properties of the final product. The use of organic solvents such as hexane, derived from a non-renewable source, has inherent safety risks to both the manufacturing facilities and personnel due to flammability and explosion hazards. In addition hexane vapors can react with nitrogen oxides in the atmosphere and increase ground level ozone (Campbell et al. 2011). The Environment Protection Agency (EPA) in the United States has classified hexane as a hazardous air pollutant so its emission to the atmosphere has to be monitored and reported (Environmental Protection Agency 2001) and is subject to costly fines if the limits are exceeded.
In the typical solvent extraction process, seeds are first cleaned by aeration and sieving (Becker 1970). After cleaning seeds are submitted to hull decortication followed by the separation of the kernels, although in the case of Canola seeds de-hulling is not performed. Size reduction is usually the next step, the seeds are cracked using a rolling mill which helps disrupt the cellular structure and increases the surface area to improve oil extraction yield. After de-hulling and size reduction, oilseeds are tempered or cooked. Usual cooking temperatures vary from 120°C for rapeseed, 100°C for canola to 65°C for soybeans (Dunford 2012). Cooking inactivates the myrosinase enzyme which prevents the hydrolysis of glucosinolates into isothyocianates and nitriles in brassica seeds. Tempering also improves pressing and solvent extraction efficiencies (Dunford 2012); it is also useful to decrease the oil viscosity prior pressing and to complete the cell disruption and facilitate the oil extraction (Ward 1984). A prepress-solvent extraction process is usually the next step (Figure 10). Lower temperatures and pressures applied in the prepressing operation compared to hard pressing reduce protein denaturation and the resulting oil concentration in the meal is about 17% to 20% (Ward 1984), which can be subsequently removed by hexane extraction. A continuous percolation type extractor is commonly used for this task, which experiences a hexane loss in the order of 1.9 to 5.7 liters per tonne of seed processed. The final meal typically contains between 0.5% and 1.0% residual oil (Lusas 1983).
Cleaning / De-hulling
Desolventizing / Toasting
Evaporation / Distillation
Solvent extracted meal
Figure : Main operations in a solvent extraction system (Becker 1970)
After hexane extraction the resulting meal has a hexane concentration around 30% (Becker 1970) and a desolventizing process is required to reduce the residual solvent to acceptable levels. In a desolventizer toaster system, the defatted meal moves through a series of trays where it is heated in order to remove most of the solvent, live steam is then injected to strip the remaining hexane, and finally the meal is toasted in the lower trays at a temperature of 107°C (Becker 1970) to reduce the moisture content of the product. It has been shown that under the same desolventizing conditions, factors such as the moisture content of the seeds prior crushing, de-hulling and solvent extraction times affect the residual hexane content (Wolff 1983).
2.3.2 Aqueous extraction process
The development of aqueous extraction processes from oilseeds to obtain both oil and protein date back to the 1950's. Chayen (1953) and Subrahmanyan (1959) considered the extraction of oil and protein with water as the main solvent in an analogous way to traditional extraction, where all or a part of the oil is first removed. Just a limited number of these methods have been fully developed to a commercial level. Further aqueous processes for the recovery of oil and protein were developed for a wide variety of oilseeds, like coconuts (Hagenmaier, Cater and Mattil 1972), sunflower seeds (R. D. Hagenmaier 1974), peanuts (Rhee, Cater and Mattil 1972), soybeans (Campbell and Glatz 2009) and rapeseed (Caviedes 1996) have also been studied.
Protein and oil can be simultaneously recovered in an aqueous system, where protein in the resulting aqueous and solid phases can be further processed and purified. The efficiency of the process greatly depends on the main operations involved: cell disruption, oil and protein extraction, centrifugation, de-emulsification (Campbell K. A., 2011; Cater, et. al 1974 and Rosenthal, 1996) and protein purification and isolation. A general process diagram is shown in Figure 11.
The conditions, methods and degree of cell disruption are fundamental in an aqueous extraction processes. Cells in the seeds to be extracted must be efficiently destroyed to increase the extraction yields of both oil and protein. Insufficient disruption may leave large quantities of oil and protein in the solid residue (Cater, et al. 1974), while excessive comminution might result in a highly stable oil and water emulsion due to the smaller oil droplets (Rosenthal, Pyle and Niranjan 1996) and an increase in the oil content of the aqueous phase. Moisture content, physical structure and chemical composition of the seed are important in deciding the disruption method, and wet or dry operations. Commonly used methods include flaking, extrusion, dry grinding and wet grinding.
Protein purification and isolation
De-emulsification and oil recovery
Figure : Main operations in aqueous extraction systems (Cater, et al. 1974)
The extraction operation consists in the agitation of a dispersion composed of the disrupted seed material and water; factors that influence the effectiveness and extent of the extraction are solid to water ratio, pH, temperature (Cater, et al. 1974), particle size, agitation degree, extraction time (Rosenthal, Pyle and Niranjan 1996), extraction stages and ionic strength. After the extraction, the dispersion is separated, usually by centrifugation, into a water in oil emulsion, a solid phase containing insoluble components such as fibers, protein and oil, and an aqueous phase with the soluble components of the seed. Studies in our food laboratory have found that for full fat yellow mustard flour, an optimum water to solid ratio of 4 to 1, pH of 12, ambient temperature, 30 minutes of extraction time and 3 stages yield the highest amount of protein and oil extraction (Prapakornwiriya 2002 and Balke 2006).
Filtration of the aqueous phase rich in soluble protein is an essential step for the recovery of protein concentrates and isolates with low levels of anti-nutritional components. Several methods have been developed in our food engineering laboratory group for the recovery of high quality products that include microfiltration, ultrafiltration and diafiltration.
The removal of water from the protein solution is the final step. The use of a freeze drying or spray drying systems may be considered depending on the scale of the production process.
The aqueous extraction processing of oilseeds has important advantages. High quality protein can be obtained, since heating and toasting steps that can irreversibly cause denaturation are omitted. Safety risks regarding the use of highly volatile solvents are eliminated which have an important impact on equipment and training costs. There is also a considerable decrease in the environmental footprint of the process and costs related to volatile organic compounds emission and control. An aqueous extraction process has a smaller number of operations than the solvent extraction, making it a simpler, more energy efficient process and having the possibility of being designed for continuous or batch operation. Even though there are important advantages, there are also some disadvantages due to the nature of the process. There is a lower oil extraction yield compared to solvent treatment and there is the need of a de-emulsification step when oil is recovered in the form of an emulsion, additionally there is an increased potential for microbial contamination because the material is wet during most of the operations (Cater, et al. 1974 and Rosenthal, 1996).
The use of enzymes in aqueous extraction systems can increase both oil and protein yields. Depending on the seed and its components different kinds of enzymes or combination of enzymes can be used. Carbohydrases, such as cellulases, pectinases and hemicellulases help degrade the cell wall materials and can increase the oil recovery, while proteolitic enzymes hydrolyze proteins including oleosins, which may increase the release of oil (Rosenthal, Pyle and Niranjan 1996).
2.4 Protein purification and isolation
In order to obtain food quality products by either a solvent or an aqueous extraction, protein must be purified. Several protein purification and separation processes rely on the differences in solubility between them, or between proteins and non-protein materials in a solution. Precipitation is one of the techniques used for the recovery of proteins, and is usually accompanied by a concentration step in order to reduce the volume of the initial solution and the level of undesired, micro-molecular components. The principles of protein precipitation are related to forces acting between the polypeptide chains in the proteins and also their interaction with the solvent molecules. Changes in the solvent-protein and protein-protein interactions which lead to precipitation can be induced by modifying the temperature, the composition of the solvating medium or the pH (Li-Chan 1996). At the isoelectric point, where there are an equal number of positive and negative charged groups, the surface of the protein will be least solvated facilitating hydrophobic interactions and aggregation.
One of the most common processes for protein precipitation is known as "salting out", where a high salt concentration leads to a decrease in the effective concentration of water. The concentration and nature of the salt used is important to determine the effect on protein-protein and protein-water interactions. In general terms, salts with high molal surface tension values are effective in protein precipitation, while salts with low values have the opposite effect, called "salting in" (Li-Chan 1996). An alternate process for protein precipitation proposed by Murray, et al. (1979) called micellization consists of the extraction of proteins from seed meals using a "salting in" technique followed by precipitation by the dilution of the concentrated extract with water and a temperature adjustment, favoring hydrophobic interactions and protein aggregation.
In some cases the solubility of proteins at their isoelectric point is low enough to allow their recovery by a pH adjustment, this process is known as isoelectric precipitation. Previous studies in our food engineering laboratory have shown that isoelectric precipitation is a suitable process to recover most of the mustard seed proteins after an alkaline extraction. The isoelectric point for the alkaline extracted proteins from defatted mustard is around a pH value of 4.75 (Lui 1998 and Xu, Lui, et al. 2003) while a value of 5.5 has been found and used for alkaline extracted full fat mustard (Prapakornwiriya 2002). It must be considered that for the mustard protein extraction, isoelectric precipitation of the protein extract would result in a product with high levels of anti-nutritional components that would limit its use for human consumption. Since the molecular weight of mustard proteins is considerably larger than most anti-nutritional components or contaminants, membrane processing via ultrafiltration and diafiltration is used as a purification step.
Ultrafiltration is a cross-flow membrane separation process. In a solution containing low molecular weight and high molecular weight solutes, the latter will be retained by the membrane, while the smaller low molecular weight particles will permeate through. The driving force in order to achieve the separation is a pressure difference applied to a solution on the feed side of a membrane. Ultrafiltration membrane pore sizes are usually classified according to the molecular weight of the species that will be retained by assigning to them a molecular weight cut off (MWCO). A schematic of this process is shown in Figure 12. The solvent and low molecular weight species passes through the membrane and constitute the permeate, while solutes with a larger weight than the MWCO are retained and form the retentate.
Figure : Ultrafiltration principle of operation
Since micro molecular components have significantly lower molecular weights, it is possible to separate them from other macromolecular compounds in aqueous solution by using ultrafiltration. Membrane molecular weight cut offs in this case are typically between 5 and 500 kDa and are able to retain proteins, polymers, and chelates of heavy metals (Cheryan 1998). Since low-molecular-weight solutes flow through the membrane, osmotic pressure is not an issue. However, since retained large molecules and colloidal particles have low diffusivities in the liquid medium, ultrafiltration membranes are more susceptible to fouling and concentration polarization than reverse osmosis or microfiltration membranes (Cheryan 1998).
Usually, not all the particles larger than the molecular weight cut off of the membrane are rejected, and some particles smaller than this parameter may be partially rejected. In order to estimate the separation degree attained by the process, a mathematical model has been developed for the rejection of the solutes (Cheryan 1998):
where R is the rejection coefficient, CP is the concentration in the permeate and CR is the concentration in the retentate. During this process, the total volume of a solution will be reduced as the solvent and low molecular weight components are being removed resulting in the concentration of the macromolecular species, whose quantity remains unchanged. The concentration and volume relationship in ultrafiltration systems are characterized by the following equation (Cheryan 1998):
Where Cf is the final concentration of the feed, C0 is the initial concentration of the feed, V0 is the initial feed volume, Vf is the final feed volume, CF is the concentration factor and R is the rejection coefficient.
Diafiltration is a method where permeable solutes are eliminated from a solution and consists in an initial volume reduction, usually performed by ultrafiltration and a subsequent addition of a suitable buffer solution or water. This process can be made in a continuous or discontinuous manner. In discontinuous diafiltration the adequate buffer solution or water quantity is added to the concentrated solution to reach the initial volume, and the ultrafiltration operation is repeated until the unwanted micro-molecular components are removed. In continuous diafiltration, buffer solution or water is added at the same rate as the permeate flux, keeping the concentrated solution volume constant during the process.
The amount of micro-molecular components that is removed is related to the volume of permeate resulting from the operation and the initial volume of retentate. This relationship is referred to as diafiltration volume (DV) (Cheryan 1998):
Where Vf is the permeate volume and V0 is the initial retentate volume. For continuous diafiltration the relationship between the initial and final concentration of the micro-molecular components is given by following equation (Cheryan 1998):
Where CR is the final concentration of the micro-molecular component, C0 is the initial concentration, and R is the rejection coefficient. As a result of continuous diafiltration, the final volume and concentration of the macro-molecular components retained by the membrane does not change. As shown by the equation, a diafiltration volume of 6 is enough to remove more than 99.5% of a micro-molecular component with a rejection coefficient of 0. The given formula also shows that when the solute is partially retained by the membrane (the rejection coefficient is greater than 0), the diafiltration volume needed to reach the same removal will increase.
The main limitations for membrane separation processes are concentration polarization and membrane fouling. Concentration polarization controls the performance of ultrafiltration. It is an effect where particles rejected by the membrane tend to form a layer near the surface causing further resistance to the flow of the permeate. The flux decrease is usually explained by two mechanisms: The first one is an increase in the osmotic pressure due to the increased solute concentration near the surface of the membrane in comparison to the bulk concentration in the feed, and the second one is the hydrodynamic resistance of the boundary layer (Cheryan 1998). To reduce the effect of concentration polarization several factors such as pressure, feed concentration, temperature and turbulence in the feed channel must be optimized.
Membrane fouling on the other hand is characterized by an irreversible decline in the flux that cannot be counteracted with fluid management techniques. It is due to the accumulation of feed components on the membrane surface or within the pores of the membrane and is influenced by the chemical natures of both the membrane and the solutes and membrane-solute and solute-solute interactions (Cheryan 1998). Usually the only way of restoring the flux of a fouled membrane is through cleaning. Fouled membranes and auxiliary equipment are generally cleaned by clean-in-place procedures (Lindau and Jönson 1994) which are usually based on various chemical or enzymatic treatments to restore the membrane to its original state.
2.5 Functional properties in protein isolates
The importance of protein isolates when used in food systems does not rely only in their nutritional value, but in the desirable properties and sensory attributes that the additives are able to provide. Emulsification capacity, water and lipid holding capacity, gelation capacity, foaming capacity and foaming stability are functional properties that enhance food sensory and organoleptic characteristics including colour, flavour, odour, texture or mouth feel. For centuries animal proteins have been traditionally the main source of functional ingredients; milk, egg and animal meat proteins have unique properties and functionality applications, however it is estimated that about 8 kg of protein from a vegetable source are needed to produce 1 kg of animal protein (Damodaran 1996). Considering this low ratio, the increasing protein demand due to population growth and land use competition between food crops, non-food crops for biofuel, and cattle, oilseeds such as mustard offer an interesting alternative as a renewable source of oil for biofuel production and protein to provide sensory characteristics and nutritional value to food products as a replacement of animal protein.
Kinsella and Melachouris (1976) defined the functional properties of proteins as those physical and chemical properties which have an influence on their behavior in diverse food systems, whether it is in their preparation, storage, cooking or consumption. The size, shape, amino acid composition and sequence, net charge, charge distribution, hydrophobicity, hydrophilicity, structural arrangements and molecular flexibility of proteins are intrinsic characteristics that define their functionality and interactions with other food ingredients.
Functional properties can be classified in three groups according to their action mechanism in food systems: properties due to hydration such as solubility and wettability, properties related to protein structure such as viscosity and gelation, and properties related to protein surface such as emulsifying and foaming capacities (Moure, et al. 2006 and Siong, et al. 2011).
2.5.1 Hydration properties
Important functional properties such as solubility, wettability, dispersibility, foaming, emulsification and gelling properties are affected by the solvation and dissolution characteristics of the protein and depend on the interaction between the molecules and the solvent. The hydration mechanism of a protein describes different states of water in hydrated proteins (Kinsella, Fox and Rockland 1986): structural water is formed by water molecules that are part of the protein structure, bound by hydrogen bonds; this water is not available for chemical reactions, is un-freezable and not relevant for the functional properties of the protein. Monolayer water is composed by water molecules bound via dipole-induced dipole, ion-dipole and dipole-dipole interactions with polar groups in the protein and hydrophobic hydration of nonpolar groups. The monolayer forms when the water activity is in the range from 0.05 to 0.3 and is unavailable for most chemical reactions. On the other hand, water states related to protein functionality include: multilayer water at water activities between 0.3 and 0.7, un-freezable water consisting of multilayer ordered water molecules up to a water activity of 0.9, capillary water bound due to capillary forces in crevices and cavities, which appears when the water activity is between 0.5 and 0.95 and finally, hydrodynamic hydration water that exists at a water activity over 0.99 and affects viscosity and diffusion properties of the protein.
22.214.171.124 Water and oil absorption
In food systems, the water absorption capacity of a protein is the ability to hold water against gravity and form network structures with other proteins via non-covalent interactions. The capacity of retaining moisture influences the texture and mouth-feel of foodstuffs (Kinsella and Melachouris 1976 and Johnson 1970) and is function of the fraction of charged residues, polar amino acid side chains and nonpolar residues of the protein (Moure, et al. 2006). Amino acid residues with charged side chains will experience strong ion-dipole interaction and bind more water. External factors like pH, ionic strength, protein concentration, temperature and particle size of protein powders have a considerable effect in water absorption (Damodaran 1996 and Johnson 1970). Most proteins have the lowest water binding capacity at their isoelectric pH. Water absorption is usually described by the water absorption capacity (WAC), the amount of water retained per unit mass of protein after mixing and centrifugation (F. Sosulski 1962) and the water hydration capacity (WHC) (Naczk, Diosady and Rubin 1985).
Similarly, oil absorption can be defined as the amount of oil retained per unit mass of protein after thorough mixing and centrifugation (Lin and Humbert 1974 and Sosulski, Humbert and Bui 1976). The importance of fat absorption by protein in food systems lies in the in the ability of lipid molecules to modify and in some cases provide odours and flavours as well as a pleasant mouth feel (Forss 1972), and an improvement in flavour transport during food processing (Kinsella and Melachouris 1976). The oil-protein binding mechanism is related to capillary forces in crevices and cavities of the protein molecule surface which are able to entrap oil molecules as well as hydrophobic interactions between non-polar side chains and lipid molecules.
126.96.36.199 Protein Solubility
Protein solubility can be described as a thermodynamic equilibrium between protein-solvent and protein-protein interactions. As protein solubility increases, it can be more easily incorporated into foodstuffs, increasing its functionality and applications. Solubility of a protein is influenced by the balance of hydrophobic and hydrophilic residues on the protein surface, given by the amino acid composition. A low number of hydrophobic residues, as well as a high number of electrostatic repulsions and ionic hydration lead to high solubility (Moure, et al. 2006). Solubility is also affected by the environmental conditions of the solution such as pH, ionic strength, ion types, temperature, solvent polarity and processing conditions, all of which interfere with the hydrophilic and hydrophobic interactions at the protein surface (Damodaran 1996). In the case of pH and ionic strength, their effect on solubility can be explained by the changes in the protein electrostatic forces (Kinsella and Melachouris 1976). At pH values around the isoelectric point, the solubility of a protein will be at its minimum value. Processing conditions that promote protein denaturation lead to conformational changes that even at a low extent can alter the hydrophobic and hydrophilic balance at the protein surface affecting solubility. The nitrogen solubility index (NSI) given as the percentage of water-soluble nitrogen from a given sample under slow stirring is the usually adopted method to determine protein solubility (AOCS, 1999).
2.5.2 Properties related to protein surface
Many processed foods are in foam or emulsion type systems. Emulsions and foams are two-phase systems consisting of a dispersed and a continuous phase. Foam can be defined as a substance formed by the dispersion gas cells in a continuous liquid phase that contains a surfactant, while an emulsion is a mixture of two or more liquids that are normally immiscible where one of the liquids is dispersed in the other. Because of the amphipathic nature of proteins, they act as macromolecular surfactants in emulsions and foam-type products. Proteins act by lowering the interfacial tensions and also are able to produce a continuous film at the interface via intermolecular interactions. The high viscosity and high dilatational modulus of protein films makes them able to withstand external forces, producing more stable foams and emulsions than low molecular weight surfactants. These characteristics are the result of protein surface activity, which is affected by molecular properties such as conformational stability, flexibility, the symmetry in the distribution of hydrophilic and hydrophobic side chains and external factors such as pH, ionic strength and temperature (Moure, et al. 2006). The dynamics of protein adsorption proceed through the sequential attachment of polypeptide segments. The first step is the transport of the protein from the bulk to the interface where the global free energy of the protein is lower. The kinetics of adsorption of proteins has been proposed to follow a diffusion controlled model that depends not only in the concentration and diffusion coefficient, but on an activation energy for adsorption at the interface that arise from physiochemical constraints of the protein related to hydrophobic, hydrophilic and conformational flexibility of the molecule (Damodaran 1996). The surfactant properties of proteins are improved when they possess a high rate of diffusion and adsorption, are able to unfold rapidly and are able to form a cohesive and viscous film at the interface. Protein adsorbs to interfaces in multiple contact points according to the degree of flexibility of the polypeptide chain and may change their conformation upon interface adsorption (Damodaran 1996).
188.8.131.52 Foaming properties
Foaming properties in proteins describe their ability to form a large and stable interfacial film between the solution and the surrounding air that will withstand internal and external forces. The foaming ability of a protein depends on its rate of adsorption at the interface, on its molecular flexibility, that is, the rates at which it can unfold and undergo molecular rearrangements to reduce the surface tension, and its capacity to form a cohesive film (Moure, et al. 2006). On the other hand, the stability of the foam is affected by the molecular rigidity of the protein and the rheological properties of protein films such as film viscosity, shear resistance, elasticity and the disjoining pressure between protein layers. Ultimately, an adequate balance of flexibility and rigidity must be present to produce stable foams (Damodaran 1996). External factors such as pH (Sathe, Deshpande and Salunkhe 1982), temperature (Richert, Morr and Cooney 1974) and the presence of other components such as sugars or lipids (Yasumatsu, et al. 1972) also affect the foaming properties and foam stability. The foam capacity can be determined by the measure of the foam volume produced after whipping of a protein dispersion with a specific concentration (Lin and Humbert 1974), while foam stability can be expressed as the volume of foam remaining after a certain amount of time has passed.
184.108.40.206 Emulsifying properties
An emulsion can be defined as a two phase system in which one liquid is dispersed as droplets in another. Thermodynamically speaking, an emulsion is an unstable system and given enough time the phases will separate, but it can be stabilized by the addition of surface active molecules or surfactants. The emulsifying potential of a protein can be described by the emulsifying activity index, the emulsion stability index and the emulsifying capacity (Kinsella and Melachouris 1976). Factors that affect the emulsifying properties of proteins include the rate of adsorption at the interface, the amount of protein adsorbed, the conformational rearrangement at the interface, molar mass, and external factors such as pH, ionic strength and temperature (Moure, et al. 2006). Disjoining forces generated by electrostatic, steric and solvation interactions in the aqueous phase also have a major role in the stability of emulsions (Damodaran 1996). Most emulsions are more stable at pH values that are far away from the isoelectric point of the protein, where electrostatic repulsion and hydration repulsion forces are maximized. Heat denaturation, as well as some chemical and enzymatic treatments, like succinylation, phosphorylation and glycosylation can improve the emulsifying properties (Damodaran 1996).
2.5.3 Properties related to protein structure: Gelation
Gels are considered as an intermediate phase between a solid and a liquid. In food systems the liquid phase is usually water and the solid phase is formed by proteins or carbohydrates. Protein gels are formed by polymeric molecules covalently or non-covalently cross-linked in a three-dimensional network. Gels provide a structural matrix able to hold water, flavours, sugars and other food ingredients. The ability to form a gel by a protein solution is affected by its molecular weight and its ability to denature (Moure, et al. 2006). The mechanism for protein gelation is a stepwise process (Damodaran 1996) in which the protein solution is first irreversibly converted to a pro-gel by heating above the denaturation temperature to expose the functional groups which interact to form the network, then the protein forms one of two types of gel networks depending on the type of protein, its amino acid composition and external factors such as pH and ionic strength: a coagulant gel is formed by proteins with high levels of nonpolar residues, and a transparent type gel is formed by proteins that contain hydrophilic amino acid residues. The minimum protein concentration necessary to form a self-supporting gel network is known as the least gelation concentration (LGC). Below this critical concentration, proteins unfolded by heat treatment undergo random aggregation which may lead to precipitation. Globular proteins usually have higher LGC values than fibrous proteins.
2.6 Lipid oxidation in lipid-protein systems
One of the disadvantages of an aqueous extraction process, as mentioned in section 2.3.2, is that proteins are recovered in a solution containing oil. Oil concentration depends on factors such as the cell disruption methods used prior extraction, the extraction conditions, the use of enzymes, and oil and protein contents of the starting material. Proteins form large molecular aggregates along with remaining oil bodies in the solution (Dendukuri and Diosady 2003) preventing oil from being permeated during membrane processing, and being unavoidably recovered in the protein isolates after isoelectric precipitation. As an oil containing ingredient, the isolates are prone to lipid oxidation and the subsequent interactions between the oxidation products and proteins.
Lipid oxidation is one of the most important processes for food deterioration as it causes the development of unpleasant odours, flavours and rancidity in both oils and oil rich foodstuffs. Lipid oxidation reactions may also decrease the nutritional value of the food (Pokorný, Kolakowska and Bienkiewicz 2005). Toxic substances can also be generated, which can be associated with health risks to consumers (Tazi, et al. 2009 and St. Angelo and Ory 1975).
Lipid oxidation in food systems is a complex process, since oxidation products may react with other components in the food system such as proteins, carbohydrates, water and vitamins. The result of lipid oxidation under these conditions has different effects in the functional properties, texture, mouth feel, aroma, nutritional value, colour and safety of food and food ingredients (Hidalgo, Zamora and Alaiz 1991). The degree of unsaturation of fatty acids, the presence of antioxidant substances, traces metals, light, temperature and oxygen availability are the main factors that affect lipid oxidation.
2.6.1 Lipid and protein oxidation mechanisms
There are four different pathways for the oxidation and formation of hydroperoxides in lipids: photo-oxidation, enzymatic oxidation, irradiation and autoxidation, the latter being the most important (Matthäus 2010). The autoxidation mechanism in unsaturated lipids begins with the reaction of a fatty acid radical with oxygen (Figure 13). The fatty acid radical is formed by hydrogen abstraction from an allylic carbon, which has a low dissociation energy. It is believed that heat, metal catalysis or ultraviolet irradiation provide the driving force for this de-protonation (Matthäus 2010). The fatty acid radical is unstable and reacts with atmospheric oxygen to produce a peroxy radical, which forms a new fatty acid radical and starts an exponential chain reaction. Bond strength vary between fatty acids, highly unsaturated fatty acids are subjected to a faster autoxidation due to the weakness of the allylic carbon-hydrogen bonds. Hydroperoxides themselves are odourless and tasteless compounds, but are unstable and react into secondary products that can be easily detected by their aroma and taste, some of them even at very low concentrations (Reindl and Stan 1982). The types of these compounds produced depend on the fatty acid composition and other components in a food system. The main pathway of hydroperoxide decomposition to volatile compounds is the ¢-scission of a carbon-carbon bond to produce oxo-compounds and an alkyl or alkenyl radical (Hidalgo, Zamora and Alaiz 1991). Secondary oxidation products include ketones, aldehydes, alcohols, hydrocarbons, acids and epoxides.
Figure : Fatty acid radical chain oxidation mechanismRH
R· + O2
R· + H·
RO2· + RH
R· + O2
RO2H + R·
R· + RO2·
Deterioration of proteins in dry food products is similar to lipid oxidation, and is promoted by the increase in water activity. Proteins are oxidized by an initial electron abstraction, by the transfer of a hydrogen atom by plant phenols or by free radical scavenging, trapping radicals by