FROM POTATO

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority from U.S. provisional patent application Serial Number 62/006413 filed on June 2, 2014 and herewith incorporated in its entirety.

TECHNOLOGICAL FIELD

This application relates to process for obtaining protein-enriched preparations from potato pulp using enzymatic degradation as well as corresponding preparations. The protein- enriched preparations comprise patatin and/or proteinase inhibitors.

BACKGROUND

Globally potatoes are one of the staple crops, used for human consumption, industrial processing, and/or agricultural regimes. Industrial processing such as the manufacturing of starch generates low value by-products known as potato fruit juice and potato pulp which contain a crude protein content of 50% and 74%, respectively. These proteins have been shown to be nutritionally comparable to egg proteins.

The potato starch industry releases large quantities of by-product, known as a potato fruit juice (PFJ), which is costly to dispose of due to its high polluting effect and provides only marginal economic value when used as animal feed and fertilizer. The phytochemical composition of PFJ is interesting as it is rich in proteins, minerals and free amino acids. Converting this by-product into high value-added ingredients would have an important economical and environmental impact. In this context, the extraction of proteins from PFJ is of particular interest as manufacturing starch from one thousand kg of potatoes releases 5- 12 m ³ of PFJ, which contains 30 to 41 % protein of total solids.

Compared to other proteins from other vegetable and cereal sources, potato proteins are considered higher quality as they contain a high proportion of lysine, which is often lacking in such crops. Potato proteins are commonly divided into three fractions patatin (up to 40%), protease inhibitors (-50%), and other high molecular weight proteins (-10%). The patatin fraction is a dimer glycoprotein with a molecular weight of 40 to 45 kDa and is present as many isoforms. In terms of beneficial properties, patatin has been shown to possess antioxidant ability and lipid acyl hydrolase. In addition, functionally, patatin has excellent foaming and emulsifying abilities. The protease inhibitors, with a molecular weight ranging from 5 to 25 kDa, have been shown to have beneficial properties such as anti-carcinogenic, anti-microbial and a high satiety property by releasing the hunger suppressant cholecystokinin. Functionally, the protease inhibitor fraction is soluble throughout a wide pH range, whereas patatin shows minimum solubility at pH 4.

Isolating proteins from PFJ without affecting their functional properties is challenging because of the aqueous nature of PFJ and its complex composition. Industrially the isolation of proteins from PFJ involves a combination of thermal coagulation and acidic precipitation. Although thermal/acidic precipitation results in a high yield of protein recovery, it often leads to complete loss of the protein functionality, which limits their application to animal feed. Other extraction techniques have been explored for the recovery of functional proteins, including salt, acid, ammonium sulphate ((NH ₄ ) ₂ S0 ₄ ) precipitations, carboxymethyl cellulose (CMC) complexation and chromatographic techniques. However, conflicting results have been reported on the efficiency of these techniques to isolate potato proteins from PFJ. These differences could be due to the use of varying potato cultivars and PFJ preparation methods. The relationship between the extracting agents and the functional properties of isolated proteins has been overlooked. Only limited studies have evaluated these extraction techniques in terms of recovery yield, purification factor, protein profile, denaturation degree and functional properties.

It would be highly desirable to be provided with a process for obtaining protein-enriched preparations which would limit the denaturation of the proteins. The process would preferably avoid submitting the treated pulp to a heating step to prevent unfolding of patatin and of the proteinase inhibitors. Such process could generate preparations having increased technical/functional properties and biological activity.

BRIEF SUMMARY

The present disclosure provides a process for making protein-enriched preparations from potato or a fraction thereof. The process uses an enzymatic treatment with galactanase and pectinase (optionally in combination with arabinanase and/or proteinase) of potato/potato fraction to retrieve proteins in a substantially native configuration. The process can be used to add value to potato residues which are not suitable for consumption.

The present disclosure also provides the protein-enriched preparations (obtained from this process) which comprises patatin and/or proteinase inhibitors. The protein-enriched preparations can be used to provide patatin and/or proteinase inhibitors in their native configuration. Such protein-enriched preparations are especially suited for pharmaceutical, nutraceutical as well as food applications. According to a first aspect, the present disclosure provides a process for making a protein- enriched preparation from a potato. Broadly, the process comprises (i) providing the potato or a potato fraction having an initial starch content; (ii) contacting the potato or the potato fraction with a first enzyme or a first enzyme mixture to reduce by at least 75% (w/w) the initial starch content of the potato or the potato fraction to provide a de-starched potato mixture; (iii) contacting the de-starched potato mixture with a second enzyme mixture comprising at least one galactanase and at least one pectinase to release at least a portion of the proteins from the de-starched potato mixture to provide a treated mixture, wherein the treated mixture comprises a protein fraction and a carbohydrate oligomer fraction; and (iv) removing the carbohydrate oligomer fraction from the protein fraction from the treated mixture to obtain a protein-enriched preparation, wherein the protein-enriched preparation comprises patatin and proteinase inhibitors. In an embodiment, the potato fraction comprises potato pulp and/or potato juice. In another example, the first enzyme or the first enzyme mixture comprises an amylase and/or an amylopectinase. In still another embodiment, the first enzyme is an amylase such as an a-amylase which can be obtained from Bacillus lichenformis. In still another embodiment, the proportion of patatin (w/w) with respect to the total proteins in the protein-enriched preparation is substantially similar to the proportion of the proteinase inhibitors (w/w) with respect to the total proteins in the protein-enriched preparation. In such embodiment, the at least one galactanase (which can be an endo-p-1 ,4- galactanase which can be obtained from Aspergillus η^ ^' βή can be provided in a purified form. Still in such embodiment, the at least one pectinase (which can be a polygalacturonase which can be obtained from Aspergillus niger) is provided in a purified form. Still in such embodiment, the second enzyme mixture can be Gamanase®, Viscozyme® or Pectinase®. In another embodiment, the proportion of patatin (w/w) with respect to the total proteins in the protein-enriched preparation is substantially dissimilar to the proportion of the proteinase inhibitors (w/w) with respect to the total proteins in the protein-enriched preparation. In such embodiment, the second enzyme mixture in step (ii) further comprises at least one arabinanase and/or at least one proteinase. For example, the second enzyme mixture can be Ceremix®, Hemicellulase®, Newlase®, Diazyme®, or Laminex® and the proportion of patatin in the protein-enriched preparation is lower than the proportion of the proteinase inhibitors. In still a further embodiment, the first enzyme is α-amylase and the second enzyme mixture is Depol®, and the proportion of patatin in the protein-enriched preparation is higher than the proportion of the proteinase inhibitors. In another example, the second enzyme mixture can be logen® and the proportion of patatin in the protein-enriched preparation is higher than the proportion of the proteinase inhibitors. In another embodiment, step (iv) further comprises filtering the treated mixture to remove the carbohydrate oligomer fraction from the protein fraction. In still another embodiment, the treated mixture further comprises a solid fraction and step (iv) further comprises removing the solid fraction from the treated mixture by centrifugation. In still another embodiment, the process further comprises, after step (iv), isolating patatin from the proteinase inhibitors in the protein-enriched preparation to obtain a patatin-enriched preparation and/or isolating proteinase inhibitors from patatin in the protein-enriched preparation to obtain a protein inhibitors-enriched preparation. In still another embodiment, the process further comprises formulating the protein-enriched preparation as a food additive, a nutraceutical composition and/or as a pharmaceutical composition. In yet another embodiment, the process further comprises glycating the protein-enriched preparation.

In a second aspect, the present disclosure provides a protein-enriched preparation obtained by the process described herein. The present disclosure provides a pharmaceutically composition comprising the protein-enriched preparation described herein and a pharmaceutically acceptable excipient. The present disclosure also provides an emulsifying or foaming agent comprising the protein-enriched preparation describes herein as well as a food additive comprising the protein-enriched preparation described herein.

In a third aspect, the present disclosure provides a patatin-enriched preparation obtained by the process described herein. The present disclosure provides a pharmaceutically composition comprising the patatin-enriched preparation described herein and a pharmaceutically acceptable excipient. The present disclosure also provides an emulsifying or foaming agent comprising the patatin-enriched preparation describes herein as well as a food additive comprising the patatin-enriched preparation described herein.

In a fourth aspect, the present disclosure provides a proteinase inhibitors-enriched preparation obtained by the process described herein. The present disclosure provides a pharmaceutically composition comprising the proteinase inhibitors-enriched preparation described herein and a pharmaceutically acceptable excipient. The present disclosure also provides a foaming agent comprising the proteinase inhibitors-enriched preparation describes herein as well as a food additive comprising the proteinase inhibitors-enriched preparation described herein.

BRIEF DESCRIPTION OF THE DRAWINGS

Having thus generally described the nature of the invention, reference will now be made to the accompanying drawings, showing by way of illustration, a preferred embodiment thereof, and in which:

Figure 1 illustrates an embodiment of the process of the present disclosure. Figure 2 illustrates the effect of starch removal from potato pulp on protein recovery, using two a-amylases: Bacillus sp. (dark grey bars) and Bacillus licheniformis (light grey bars) on the recovery of proteins upon the enzymatic treatment with polygalacturonanase 1 (16.66 U/g pulp) and endo-1 ,4-p-D galactanase (16.66 U/ g pulp). Results are shown as the of protein recovery (in percentage) in function of units of a-amylase (U7 g pulp).

Figures 3A-B illustrate the effect of the process used on the structural stability of different protein preparations and the sigmoidal transition curve of relative fluorescence intensity per milligram protein. (A) Structural stability of the protein isolate preparations obtained upon combined thermal/acidic treatment (e.g., combination, light grey bars) and protein preparations recovered upon enzymatic treatment (e.g., enzymatic, dark grey bars). Results are provided as fluorescence intensity per mg of protein in function of temperature (in °C). Sigmoidal transition curves of protein isolate preparations (e.g., combination, ♦, light grey line) and protein preparations (enzymatic,■, black line) obtained from processes conducted at pH 4 (B) or at pH 9 (C). Results are provided as fluorescence intensity per mg of protein in function of temperature (in °C).

Figures 4A-E show response surface plots on the protein recovery yield, gram of patatin extracted, gram protease inhibitors extracted as affected by temperature (°C), time (hrs), pulp concentration (mg/mL), units of polygalacturonase (U), units of endo- -1 ,4-galactanase: yield response: interaction of units of polygalacturonase and temperature (A), gram of patatin/gram of pulp response: interaction of units of polygalacturonase and temperature (B), gram of patatin/gram of pulp response: interaction of pulp concentration and time (C), gram of protease inhibitors/gram of pulp response: interaction of pulp concentration and temperature (D), gram of protease inhibitors/gram of pulp response: interaction units of endo- β-1 ,4-galactanase and temperature (E).

Figures 5A-B illustrate the effects of using different enzyme mixtures on the protein yield and proportion of the protein preparations obtained. (A) Protein yield (provided in weight percentage) is presented in function of enzymatic mixture used (Gamanase®, Depol®, Ceremix®, Hemicellulase®, logen®, Viscozyme®, Pectinase®, Newlase®, Diazyme® or Laminex®) and of incubation time (10 hours = light gray bars, 20 hours = dark grey bars). (B) Patatin (light gray bars) and protease inhibitors (black bars) yield (provided in weight percentage) is presented in function of enzymatic mixture used (Gamanase®, Depol®, Ceremix®, Hemicellulase®, logen®, Viscozyme®, Pectinase®, Newlase®, Diazyme® or Laminex®).

Figure 6 illustrates the amino acid sequence of one of patatin's isoforms. Figure 7 illustrates a flow diagram of a pilot-scale potato protein isolation of Example IV using three methods: ultrafiltration (PPC UF), 60% ammonium sulphate saturation (PPI AS), and commercial multi-enzymatic system Depol 670L (PPC Enz).

Figure 8 illustrates the size exclusion chromatography (Optical density (OD ₂ eo _nm ) in function of retention time (min)) of potato protein isolates extracted in pilot scale using ultrafiltration (PPC UF), 60% ammonium sulphate saturation (PPI AS, and multi-enzymatic product Depol 670L (PPC Enz). Std- ixture (Standard Mixture) = Thyoglobulin-670 kDa, y-Globulin-158 kDa, Ovalbumin-44 kDa, Myoglobulin-17 kDa, and Vitamin B12-1 .35 kDa.

Figure 9 illustrates differential scanning calorimetry Tthermograms (heat flow (W/g) in function of temperature (°C)) for potato proteins extracted by 60% Ammonium Sulphate Saturation (PPI AS) and Multi-Enzymatic Product Depol 670L (PPC Enz), at pH 7.

Figures 10A-D illustrate Fourier Transform Infrared Spectroscopy spectra of potato protein isolate and concentrate extracted on pilot plant scale using (A, C) 60% Ammonium sulphate saturation or (B, D) multi-enzymatic system Depol® 670L. Thermal denaturation curves are shown in A and B. Effect of extracting agent on the secondary structural changes as obtained by Fourier-transform Infrared Spectroscopy spectra are shown in C and D, intermolecular beta-sheet (·) and aggregation (Δ).

DETAILED DESCRIPTION

The present disclosure provides a process for obtaining protein-enriched preparations from potatoes as well as such preparations comprising patatin and/or proteinase inhibitors. The process broadly comprises enzymatically treated de-starched potato-derived material with at least one galactanase and at least one pectinase (optionally in combination with at least one arabinanase and/or at least one proteinase) to obtain such protein-enriched preparations. The present disclosure also provides pharmaceutical, nutraceutical and food compositions comprising such protein-enriched preparations as well as pharmaceutical, nutraceutical and food applications of such protein-enriched preparations.

The processes described herein are advantageous because they provide protein-enriched preparations comprising both patatin and potato proteinase inhibitors at a low cost. The process described herein do not include a heating step above 45°C and/or a pH-lowering step below 4.0. In some embodiments, the processes described herein can be used with potatoes, potato waste or potato by-products to extract and used the proteins contained therein. In an embodiment, some of the processes described herein can be designed to minimize modifications in the native potato proteins' conformation so as to provide protein- enriched preparations having more biological activity than protein-enriched preparations obtained using conventional techniques (such as heating above 45°C or treating with an acid to obtain a pH below 4.0). In an embodiment, the protein-enriched preparations described having more than 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100% or more biological activity than protein-enriched preparations obtained using conventional techniques (such as heating above 45°C or treating with an acid to obtain a pH below 4.0). Further, in some embodiments, the process can be tailored to obtain either patatin-enriched or proteinase inhibitors-enriched preparations. In addition, in other embodiments, the processes described herein can also be used to produce natural ingredients.

In the context of the present disclosure, the protein-enriched preparations comprises patatin, potato proteinase inhibitors (also referred herein simply to proteinase inhibitors) as well as combinations of patatin and proteinase inhibitors. The term "patatin" refers to a family of proteins derived from the patatin gene. Native "patatin" members are homodimers having a relative size of about 40 to 45 kDa. Patatins represents about 40% (in weight percentage) of the total potato proteins. Patatin has an isoelectric point of 4.9, exhibits at least 4 different isoforms, one of which having 366 amino acids (shown in Figure 6). In its native configuration, patatin exhibits lipid acyl hydrolase activity. Patatin can be used in pharmaceutical applications to modify (e.g., increase) HDL cholesterol levels, to provide ACE inhibition, to provide anti-oxidant activity. Patatin can be used in food/nutraceutical applications as an anti-oxidant, as an emulsifying agent, as a foaming agent, as a protein substitute. Its use in food/nutraceutical applications is especially useful because patatin exhibits low allergenicity.

On the other hand, "proteinase inhibitors" are a heterogeneous class of proteinases inhibitors having a relative size ranging from 5 to 29 kDa. In the potato, native protease inhibitors content is about 50% (in weight percentage) with respect to the total potato protein content. Proteinase inhibitors have an isoelectric point between 8 and 9. In their native conformation, proteinase inhibitors hinder aspartate proteases and/or metalloproteases. The potato protease inhibitors comprise different proteins which are able to act on a variety of proteases and other enzymes. Known potato protease inhibitors include, but are not limited to, protease inhibitor I (PI-1), protease inhibitor II (PI-2) and potato carboxypeptidase inhibitor (PCI). PI- 1 is a pentameric serine protease inhibitor composed of five 7-8 kDa isoinhibitor protomers. Pl- 1 has a strong affinity for chymotrypsin and inhibits its activities. PI-2 is a dimeric serine protease inhibitor composed of two 10.2 kDa proteins that are linked together by disulfide linkages. PCI is a thermostable 4.3 kDa single subunit peptide. In pharmaceutical applications, proteinase inhibitors can be used to provide satiety, as an anti-microbial agent and/or as an anti-cancerogenic agent. In food/nutraceutical applications, proteinase inhibitors exhibit an excellent solubility over a wide pH range and are sensible to thermal coagulations. In food/nutraceutical applications, proteinase inhibitors can be used as foaming agents, emulsifiers and/or preservatives. Their use in food/nutraceutical applications is especially sought because proteinase inhibitors exhibit low allergenicity.

Process for making protein-enriched preparations from potato

The present disclosure provides a process for making protein-enriched preparations from potato or a potato fraction. An exemplary process of the present disclosure is provided in Figure 1. As a first step 010, the process includes providing a source of potato protein. In the context of the present disclosure, the source of potato protein is a native source of potato proteins, e.g., it is derived from the potato plant and comprises at least one potato cell. Still in the context of the present disclosure, the source of potato protein is not a genetically- modified organism capable of expressing a potato protein, such as a patatin or a proteinase inhibiter. The source of potato protein can be solid, liquid or a combination of solids and liquids (e.g., a suspension). One exemplary source of potato proteins can be the potato tuber itself, including a processed potato tuber. The potato tuber comprises, on average, in weight percentage, a crude protein content of 8.1 %, a carbohydrate content of 87% and a minerals content of 1.86 %. Another source of potato proteins can be a potato fraction such as, for example, potato juice. The soluble solids contained in potato juice comprise, on average, in weight percentage, a crude protein content of 35%, a carbohydrate content of 35% and a minerals content of 20%. The source of potato proteins can be modified prior to being submitted to the process. For example, it can be crushed, filtered, lyophilized, dehydrated (wholly or partly), rehydrated, powdered, etc. Also, some of the parameters of the source of potato proteins can be adjusted prior to being submitted to the process. For example, the mass content, the pH, the temperature and/or the starch content of the source of potato proteins can be adjusted. If the source of potato proteins is modified or one of its parameters being adjusted prior to being submitted to the process, it must be done in way so as to preserve the integrity and native configuration of the proteins contained therein or limit degradation/unfolding of the proteins contained therein. As such, the source of potato protein is preferably not submitted to a heating step above 50°C or to an acidic treatment below a pH of 4.0 prior to being submitted to the process.

Once the source of potato proteins has been provided at step 010, then, at step 020, starch is removed from the source of potato proteins to provide a de-starched mixture. When compared to the material prior to step 020, the de-starched mixture has a starch amount equal to or less than 70%, than 71 %, than 72%, than 73%, than 74%, than 75%, than 76%, than 77%, than 78%, than 79%, than 80%, than 81 %, than 82%, than 83%, than 84%, than 85%, than 86%, than 87%, than 88%, than 89%, than 90% (on a weight basis). For example, when the starch content of the starting material is about 19 to 20% (on a weight basis), the starch content of the de-starched mixture would be between about 6% and 0.19%, and in some embodiments, between about 4% and 5% (still on a weight basis). Many techniques are known in the art to determine the starch level (for example using the potassium iodide colorimetric method) and step 020 can include an optional substep of determining the starch content in the material prior to step 020 and/or after step 020. As indicated below in the Examples section, the removal of starch is important to improve the protein yield. Step 020 is not limited to any particular process known for removing starch from a potato or a potato fraction. Methods known in the art for removing starch from potatoes or fractions thereof include an alkaline method as well as a wet milling procedure. In one particularly advantageous embodiment of the present process, the starch from the source of potato proteins in step 020 is removed enzymatically using an amylase (or a combination of amylases) or a combination of an amylase (including a combination of amylases) and an amylopectinase (including a combination of amylopectinase). In the context of the present disclosure, the enzymes used to break-down starch are referred to a first enzyme (when a single type of enzyme is used) or a first enzyme mixture (when more than one type of enzymes are used). In such embodiment of the process, the source of potato proteins, at step 020, is preferably provided in a liquid (e.g., a solution or a suspension) to favor the enzymatic treatment to breakdown the starch. The use of enzymes to breakdown starch in the source of potato proteins is advantageous in the process described herein because it has limited (e.g., an in some embodiment has no) deleterious side effects on the secondary structure (e.g., folding) and/or the biological activity of the potato proteins. Further, the use of amylase and/or amylopectinase is compatible with further processing steps described herein and as such, there is no need to remove such enzymes from the de-starched mixture. In an embodiment, step 020 is conducted in the presence of amylase only (e.g., in the absence of amylopectinase), for example, in the presence of an oamylase from Bacillus lichenformis. The skilled person will recognize that the conditions (e.g., buffer, temperature, duration of enzymatic treatment, etc.) for conducting step 020 can be varied to achieve the intended starch level to provide the de-starched mixture. One embodiment of the process of the present disclosure includes, in step 020, conducting the enzymatic reaction at a pH of between about 4.5 to 6.5 (and in some further embodiments at a pH of 6.5) and at a temperature of at least 40°C (but not more than 45°C) and in one further embodiment, at a temperature of 40°C). In an embodiment, step 020 can also comprise an optional substep of removing the enzyme or the enzyme mixture used for de-starching the source of potato proteins from the de-starched mixture prior to conducting step 030. However, removing the enzymes at step 020 is not necessary because they do not interfere with downstream operations.

Some of the parameters of the de-starched mixture can be adjusted prior to being submitted to step 030 of the process. For example, the solid content, the pH, the temperature and/or the buffering agent of de-starched mixture can be adjusted. If the de-starched mixture is modified or one of its parameters being adjusted prior to being submitted to the process, it must be done in a way so as to preserve the integrity and native configuration of the proteins contained therein or limit degradation/unfolding of the proteins contained therein. As such, the de-starched mixture is preferably not submitted to a heating step above 45°C or to an acidic treatment around and below a pH of 4.0.

Further, prior to step 030, the de-starched mixture can optionally be submitted to a filtration step to remove, for example, mono- and disaccharides from the de-starched mixture. Such filtration step can include nanofiltration and/or diafiltration.

In the process described herein, the de-starched mixture obtained in step 020 is submitted to an enzymatic reaction using a combination of enzymes at step 030 to provide a treated mixture. It is believed that, in step 030, the combination of enzymes will cause a release of the potato proteins from the potato cells/walls. In step 030, at least two different enzymes are contacted with the de-starched mixture to cause such release: a galactanase (or a combination of galactanases) and a pectinase (or a combination of pectinases). In step 030, the galactanase can be first used to treat the de-starched mixture which can then be submitted to a pectinase treatment. Alternatively, the pectinase can be first used to treat the destarched mixture, which can then be submitted to a galactanase treatment. However, because in the context of the present disclosure galactanase and pectinase treatments are compatible with one another, step 030 can include treating the de-starched mixture simultaneously with at least one galactanase and at least one pectinase. The enzymes used in step 030 are referred herein as the second enzymatic mixture.

The enzyme galactanase, which can be an endo-galactanase or an exo-galactanase, can be provided in a purified form (e.g., in a form which is free from other enzymes) or as an enzyme mixture (e.g. , usually in the form of an enzymatic extract comprising other types of enzymes). In an embodiment, the galactanase is an endo-galactanase and, in a further embodiment, the galactanase can be an endo- 1 ,4-galactanase (from Aspergillus niger for example). Galactanase is an enzyme capable of breaking down the galactan present in the pectic rhamnogalacturonan. Unlike pectin from other sources, pectic polysaccharides obtained from potato have a high proportion of rhamnogalacturonan I (RG I, 75%) and low amount of polygalacturonan (HG, 20%). 67% of potato rhamnogalacturonan I consists of β-linked galactan side chains; while only 10-24% of RG I from other sources are galactan side chains.

The enzyme pectinase, which can be an endo-pectinase or an exo-pectinase, can be provided in a purified form (e.g., in a form which is free from other enzymes) or as an enzyme mixture (e.g. , usually in the form of an enzymatic extract comprising other types of enzymes). In an embodiment, the pectinase is an polygalacturanase, an enzyme capable of breaking down the polygalacturonan present in pectin (from Aspergillus niger for example).

The skilled person will recognize that the conditions (e.g., buffer, temperature, duration of enzymatic treatment, etc.) for conducting step 030 can be varied to achieve the intended level of breakdown of the carbohydrates present in the de-starched mixture. One embodiment of the process of the present disclosure includes, in step 030, conducting the enzymatic reaction in a sodium acetate buffer, at a pH between 4.0 and 5.0 and at a temperature between 37°C and 45°C. In still another embodiment, step 030 is conducted at a pH of at least 4.0, 4.1 , 4.2, 4.3, 4.4, 4.5, 4.6, 4.7, 4.8, 4.9, 5.0, 5.1 , 5.2, 5.3, 5.4, 5.5, 5.6, 5.7, 5.8, 5.9., 6.0, 6.1 , 6.2, 6.3, 6.4 and/or less than 6.5, 6.4, 6.3, 6.2, 6.1 , 6.0, 5.9, 5.8, 5.7, 5.6, 5.5, 5.4, 5.3, 5.2, 5.1 , 5.0, 4.9, 4.8, 4.7, 4.6, 4.5, 4.4, 4.3, 4.2 or 4.1 . In a further embodiment, step 030 is conducted at a pH of about 5.0. In yet another embodiment, step 030 is conducted at a temperature of at least 37°C, 38°C, 39°C, 40°C, 41 °C, 42°C, 43°C, 44°C and/or of less than 45°C, 44°C, 43°C, 42°C, 41 °C, 40°C, 39°C, 38°C, 37°C. In yet another embodiment, step 030 is conducted at a temperature of 40°C. In an embodiment, step 030 can also comprise an optional substep of removing the second enzyme mixture used for providing the treated mixture prior to conducting step 040. However, enzymatic removal at step 030 is not necessary since the galactanase and pectinase can remain in the mixture without altering downstream operations.

Optionally, step 030 can include, in the second enzyme mixture, at least one arabinanase (or a combination of arabinanases) and/or at least one proteinase (or a combination of proteinases). Such optional enzymes can each be added prior to the treatment with the at least one galactanase and/or the at least one pectinase, between the treatment with the at least one galactanase and/or the at least one pectinase or after the treatment with the at least one galactanase and/or with the at least one pectinase. However, because the inclusion of the at least one arabinanase and/or the at least one proteinase is compatible with the galactanase/pectinase treatment, all the enzymes of the second mixture can be added simultaneously to the de-starched mixture to provide the treated mixture. The enzyme arabinanase, which can be an endo-arabinanase or an exo-arabinanase and can be provided in a purified form (e.g., in a form which is free from other enzymes) or as an enzyme mixture (e.g. , usually in the form of an enzymatic extract comprising other types of enzymes). In an embodiment, the arabinanase is an endo-arabinanase (from Aspergillus niger for example). The enzyme proteinase (also referred to as a protease or a peptidase), which can be an endo-proteinase or an exo-proteinase, can be provided in a purified form (e.g., in a form which is free from other enzymes) or as an enzyme mixture (e.g., usually in the form of an enzymatic extract comprising other types of enzymes). In an embodiment, the proteinase breaks down more efficiently or more specifically patatin and can be used to provide a protein-enriched preparation having a higher content of proteinase inhibitors. In another embodiment, the proteinase breaks down more efficiently or more specifically potato proteinase inhibitors and can be used to provide a protein-enriched preparation having a higher content of patatin.

In still another embodiment, the enzyme mixture of step 030 can include purified enzymes (endo-galactanase and endo-arabinanase, polygalacturonissadase) as well as protease inhibitors highly-specific protease(s) (to further increase the proportion of patatin in the protein-enriched preparation) or patatin highly-specific protease(s) (to further increase the proportion of protease inhibitors in the protein-enriched preparation).

The process designed herein can be designed to provide protein-enriched preparations having different content (in weight percentage) of patatin and proteinase inhibitors. In a first embodiment, the protein-enriched preparation can have a substantially similar content in patatin and in proteinase inhibitors (e.g., 50% ± 10% (w/w) for each of the protein fractions or types). In such first embodiment, a combination of purified galactanase and of purified pectinase can be used. Alternatively, an enzyme mixture such as Gamanase®, Viscozyme® or Pectinase® can be used to obtain such first embodiment. In another example, the enzyme mixture can exhibit low arabinanase activity (0.06-013 U/g pulp), low rhamnoglacturonase activity (0.13-0.266 U/g pulp), high galactanase activity (6.6-8.4 U/g pulp) and/or high pectinase activity (3.33-8.8 U/g pulp) can be used to obtain such first embodiment. In a second embodiment, the protein-enriched preparation can have a substantially dissimilar content in patatin and in proteinase inhibitors (e.g., <40% or >60% (w/w) for each of the protein fractions or types). In such second embodiment, the second enzyme mixture usually further comprises at least one arabinanase and/or at least one proteinase. As a first example of this second embodiment, the protein-enriched preparation can have less patatin content (e.g., a patatin content of <40% when compared to the total protein content of the treated mixture) than proteinase inhibitor content (e.g., a proteinase inhibitor content of >60% when compared to the total protein content of the treated mixture). In such first example, a second enzyme mixture such as Ceremix®, Hemicellulase®, Newlase®, Diazyme® or Laminex® can be used to obtain such second embodiment. In another example, a second enzyme mixture exhibiting high arabinanase activity (0.8-4.7 U/g pulp), high rhamnoglacturonase activity (1.13-6.0 U/g pulp), high galactanase activity (6.6-139 U/g pulp) and/or low pectinase activity (0-3.3 U/g pulp) as well as a patatin-highly specific protease can be used. As a second example of such second embodiment, the protein-enriched preparation can have more patatin content (e.g., a patatin content >55% when compared to the total protein content of the treated mixture) than proteinase inhibitor content (e.g., a proteinase inhibitors content <45% when compared to the total protein content of the treated mixture). In such second example, a second enzyme mixture such as logen® and Depol® 670 L or an enzyme mixture exhibiting high galactanase activity (6.6-68 U/g pulp) and more or less similar levels of arabinanase, rhamnoglacturonase and/or pectinase activities (0.04 - 2.7 U/g pulp) as well as protease inhibitors highly specific proteases can be used.

As shown in the Examples below, and in some embodiments, the following multienzymatic systems can be used in the methods and processes described herein: Gamanase® 1 .5L (GAMase), Depol® 670L (DEP), Ceremix® 2XL (CER), Hemicellulase® CE-1500 (HEMase), logen® HS 70 (IOG), Viscozyme® (VIS), Pectinex® Ultra SPL (PEC), Newlase® II (NEWase), Diazyme® L-200 (DIA) and/or Laminex® DG (LAM). Amonst such enzymes, DEP exhibits the highest combined endo^-1 ,4-galactanase; 1 ,5-a-L-arabinofuranosyl/endo- polygalacturonase; 1 ,4-a-D-galactanase-pA activity ratio of 1 15.0, revealing its higher hydrolyzing activity toward side chains of pectic polysaccharides as compared to their backbones. The use of GAMase, HEMase, and CER, expressing an enzyme activity profile with high specificity toward the side chains than the backbones (e.g., βη(_ο-β-1 ,4- galactanase; 1 ,5-a-L-arabinofuranosyl/endo-polygalacturonase; 1 ,4-a-D-galactanase-pA activity ratio of 2.4, 6.3 and 29.0, respectively) led a protein recovery yield of 60.0, 46.7, and 61 .05%, respectively. However, the lower yield (31.2%) obtained with IOG (endo- -1 ,4- galactanase; 1 ,5-a-L-arabinofuranosyl/ endo-polygalacturonase; 1 ,4-a-D-galactanase-pA activity ratio of 12.3) may be attributed (a) to the presence of the proteolytic activity, which may have hydrolyzed the recovered proteins, (b) to the high substrate inhibition and/or (c) to enzyme denaturation due to protein/protein or protein/carbohydrate interactions. On the other hand, NEWase, which had the highest endo-polygalacturonase activity level and endo- polygalacturonase/endo-p-1 ,4-galactanase activity ratio of 4.4, resulted in a higher protein recovery yield of 70.7%. Without wishing to be bound to theory, these results reveal that higher hydrolyzing activity towards the backbones of potato pectic polysaccharides also favored the recovery of potato proteins. As shown in the Examples below, VIS and PEC showed more or less similar efficiency towards the hydrolysis of side chains and the backbones of pectic polysaccharides, with a endo-polygalacturonase/endo-β-1 ,4-galactanase activity ratio of 1.1 -1 .3 and a endo-p-1 ,4- galactanase; 1 ,5-a-L-arabinofuranosyl/endo-polygalacturonase; 1 ,4-a-D-galactanase-pA activity ratio of 0.8-0.9; as a result of this enzyme activity profile, they led to similar protein recovery yield of 65-68%. The comparison of the efficiency of DIA and LAM as a function of their enzyme activity profile indicates the importance of endo-polygalacturonase activity as compared to 1 , 4-a-D-galactanase-pA for the efficient recovery of potato protein. Indeed, DIA and LAM exhibited more or less similar levels of endo-p-1 ,4-galactanase and 1 ,5-oL- arabinofuranosyl; however, DIA showed higher level of 1 , 4-a-D-galactanase-pA, whereas LAM expressed higher endo-polygalacturonase activity level.

One of the advantage to the methods and processes described herein is the ability, in some embodiments, to enrich the protein extract with either patatin or protease inhibitors, as the result of the presence of highly specific proteolytic activity. As shown in Figure 5B, no enrichment was obtained with GAMase, VIS, and PEC multi-enzymatic systems, which were able to isolate both fractions, patatin and protease inhibitors in about the same proportions. Without wishing to be bound to theory, these results can be attributed to the presence of non-significant proteolytic activity toward potato proteins in GAMase, VIS, and PEC multi- enzymatic systems at the investigated concentrations.

On the other hand, DEP and IOG were found to both have an enrichment effect on the recovery of patatin at both incubation time. However, the recovered patatin (73.5-97.1 %) achieved with IOG was very high as compared to DEP (54.8-57.6%) (Figure 5B). This difference may be due, at least in part, to the extremely high proteolytic activity present in IOG at the investigated concentration. Regardless of IOG achieving higher patatin recovery with time, the substantially lower protein recovery rendered IOG inadequate when compared to DEP. Therefore DEP was further optimized for the recovery of patatin.

Further, the use of CER, HEMase, NEWase, DIA, and LAM led to more enrichment of the protein extract with protease inhibitors (> 61 .4%) (Figure 5B). These multi-enzymatic systems have shown different levels of proteolytic activity with DIA, LAM and CER expressing the highest ones. Of those multi-enzymatic systems the one which consistently at both incubation times extracted more protease inhibitors was CER, indicating the high specificity of its proteolytic activity towards the hydrolysis of patatin. All others had an effect on the enrichment with less protease inhibitors with time. Some of the parameters of the treated mixture can be adjusted prior to being submitted to step 040 of the process. For example, the solid content, the pH and/or the temperature of the treated mixture can be adjusted. If the treated mixture is adjusted prior to step 040, it must be done in way so as to preserve the integrity and native configuration of the proteins contained therein or limit degradation/unfolding of the proteins contained therein. As such, the treated mixture is preferably not submitted to a heating step above 45°C or to an acidic treatment below a pH of 4.0 prior to being submitted to the process.

The treated mixture obtained at step 030 comprises oligomers (e.g., fragments of potato carbohydrates, also referred to as a carbohydrate oligomer fraction) as well as native (e.g., uncleaved) potato proteins (e.g., patatin and proteinase inhibitors also referred to as a protein fraction). Once the treated mixture has been obtained at step 030, it is submitted to a step 040 of separating the oligomers from the native potato proteins to obtain the protein- enriched preparations. Step 040 is not limited to any specific technique and includes centrifugation, precipitation and/or filtration. However, to preserve the integrity and native configuration of the proteins contained therein or limit degradation/unfolding of the proteins contained therein, step 040 preferably does not include submitting the treated mixture to a heating step above 45°C or to an acidic treatment below a pH of 4.0. At step 040, filtration (for example, by using a filter capable of filtering out fractions above and below 800 Da) is however preferably used at step 040 so as to maintain the native configuration of the potato proteins. The treated mixture can be submitted to a filtration step, such as, for example, a diafiltration or ultrafiltration step. Diafiltration and ultrafiltration selectively utilize permeable membrane filters (e.g., such as a filter having a molecular weight cut-off of 800 Da and another filter having a molecular weight cut-off of 35 kDa) to separate the components of the treated mixtures (e.g., solutions and suspensions) based on their molecular size. An ultrafiltration membrane retains molecules that are larger than the pores of the membrane (e.g. the protein fraction) while smaller molecules such as salts, solvents, oligomers and water (e.g., the carbohydrate oligomer fraction) freely pass through the membrane. When the treated mixture comprises a solid fraction, step 040 can also include a centrifugation step (either before or after the filtration step) to remove the solids from the treated mixture (not shown in Figure 1).

The protein-enriched preparations can optionally be submitted to a further enrichment step 050, and in an embodiment, a purification, in either patatin (to obtain a patatin-enriched preparation) or proteinase inhibitors (to obtain a proteinase inhibitors-enriched preparation). This step 050 is not limited to any particular technique, as long as such techniques can preferentially isolate patatin from a patatin/proteinase inhibitors mixture or proteinase inhibitors from a patatin/proteinase inhibitors mixture. In the context of the present disclosure, it is suggested to conduct chromatography in step 050 to enrich or purify a specific protein fraction from the protein-enriched preparation obtained at step 040. For example, to enrich or purify patatin from the protein-enriched preparations of step 040, a combination of gel filtration (e.g., using Sephacryl® S-100 HR), anionic chromatography (e.g., using DEAE- Sepharose-CL6B®) and affinity chromatography (e.g., using concanavalin-A-Sepharose 4B and methyl oD-glucopyranoside) can be used. In another example, to enrich or purify proteinase inhibitors from the protein-enriched preparations of step 40, a combination of gel filtration (e.g., using Sephacryl S-100 HR®), anionic chromatography (e.g., using DEAE- Sepharose-CL6B®), cationic chromatography and hydrophobic interaction chromatography can be used.

The protein-enriched preparation described herein can optionally be submitted to a glycation step (e.g., glycating) to reduced its bitterness and/or its allegernicity (Seo et a/. , 2014). This glycation step is especially useful when the protein-enriched preparations comprise patatin. As it is currently known in the art, the glycation step can be conducted via a Maillard reaction.

Optionally, as shown in Figure 1 , the protein-enriched preparations obtained at step 040, the patatin-enriched preparations obtained at step 050 or the proteinase inhibitors-enriched preparations obtained at step 050 can be submitted to a formulation step 060 depending on their intended uses. For example, when the preparations are intended to be included in pharmaceutical compositions, step 060 can include a step of formulating the preparations with a therapeutic agent and/or pharmaceutically acceptable excipients. In another example, when the preparations are intended to be included in nutraceutical compositions, step 060 can include a step of formulating the preparations with a nutraceutical agent and/or an appropriate nutraceutical excipient. In still another example, when the preparations are intended to be included in food compositions (intended for human consumption or animal feed), step 060 can include a step of formulating the preparations with a food and/or an appropriate food excipient. In this example, step 060 can also include formulating the preparations as foaming agents or as emulsifiers to be included in the food composition.

Uses of the protein-enriched potato preparations

In the context of the present disclosure, and as indicated above, the protein-enriched preparations obtained at step 040 can be used without further purification or alternatively can be submitted to a further enrichment step 050 to provide a patatin-enriched preparation or a proteinase inhibitors enriched-preparations. In an embodiment, the protein-enriched preparation can comprise patatin (either in a substantially purified form, or in a mixture with proteinase inhibitors). Such protein-enriched preparations can be included in pharmaceutical compositions in combination with pharmaceutically-acceptable excipients. The pharmaceutical compositions comprising such protein-enriched preparations can be used to provide lipid acyl hydrolase activity. The pharmaceutical compositions can also include other therapeutic agents which can either have lipid acyl hydrolase activity or complement lipid acyl hyrolase activity. Such pharmaceutical compositions can be used for example to treat hypercholesterolemia (for example by increasing HDL cholesterol levels), to treat hypertension (for example by providing angiotensin-converting-enzyme inhibitor (ACE) inhibition) as well as to provide anti- oxydant activity (which may be linked to the prevention of cancer). Alternatively, such protein-enriched preparations can be in nutraceutical/food compositions in combination with other food or nutraceutical elements. Such protein-enriched preparations can be used as:

• an emulsifier or as an emulsifying agent (e.g., a stabilizer of an oil and water mixture (such as an oil-in-water mixture or a water-in-oil mixture) alone or in combination with other emulsifiers;

• a foaming agent (e.g., a stabilizer which limits or prevents the coalescence of bubbles) alone or in combination with other foaming agents;

• an anti-oxidant (e.g., an agent which limits or prevents the oxidization of food) alone or in combination with other anti-oxidants; and/or

• as a protein substitute alone or in combination with another protein source.

In an embodiment, the protein-enriched preparation can comprise proteinase inhibitors (either in a substantially purified form, or in a mixture with patatin). Such protein-enriched preparations can be included in pharmaceutical compositions in combination with pharmaceutically-acceptable excipients. The pharmaceutical compositions comprising such protein-enriched preparations can be used to provide:

• satiety either alone or in combination with another therapeutic agent providing satiety or complementing satiety;

• an anti-microbial agent alone or in combination with another antimicrobial agent; and/or • an anti-cancerogenic agent alone or in combination with another anti-carcinogenic agent.

Alternatively, such protein-enriched preparations can be in nutraceutical/food compositions in combination with other food or nutraceutical elements. Such protein-enriched preparations can be used as:

• an emulsifier or as an emulsifying agent (e.g., a stabilizer of an oil and water mixture (such as an oil-in-water mixture or a water-in-oil mixture) alone or in combination with other emulsifiers;

• a foaming agent (e.g., a stabilizer which limits or prevents the coalescence of bubbles) alone or in combination with other foaming agents; and/or

• a preservative (e.g., which would limit or prevent microbial growth) alone or in combination with another preservative. Their use in food/nutraceutical applications is especially useful because proteinase inhibitors exhibit low allergenicity.

The present invention will be more readily understood by referring to the following examples which are given to illustrate the invention rather than to limit its scope.

EXAMPLE I - PROCESS USING PURIFIED ENZYMES

Materials. Fresh potatoes of Russet Burbank variety were purchased from a local supermarket. Sodium metabilsulfite, sulphuric acid (H ₂ S0 ₄ ), trifluoroacetic acid and hydrochloric acid were purchased from Sigma Chemical Co. (St-Louis, MO). Bradford reagent and SDS-PAGE Broad Molecular weight standard were purchased from Bio-Rad (Mississauga, Ontario). Bovine serum albumen (BSA), Tris base and potassium phosphate dibasic were purchased from Fisher Scientific (Fair Lawn, NJ). Potassium phosphate monobasic was acquired from MP Biomedicals, LLC (Solon, Ohio).

Preparation of potato pulp. Potato pulp was prepared with potatoes of Russet Burbank variety. The potatoes were washed and finely chopped into 0.5 g/mL samples. The potato pieces were ground with a mortar and pestle for 1 min with 1.315 mM sodium metabisulfite. The ground pieces were homogenized using a Warring commercial Blender on low speed for 1 min. Slurry was lyophilized prior to use.

Preparation of protein isolates using heat and acid treatment of potato pulp. Potato fruit juice underwent a combination of thermal/acidic precipitation as previously described by Knorr, Kohler, and Betschart (1977). The pH of potato fruit juice was adjusted to 4.8 and 5.5 with 0.5 H ₂ S0 ₄ . The suspensions were then incubated at 100°C for 2 min. After 5 min of cooling on ice, the suspensions were centrifuged at 8 000g for 50 min at 4°C to recover the protein isolates.

Starch removal. Two selected a-amylases from Bacillus licheniformis (Termamyl) and Bacillus sp. were evaluated for the removal of starch. Dried potato pulp (15% w/v) was consistently weighed and suspended in 10 mM potassium phosphate buffer at pH 6.5. Selected units of a-amylase were added to the potato pulp suspension to yield 0-3 U/mg pulp. Reactions were carried out at 40°C with constant stirring at 220 rpm for 16 hrs. The remaining starch was determined using potassium iodide colorimetric method.

Enzymatic process for obtaining protein preparations. Following starch removal, destarched potato pulp (2.86- 20% w/v) was pH adjusted with 100 mM sodium acetate at pH 5.0. The enzymatic reactions were initiated by combinations of endo-arabinanase (e.g., endo-A) from Aspergillus niger (0.1-0.3 units/mg pulp), polygalacturonanase M1 (e.g., PG) from A. niger (0.008-0.5 units/mg pulp) and endo-1 ,4- -D-galactanase (e.g., endo-G) from A. niger (0.008- 0.5 units/mg pulp) were added to the destarched potato pulp suspension. The reaction mixtures were incubated at 40°C for selected reaction times of 6-48 hrs. Selected enzyme/pulp ratio (0.008-0.5 units/mg potato pulp) and pulp concentration (28.6-200.0 g potato pulp/L) were investigated. After incubation, the reaction mixtures were vacuum filtered using .2 μηι GF/C Whatman filters and the supernatant containing proteins were recovered. The protein content of the recovered pulp after enzymatic treatment and supernatant were determined using Dumas method described by Kirsten and Hesselius (1983).

Protein content determination. Nitrogen content was determined using Leco® TruSpec N (Leco Corporation, St-Joseph, Michigan). Prior to nitrogen determination samples were freeze dried and stored at -80°C. Nitrogen content was multiplied by a factor of 6.25 to determine the total crude protein content.

Sodium Dodecyl Sulphate polyacrylamide gel electrophoresis (SDS-PAGE). Protein isolates were analyzed by SDS-PAGE according to the method of Laemmli (1970) using 5 and 15% acrylamide content in the stacking and resolving gels, respectively. Sample loading was achieved in a mini protein gel apparatus (Bio-Rad) with a 1.5 mm-thick gel. The electrophoresis was conducted at a constant voltage of 120 mV. The gels were then stained for 2 hr with a staining solution of Coomassie brilliant blue R250 (1 %, w/v) in methanol: water: acetic acid (45: 45: 10, v/v/v), followed by destaining in methanol: acetic acid: water (1 : 1 : 8, v/v/v). The analyses of the electrophoretic patterns to obtain the protein profiles were carried out using Red™ Imaging system equipped with Alpha-View™ SA Software. The recovered patatin and protease inhibitors are represented in the actual relative proportion and the calculated extracted patatin or protease inhibitor per gram of initial pulp according to the following calculation:

Gram patatin/gram pulp protein = (Protein recovery yield/100) ^* (relative proportion patatin/100)

Structural characterization of protein preparations using fluorescence spectroscopy. The fluorescence spectra of protein hydrolysate and isolate were recorded using a Fluoromax-4™ spectrofluorometer (Horiba Jobin Yvon system). The spectra were recorded at pH 4 and 9 while varying temperatures from 25 to 85°C with a 5°C interval increase. Excitation was at 295 nm and the resulting emission was measured at 305-450 nm with a scan speed of 120 nm/min. Both the excitation and emission slits were set at 3.5 nm. Protein isolate was prepared using combination of thermal treatment with an acidic pH 4.8. Protein hydrolysate was prepared and run in triplicate; conditions for reaction were 150 mg pulp/mL buffer, 1790 U Termamyl™, 5.56 U/g pulp polygalacturonase and endo- -1 ,4-galactanase.

Experimental design. Optimization of the protein recovery yield, grams of patatin per gram of potato pulp and grams of protease inhibitors per gram of potato pulp was achieved using response surface methodology (RSM). A five-level, five variable central composite rotatable design (CCRD) was used. The full design consisted of 16 factorial points, 10 axial points, and 8 center points, resulting in 34 sets of experiments. The five variables with their corresponding variables consisted of x1 temperature (30, 33.75, 37.5, 41.25, 45°C), x2 time (1.5, 9, 16.5, 24, 31.5 hr), x3 pulp/buffer (80, 1 10, 140, 170, 200 mg/mL), x4 polygalacturonase amount (1.5, 1 1 , 20.5, 30, 39.5 units ) and x5 endo-p-1 ,4-galactanase amount (1.5, 1 1 , 20.5, 30, 39.5 units).

Statistical analysis. For approximating response surface, the obtained yield (%), proportion of patatin/mg pulp and proportion of protease inhibitors/mg pulp based on the described design, was fitted to the general model Equation 1 using the software Design-Expert 8.0.2 (Stat- Ease, inc. Minneapolis, MN, USA): β _ο, Xs (i=1 -5), β _ίη βπ, and are the constant coefficient, coded independent variables, the coefficient for the linear effect the coefficient for the quadratic effect and the coefficient for the interaction effect, respectively. Contour plots were obtained from the fitted model by keeping the independent variables at a constant value while changing the other two variables.

Effect of starch removal on the isolation of potato proteins. The enzymatic approach is based on the degradation of the plant cell wall polysaccharides to release the proteins. Starch removal is an essential process to allow polysaccharide degrading enzymes to have an increased access to the potato cell wall (Thomassen and Meyer, 2010). Selected a-amylases from B. licheniformis (Termamyl™) and Bacillus sp. were evaluated for the removal of the starch. Contrary to the Association of Official Analytical Chemists AOAC method 45.4.07 and the improved one developed by Thomassen and Meyer (2010), lower temperature of 40°C was used to prevent the thermal denaturation of the potato proteins. The necessity of starch removal is better emphasized in Table 1 , where initially at low pulp concentration 75 mg/mL with treatment involving no starch removal the protein yield is 74.6% higher than with starch removal of 66.4 and 70.8% for Bacillus licheniformis and Bacillus sp., respectively. However increasing the pulp concentration to 150 mg/mL the protein yield decreases to 46.8% for no starch removal, whereas with starch removal protein yield is 75.6 and 63.2% for Bacillus licheniformis and Bacillus sp., respectively. Whereas increasing pulp concentrations further to 160 mg/mL resulted in significantly lower protein recoveries when compared with starch removal using both a-amylases.

Table 1 . Effect of starch removal on protein recovery yield (% w/w with respect to the initial weight of protein in the potato pulp)

Starch removal efficiency of these two enzymes was related to the effectiveness of protein recovery after the enzymatic treatment with polysaccharide-hydrolyzing enzymes: polygalacturonase (16.66 U/ g pulp) and endo-1 ,4- -D-galactanase (16.66 U/ g pulp). As shown in Figure 2, the removal of starch with α-amylase from Bacillus sp. resulted in an increase of the protein recovery yield upon enzymatic treatment to a maximum value of 48.7% at 1 U/mg pulp; whereas above this concentration, a sharp decline of the protein recovery yield resulted. A previous study conducted by Thomassen and Meyer (2010) found that enzyme dose, amount of dry matter and enzyme-enzyme interactions had a significant impact on starch reduction. This sharp decline of protein recovery could be affected by the interactions between the a-amylase Bacillus sp. at higher concentrations and the polysaccharide degrading enzymes. This interaction effect was not observed in the presence of α-amylase B. Iicheniformis at higher concentrations. Figure 2 shows that α-amylase from B. Iicheniformis resulted in an optimal protein yield greater than 50% at volume addition of 1 to 2 U/mg pulp. Therefore, in these experimental conditions, α-amylase from 8. Iicheniformis was proven to be more effective for starch removal as high protein recovery was achieved at broad range of enzyme addition. Using a smaller amount of enzyme while generating the same protein recovery is a desirable property when an enzymatic reaction is scaled up. This process was scaled up to double amounts of potato pulp (150 mg potato pulp/ml) and 1.4 U/mg pulp of α-amylase from B. Iicheniformis, which were proven effective for starch removal and led to an improved efficiency of the polysaccharide degrading enzymes (data not shown).

Enzymatic isolation of potato proteins and their structural characterization. Following starch removal the plant cell wall is mostly composed of polysaccharides cellulose, hemicelluloses and pectin. Cellulose is present to strengthen and add rigidity to the cell wall and is composed of β-1 ,4 linkages of glucose molecules. Like cellulose, hemicellulosic polysaccharides are present in the cell wall to strengthen the wall; however they are commonly classified as four categories xylans, mannans, β-glucans with mixed linkages, and xyloglucans. The pectic polysaccharides are present within the plant cell wall to regulate water flow and maintain cell placement. Potato plant cell wall pectic polysaccharides are mainly composed of rhamnogalacturonan I and homogalacturonan. Several studies have examined the degradation of plant cell walls for the assessment of enzymes specificity to monitor the pathogenesis, functions with plant physiological development (ripening, germination, cell wall extension). Investigation of the release of cell wall protein from potato, carrot, and cotton using several glycosyl -hydrolysing enzymes underlined that the cleavage of the galacturonic linkages and the opening of cell wall through pectic polysaccharide degradation are crucial for protein recovery. No studies have taken advantage of this subtractive isolation method to provide plant proteins having improved functionality for their use as, for example, health promoting ingredients.

Initially the exploration of three pure enzymes polygalacturonase M1 (Aspergillus niger), endo-p-1 ,4-galactanase (Aspergillus niger) and endo-arabinanase (Aspergillus niger) were deemed necessary for effective protein recovery. The addition of endo-arabinase (endo-A) did not improve the protein recovery (data not shown). Due to this results, arabinanase was deemed not necessary for improved protein recovery and was not further optimized. To study the effect of the process on the structural stability of proteins extracted, fluorescence spectroscopy was used to monitor conformational changes of the aromatic acid environment as a function of temperature. The wavelength of the maximum intensity varied 345 and 360 nm for both protein preparations recovered upon enzymatic treatment and protein isolate preparations obtained upon combined acidic/thermal treatment at room temperature, respectively (data not shown). Fluorescent spectroscopy results show that protein isolate preparations resulted in higher fluorescence intensity when compared to protein hydrolysate preparations (Figure 3A). Without wishing to be bound to theory, it is believed that the thermal and acidic adjustment used to generate the isolate preparations favors protein unfolding. It is also believed that the acid- and thermal-treatment favors exposition of the tryptophan residues in the polar environment resulting in higher fluorescence intensity with further quenching due to the formation of aggregates at increasing temperatures. In contrast, the protein preparations recovered upon enzymatic treatment contained less protein denaturation as the fluorescence intensity of these preparations is lower (Figure 3A). It is believed that in protein preparations obtained upon enzymatic treatment, protein unfolding and exposure of the tryptophan residues was not as pronounced as in the combined acidic/thermal-based protein isolate preparations. The gradual decrease in fluorescence intensity for the enzymatic-based protein preparation had a lower velocity of 2.2 fold compared to the acidic/thermal-based protein isolate preparation where a reduction of 3.8 fold resulted. Emphasizing the minimal deleterious effect of the enzymatic isolation technique compared to the industrially performed thermal and acid adjustment.

At both pH 4.0 and 9.0 (Figures 3B and 3C), the thermal conformational changes of protein preparations obtained by enzymatic extraction showed a linear behavior with a decrease in fluorescence intensity as temperature increases. With increasing temperature the conformational changes of the tryptophan environment associated with the enzymatic-based proteins was less extensive when compared to the proteins of the acidic/thermal-based isolate preparation. In contrast, the proteins of the acidic/thermal-based isolate preparations exhibited an exponential behavior of the maximum tryptophan fluorescence intensity up to 45 °C, followed by a decrease with lower rate; suggesting a major change in the molecular conformation took place at temperature below 45°C. It has been reported that patatin began unfolding at 46-55°C, whereas above 55°C the polarity surrounding the tryptophan residue is unchanged, as they are not available to react with water. The results shown in Figures 3B and 3C also indicate that acidic/thermal-based protein isolate preparations showed the highest substantial changes in tryptophan fluorescence intensity at pH 4.0 and 9.0. These results confirm the influence of thermal/acidic conditions on the protein stability of the isolate preparations compared to the stability of the enzymatic-based proteins which is less affected. According to the initial linear rate of intensity changes, the protein preparations seem to exhibit more structural unfolding effect on the tryptophan environment at pH 9.0 than at pH 4.0. At pH 4.0 (close to patatin's isoelectric point), most patatin may be insoluble, the differences in intensity at this pH can be attributed to the thermal stability of the protease inhibitors. On the contrary because pH 9.0 is within the range of the isoelectric point of the protease inhibitors (pH 7 to 9), it can be assumed that the recorded thermal unfolding is related to a greater extent to the soluble patatin.

Effects of selected parameters on the protein yield and profile. In order to understand the effect of the reaction parameters including, temperature, time, pulp concentration, units of PG and units of endo-G on the recovery of potato proteins and their corresponding fractions was optimized using a central composite rotatable design (CCRD). The design selected contained five levels and five factors temperature (30-45°C), time (9-31 .5 hrs), pulp concentration (80-200 mg/mL), units of PG (0.74- 35.27 U/ g pulp) and units of endo-G (1.34- 35.68 U/ g pulp ). The levels selected for each parameter was chosen based on preliminary trials. Optimization was performed to increase protein recovery yield and proportion of patatin and protease inhibitors. The central point conditions of the CCRD design was chosen to be 37.5°C, 16.5 hrs, 140 mg/mL, 18.3 units of PG and 18.3 units of endo-G (Table 2). In Table 2 the experimental conditions, the actual experimental and the predicted values for protein recovery and proportion of patatin and protease inhibitors extracted according to the CCRD, is presented.

Table 2. Reaction parameters for protein recovery and protein proportion from potato pulp in the statistically designed screening experiments.

aActual factor

bCoded factor

cExperimentally obtained value

dRelative proportion (%) obtained from Alpha View software eExperimentally calculated gram of patatin/gram of potato pulp protein

When examining Table 2 the protein yield ranged from 42.41 -73.91 %. The lowest protein yield resulted from low temperature (-1), low time (-1), high pulp/buffer (+1), low units of PG (-1) and endo-G (-1), whereas highest protein yield was obtained for center point temp (0), mid time (0), lowest pulp/buffer (-2), center point units of PG (0) and endo-G (0). A general trend shows that increasing temperature resulted in improved protein yield although temperature had to be limited to 45°C as patatin has been shown to start unfolding at higher temperatures. Without wishing to be bound to theory, when pulp concentrations were low, higher protein recovery yield resulted which could be attributed to the increased movement of the enzyme in the aqueous environment resulting in higher substrate accessibility (Table 2). However optimally for improved industrial applications, the use of high pulp concentration would be more desirable (Molden and Sakthivadivel, 1999).

Table 2 shows the relative proportion of patatin and protease inhibitors recovered for each experiment run. The relative proportion of patatin was found to vary between 26 and 65.5% (Table 2). Lowest patatin recovered (26%) was extracted with low temperature (-1), long time (+1), low pulp/buffer (-1), and low enzyme amounts for both PG and endo-G (-1). Higher patatin (65.5%) was extracted with lowest temperature (-2) and midpoint values for the remaining parameters (0), experimental run 1. When increasing temperatures to highest axial point of 45°C while maintaining center points for all other parameters patatin extraction decreased by 1.24 fold from 65.5%.

Therefore temperature was extremely important for extracting patatin and this was confirmed by the results provided in Table 3. This could be due to the unfolding of patatin which begins at 45°C rendering it less or insoluble, therefore remaining in the degraded pulp and in turn not recovered. However this occurrence was limited as temperature was set to not exceed 45°C and buffering systems were maintained at 50 m and would therefore preserve patatin's structure. A similar phenomenon has been described by van Koningsveld et al. (2001); increasing salt concentrations from 15 to 200 mM increased the denaturation temperature for some potato proteins by shielding the protein from electrostatic repulsive forces.

The relative proportion of protease inhibitors recovered varied from 20.1 to 74.8% (Table 2). Contrarily to patatin recovery, lowest proportion (20.1 %) was attained for experimental run 1 where lowest temperature (-2) and midpoint values for the remaining parameters (0 ). Highest protease inhibitor recovery was achieved in run 18, which is high temperature (+1), short time (-1), high pulp/buffer (+1), high PG (+1 ), and low endo-G (-1). Therefore as shown temperature seems to have a strong effect on protease inhibitors extracted (Table 3). Temperature was maintained at these conditions for preservation of patatin but temperature could have increased for protease inhibitor recovery as they have been shown to denature between 58 and 71 °C.

Model fitting and analysis of variance. Multiple regression analysis was used to evaluate the best fitting model using the software Design-Expert™ version 8.0.2. The models statistical significance was investigated based on F-values, P values, lack of fit, and predicted and adjusted R ² values (results not shown). For protein recovery yield the quadratic model was selected with a backward elimination (or backward reduction algorithm) and an alpha-score of 0.05; the model selected for gram patatin/ gram pulp was found to be quadratic with a backward treatment and an alpha-score of 0.1 ; and the model selected for gram protease inhibitors/ gram pulp 2FI interaction with a backward treatment and an alpha-score of 0.1.

Table 3 shows the analysis of variance (ANOVA) for the CCRD concerning the responses protein recovery yield, patatin and protease inhibitors recovery per gram of pulp, the F-values of 5.18, 4.94, and 5.16 with corresponding P-values 0.0012, 0.0023, and 0.0007, respectively show that all three models are statistically significant. As well as the non-significant lack of fit of 0.44 (P-value 0.9289), 0.69 (P-value 0.7564), and 1.02 (P-value 0.5262) for protein recovery yield, patatin recovery, and protease inhibitors recovery, respectively. All responses showed a predicted R ² in reasonable agreement with adjusted R ² (results not shown ) therefore the models selected were appropriately chosen to analyze the corresponding responses protein recovery yield, patatin recovery and protease inhibitors recovery.

Table 3. Analysis of variance results presented in Table 2.

a Not significant

b a-value 0.05

c a-value 0.1

d a-value 0.1

For protein recovery the two significant linear terms include pulp/buffer (F-value 10.31 , P- value 0.0034) and incubation temperature (F-value 7.96, P-value 0.0088). Both enzyme amounts had no significant effect on protein recovery; however the interaction effect of PG and incubation temperature (F-value 5.56, P-value 0.0258 ) showed significant effect on protein recovery. Incubation time had a significant quadratic effect towards protein recovery. For patatin recovery, the significant linear effects included all parameters with the exception of incubation time and nits of PG whereas the only linear significant term for protease inhibitor recovery included incubation temperature. Like protein yield, patatin recovery also demonstrated a quadratic effect towards incubation time (Table 3). Longer time was found to negatively impact protein recovery yield to a greater extent when compared to patatin recovery. For all responses incubation temperature exhibited a positive effect, where increased temperature improved protein yield and protein profile recovery, although temperature was not allowed to exceed 45°C as patatin has been shown to unfold at that temperature. Pulp concentration showed a positive effect for both protein yield and patatin recovery. In terms of enzyme additions, patatin was significantly affected by units of endo-G with a negative correlation. Recovered protease inhibitors were affected by the interaction terms for incubation temperature with pulp concentration and with units of endo-G. Incubation temperature and pulp concentrations had an improving effect on protease inhibitor recovery, whereas incubation temperature and units of endo-G had a detrimental effect to protease inhibitor recover .

Effect of reaction parameters on optimal conditions. The significant interaction effects of selected parameters and responses towards protein yield, patatin recovery, and protease inhibitor recovery is shown in Table 3. For yield, a significant interaction was temperature with units of PG (Figure 4A). The interaction shows that when the units of PG are low and the temperature is high a high protein yield resulted. As shown in Table 2, this relationship was seen in several of the experimental runs specifically run 10, 21 , 22, 23, 25 with corresponding yields of 69.7%, 71 .1 %, 64.2%, 63.1 %, 70.9%, respectively. A previous study confirms the necessity in hydrolysing the galacturonide segments to release cell-wall proteins which was previously conducted with polygalacturonase purified from Verticillium alboatrum (Strand, Rechtoris, and Mussell, 1976). Like protein yield, extracted patatin was also affected by the interaction of units of PG and temperature (Figure 4C) where midpoint temperatures in combination with high units of PG extracted the high proportion of patatin. Indeed, extraction of protease inhibitor recovery also exhibited similar trend as protein yield for the interaction of the units of PG and temperature, shown in Figure 4D. High units of PG and increased temperature resulted in high proportion of protease inhibitors. To preserve the patatin extracted temperatures were maximally set at 45°C as patatin begins to unfold at these conditions (Waglay, 2014). This denaturation effect could lead to patatin being retained in the degraded pulp instead of filtering into the supernatant.

Time was found to have a large negative quadratic effect towards protein yield (Table 3). This is also shown in experimental run 17 (Table 2) where all parameters were kept at the center points whereas time was at the high axial point resulting in a protein yield of 54.7%. Indeed, this low protein yield could be due to an antagonistic effect between starch removing amylases and glycosyl-hydrolases over prolonged time.

As shown in Figure 4B, extracted patatin recovery was significantly affected by the interaction of pulp/buffer and time where midpoint pulp buffer and time resulted in high extraction of patatin recovery. Increasing temperature with low pulp/buffer concentrations resulted in decreased patatin recovery, which could be explained by the increased diffusion between enzymes and thereby enhancing the negative interaction effect between starch removing and polysaccharide degrading enzymes.

The extraction of protease inhibitors was affected by the interaction of pulp/buffer and temperature (Figure 4E) where increasing both factors resulted in high protease inhibitor recovery. Potato protease inhibitors tend to be relatively heat stable compared to patatin where they begin to denature at temperature above 50-60°C (van Koningsveld, Gruppen, de Jongh, Wijingaards, van Boekel, Walstra, Voragen, 2001).

Taking into consideration these models significant linear parameters, interactions, and quadratic effects two optimal conditions were chosen to validate the models and were run in triplicate. Table 4 shows the two optimal experimental runs where run 1 , x1= 42.6°C, x2=21.03 hrs, x3= 126.4 mg/mL, x4= 1.89 units, x5= 1.5 units or run 2, x1= 43°C, x2= 19.72 hrs, x3- 182.6 mg/mL, x4- 1 .53 units, x5- 8.88 units. Both conditions were chosen for their high predicted protein yields of 79.4 and 73.9%, respectively and their high predicted patatin recovery 0.355 and 0.433 g patatin/g pulp, respectively. As shown in Table 4 both reactions run 1 and 2 resulted in actual yields of 67 and 59.5%, respectively. Run 1 resulted in a higher extraction of patatin recovered of 0.325 whereas Run 2 resulted in higher recovery of protease inhibitors 0.328. These yields and extracted patatin and protease inhibitor recovery can be accepted as they fall into the predicted interval range (Table 4). Therefore the actual obtained results confirm the validity in the model. Table 4. Model Validation

The approach suggested in this example is an optimized enzymatic approach to isolate potato proteins by degrading plant cell wall constituents using pure enzymes polygalacturonase M 1 and endo-p-1 ,4-galactanase. This process required an initial step to remove of starch by using α-Amylase from B. licheniformis as it was proven to be more effective due to the high protein recovery which was achieved at broad range of enzyme addition. Protein yield, extracted patatin recovery, and extract protease inhibitors recovery were affected by temperature; however it was necessary to limit temperature as it has been shown that patatin begins to unfold at 45°C. The developed enzymatic approach has the potential to isolate potato proteins with minimal deleterious effects. This approach will help broaden their applications as a value added ingredients.

EXAMPLE II - PROCESS USING ENZYME MIXTURES

Enzymatic process for obtaining protein preparations. Following starch removal, destarched potato pulp (obtained following the methodology of Example I, 16% w/v) was pH adjusted with 100 mM sodium acetate at pH 5.0. The enzymatic extraction of proteins were initiated by adding selected multi-enzymatic products, such as Gamanase®, Depol®, Ceremix®, Hemicellulase®, logen®, Viscozyme®, Pectinase®, Newlase®, Diazyme® or Laminex® to destarched potato pulp at 40-45°C for selected reaction times of 0.4-48 hrs. Gamanase®, Viscozyme® and Pectinase® exhibit low arabinanase activity (0.06-013 U/g pulp), low rhamnoglacturonase activity (0.13-0.266 U/g pulp), high galactanase activity (6.6-8.4 U/g pulp) and high pectinase activity (3.33-8.8 U/g pulp). Ceremix®, Hemicellulase®, Newlase®, Diazyme® or Laminex® exhibit high arabinanase activity (0.8-4.7 U/g pulp), high rhamnoglacturonase activity (1.13-6.0 U/g pulp), galactanase activity (6.6-139 U/g pulp) and low pectinase (0-3.3 U/g pulp) and patatin-highly specific protease activity, logen® and Depol® 670 L showed high galactanase activity (6.6-68 U/g pulp) and more or less close levels of arabinanase, rhamnoglacturonase and pectinase activities (0.04 - 2.7 U/g pulp) and protease inhibitors highly specific proteases. After incubation, the reaction mixtures were vacuum filtered using 1.2 pm GF/C Whatman filters and the supernatant containing proteins were recovered. The protein content of the recovered pulp after enzymatic treatment and supernatant were determined using Dumas method described by Kirsten and Hesselius (1983).

As shown on Figure 5A, an overall high yield was achieved at reaction time of 20 hrs, with the exception of Diazyme® (data not shown). Highest yields of >65% was obtained with the enzymes with the highest galactanase activity level (1210-6751 U/ml), such as Depol® 670L, Viscozyme® and Pectinase®. As shown on Figure 5B, in terms of protein fractionation, a high proportion of patatin (74% w/w) was obtained with the enzyme mixture, such as logen® having a galactanase: rhamnogalacturonase and rhamnogalacturonase: arabinanase enzymatic ratios of 9.3 and 1 .4 (U: U) as well as a highly protease-inhibitors-specific protease. While high proportions of protease inhibitors (90% w/w) was recovered using enzyme mixtures, such as Ceremix®, Hemicellulase®, Newlase®, Diazyme® or Laminex®, having high arabinanase activity (0.8-4.7 U/g pulp), high rhamnoglacturonase activity (1.13- 6.0 U/g pulp), and low pectinase (0-3.3 U/g pulp) and patatin-highly specific protease.

Statistically designed screening experiments (as described in Example I) were conducted with the Ceremix® enzymatic mixture and the results are shown in Table 5. Table 5 shows that the protein recovery yield upon treatment with Ceremix® and the purity of factor of protease inhibitors (PI) were highly dependent on the reaction time and the amount of enzyme. The highest yield of 67.9% was achieved upon treatment of potato pulp with 151 .84 U/g pulp (galactanase activity) for 26 hr. the use of shorter reaction time resulted on low yield of 27.6%. On the other hand, the highest enrichment level (purity factor of 0.85 and 0.78) with protease inhibitors (PI) was obtained upon longer enzymatic treatment (26-30 hrs).

Table 5. Reaction parameters for protein recovery and protein proportion from potato pulp using commercial multi-enzymatic system in statistically designed screening experiments.

aActual factor

Coded factor

cExperimentally obtained value

dRelative proportion (%) obtained from Alpha View software

eExperimentally calculated gram of patatin/gram of potato pulp prote

'Center points (more than triplicate)

One of the benefit of using enzyme mixture major benefit to this technique is the elimination of a potentially costly purification step; indeed, these multi-enzymatic systems have the ability to isolate proteins, while fractionating them due to their specificity and selectivity with minimal proteolytic activities.

EXAMPLE III - FUNCTIONAL CHARACTERIZATION OF PROTEIN-ENRICHED

PREPARATIONS ENZYME MIXTURES

The proteins preparations have been obtained using a thermal/acid treatment (as described in Example I), a combination of purified galactanase/pectinase (as described in Example I) or an enzyme mixture (as described in Example II). The lipid acyl hydrolase activity (e.g., one of patatin's biological activity) was investigated using p-nitrophenyl butyrate as a substrate. The results of such characterization are shown in Table 6. Further, the proteinase inhibition activity against trypsin and chymotrypsin was estimated from the decrease in the protease activity in the presence of potato isolates containing protease inhibitors. The proteinase inhibition activity of the protein preparations is shown in Table 7.

Table 6. Lipid acyl hydrolase activity expressed in protein recovered from potato pulp obtained using enzymatic isolation.

Extraction Techniques p-Nitrophenyl Butyrate

Specific Activity ³

(Mmole/(min. mg protein))

Thermal/Acidic Combination Not detected

Pure enzyme treatment recovery 0.10

Multi-enzymatic (Depol® 670) treatment

0.18

recovery

Table 7. Proteinase inhibition activity of various protein preparations expressed protease inhibited by g potato protein isolates.

.EXAMPLE IV - RECOVERY OF POTATO PROTEIN ISOLATES USING LABORATORY

ISOLATION AND PILOT-PLANT SCALE UP ISOLATION

Materials Fresh potatoes of Russet Burbank variety were purchased from a local supermarket. Sodium metabisulfite and termamyl alpha-amylase (Bacillus licheniformis) were purchased from Sigma Chemical Co. (St-Louis, MO). Food grade sodium sulfite and ammonium sulphate was purchased from Quadra Ingredients (Montreal, Quebec). Bradford reagent and SDS-PAGE broad molecular weight standard were obtained from Bio-Rad (Mississauga, Ontario). Tris base and potassium phosphate dibasic were acquired from Fisher Scientific (Fair Lawn, NJ). Potassium phosphate monobasic was purchased from MP Biomedicals, LLC (Solon, Ohio). Commercial multi-enzymatic system Depol 670L from Trichoderma reesei, was provided by Biocatalyst Ltd (Bensenville, Illinois).

Ammonium sulphate precipitation (Laboratory Scale). Potato fruit juice (PFJ) was prepared according to the modified method of van Koningsveld et al. (2001). Potatoes of Russet Burbank variety were washed and chopped into large pieces. Potato samples (100 g) were suspended in 50 mL sodium sulfite solution (1 g/L) to prevent polyphenol oxidation and homogenized using a Waring Commercial Blender on low speed for 5 min, after which the potato slurry was subjected to cheese cloth filtration. The resulting turbid liquid was centrifuged at 8 OOOg for 30 min at 4°C using a Beckman Centrifuge Model J2-21. The yellowish filtrate is known to be similar to industrial PFJ, and was lyophilized prior to use. Lyophilized PFJ was then suspended in water and subjected to 60% (NH ₄ ) ₂ S0 ₄ saturation with constant stirring for 1.5 hrs at 4°C. The precipitate was recovered by centrifugation (8 OOOg, 50 min). The recovered protein precipitate was then dialysed for 2 days with a molecular weight cut off of 3 000 - 6 000 Da. The protein content of the recovered precipitate and supernatant were determined. Enzymatic-based isolation (Laboratory Scale). Potato pulp was prepared with Russet Burbank potatoes. The potatoes were washed and finely chopped into 0.5 g/mL samples. The potato pieces were ground with a mortar and pestle for 1 min with 0.25 g/L sodium metabisulfite. The ground pieces were homogenized using a Waring commercial Blender on low speed for 1 min. Termamyl was added to the slurry to yield 1220.63 U of a-amylase/ g potato pulp. Reactions were stirred for 17 hrs at 25°C. Following the incubation with the starch degrading enzyme, Depol 670L multi-enzymatic system (2.03 U of galactanase/ gram potato pulp) was added, and the mixtures were stirred for 5 hrs at 25°C. After incubation, the reaction mixtures were vacuum filtered using 1.2 pm GF/C Whatman filters and the supernatant containing proteins were recovered. The degraded pulp was dried in an oven for one week (60°C), whereas the recovered supernatant was lyophilized. The protein content of the enzymatically treated pulp and supernatant were determined.

Ultrafiltration and ammonium sulphate precipitation (Pilot Plant Scale). PFJ was prepared according to the modified method of van Koningsveld et al.(2001) (Figure 7). Potatoes were washed and chopped into large (1 cm ³ ) pieces using a Big Chop C15 (Stephan Mikrocut, Hamelan, Germany). The entire potato sample (170 kg) was suspended in 98 L of sodium sulfite solution (0.77 g/L) to prevent polyphenol oxidation and homogenized to cubes of 1 mm ³ , using a chopper GK (Urshel Laboratories Inc., Indiana, USA). The potato slurry was subjected to constant stirring for 1 hr at 2°C, which allowed time for the proteins to solubilize. he resulting turbid liquid was decanted at a speed of 4 800 rpm for 20-30 sec, with a feed rate of 380 L/hr using a CBB Decanter s.r.1 DR250-EF (Drycake, Surrey, Canada). The opaque juice was then subjected to centrifugation with speed of 1 1 000 rpm and feed rate of 200 L/hr using a Dexter MiSR 1010 (SRS A USI Company, Michigan, USA). The clear yellowish filtrate is known to be similar to industrial PFJ.

A fraction of PFJ was ultrafiltrated using a Koch Hollow Fiber Cartridge (Koch Membrane Systems Inc., Massachusetts, USA) with a molecular weight cut off of 5 000 Da. Following the ultrafiltration, the concentrate was diafiltered two times with equal water additions. The collected retentate was lyophilized and the protein content of the retentate was determined. The freeze dried powder of potato protein isolated by ultrafiltration of PFJ is further abbreviated as PPC UF.

The other fraction of PFJ was subjected to 60% saturation with (NH ₄ ) ₂ S0 ₄ at 2°C with constant stirring for the precipitation of proteins. The precipitation occurred for 17 hrs, after which the suspension was centrifuged with a speed of 1 1 000 rpm and feed rate of 200 L/hr using a Dexter MiSR 1010 (SRS A USI Company, Michigan, USA). A subsequent ultra- and dia-filtration steps were performed, on the recovered precipitate, using a Koch Hollow Fiber Cartridge (Koch Membrane System Inc., Massachusetts, USA) with a molecular weight cut off of 5 000 Da. The diafilteration step was performed with 3 times equal volume additions of water. The retentate was collected on ice and lyophilized. The protein content of the recovered precipitate was determined. The freeze dried powder of potato protein isolated by 60% (NH ₄ ) ₂ S0 ₄ saturation is further abbreviated as PPI AS.

Enzymatic-based isolation. Potatoes of Russet Burbank were used to prepare potato slurry as described above. Starch degrading enzyme, Termamyl (a-amylase) (250 ml_), was added to the slurry, and the mixture was incubated at 25°C for 17 h. This was followed by the addition of 2.03 galactanase units/ gram of potato of Depol 670L, after which the mixture was incubated for 5 h at 25°C, with constant stirring (100 rpm). The resulting turbid slurry was decanted at a speed of 4800 rpm for 20-30 sec, feed rate 380 L/hr using a CBB Decanter s.r.1 DR250-EF (Drycake, Surrey, Canada). The opaque juice was then subjected to centrifugation with speed of 1 1 000 rpm and feed rate of 200 L/h using a Dexter MiSR 1010 (SRS A USI Company, Indiana, USA). A Koch Hollow Fiber Cartridge (Koch Membrane Systems Inc., Massachusetts, USA) was used with a molecular weight cut off of 5 000 Da for the ultra- and dia-filtration steps. The collected supernatant was ultrafiltered to concentrate the retentate 2.4 times, which was followed by diafiltration steps with 3 times equal volume additions with water. The recovered supernatant and degraded pulp were lyophilized and their protein content were determined. The freeze dried powder of potato protein isolated by enzymatic method is further abbreviated as PPC Enz.

The extraction methods were performed in a continuous flow pilot plant facility, commencing with the preparation of imitation industry by-products PFJ and potato pulp (Figure 7). However it should be noted that in real applications these by-products would be supplied. Figure 7 shows the conversion of 170 kg of potatoes to either PFJ or potato pulp suspension. The major difference between PFJ and potato pulp is the incorporation of water and the decanting process. For the preparation of PFJ potato to water followed a 1.6 times relationship, whereas pulp was slightly higher with a ratio of 2 (data not shown). The decanting process for the potato pulp occurred following enzymatic hydrolysis, whereas the decanting process for PFJ preparation was prior to ultrafiltration and precipitation steps (Figure 7). The overall process of preparing PFJ resulted in a 22% loss of the total volume (data not shown), this loss is expected due to the continuous process of the decanter and centrifugation steps. Despite these losses, improvements to the manufacturing are not necessary as these by-products are available in large supply. PFJ was subjected to ultrafiltration and several diafiltration steps (Figure 7), in order to concentrate and purify the proteins. This resulted in an extract (PPC UF) with a high carbohydrate content, specifically of the oligomer consisting of approximately 40-subunits of glucose as indicated by HPLC size exclusion (results not shown).

In addition, PFJ underwent 60% (NH ₄ ) ₂ S0 ₄ saturation, with constant gentle stirring at 2°C (Figure 7). The stirring process needed to be kept minimal (100 rpm) and at low temperatures in order to preserve the proteins structure as well as to decrease the occurrence of foaming. Foaming was a prominent occurrence throughout the extraction process which in turn led to protein loss, exact quantification was difficult to determine due to interference with the (NH ₄ ) ₂ S0 ₄ addition. The centrifugation step employed to separate the precipitated protein from the supernatant also resulted in protein loss, again, exact quantification was difficult to establish due to (NH ₄ ) ₂ S0 ₄ interference. In order to decrease losses a batch centrifuge could be used, however this would decrease load capacity and thereby decrease process efficiency. Similar to ultrafiltration, substantial protein loss resulted from the ultra- and dia-filtration steps, where again interference resulted from the (NH ₄ ) ₂ S0 ₄ present in the waste-stream permeate. While considerable diafiltration steps were required in order to remove the interfering (NH ₄ ) ₂ S0 ₄ , in industrial applications this waste stream could be used to recover the salt, thereby decrease associated purchasing cost.

The third process consisted of the use of commercially available multi-enzymatic system Termamyl (a-amylase) and Depol 670L (glycosyl-hydrolase enzymes), for starch and polysaccharide degradation, respectively.

Table 8 shows the recovery efficiency of PPI and PPCs on the lab- and pilot- scales. The pilot-scale processes were run in a continuous system, whereas at the lab-scale, they were processed as an individual batch system. The results show that the protein recovery yield by (NH ₄ ) ₂ S0 ₄ precipitation (74.4%) was significantly higher on laboratory scale; these

Table 8. Effects of Extraction Process Scale-Up of Potato Protein Isolates on the Protein Recovery Yield, Protein Content, and Relative Proportion of Potato Protein Fractions

aUltrafiltration occurred on potato fruit juice with a molecular weight cut-off of 5 kDa, abbreviated PPC UF

Ammonium sulphate occurred on potato fruit juice with 60% saturation, abbreviated PPI AS

Multi-enzymatic system used was Depol 670L, abbreviated PPC Enz

Not determined

results can be attributed to the high accessibility of the salt to the proteins. In addition, the lower yield (PPI AS, 44.6%) on pilot scale could be due to losses during decanting and centrifuging as both these machines follow a continuous flow. Similarly, lower yields were previously reported during continuous pilot plant fractionation of soybean proteins when compared to laboratory extraction (Wu, Murphy, Johnson, Fratzke, & Reuber, 1999). The enzymatic-based isolation process on pilot-scale (PPC Enz, 29.1 %) resulted in more than half the protein recovery efficiency as that by lab-scale (50.9%). These results reveal that the efficiency of both (NH ₄ ) ₂ S0 ₄ and the enzymatic approach was not affected significantly by the process scale up. Interestingly, higher extract recovery yields per kilogram potato of 7.3 and 5.6 g extract/kg potato were obtained for both PPI AS and PPC Enz, respectively, outlining the efficiency of both processes. Contrarily, the protein recovery yield and extract recovery yield for ultrafiltration were low, 21.6% and 3.8 g extract/ kg potato, respectively (Table 8), which emphasizes the need for the addition of extracting agents for the efficient recovery of potato protein. In comparison, PPI AS resulted in a high protein content of 86.5; and therefore referred to as marketable protein isolate. Whereas the lower protein content possessed by PPC UF and PPC Enz of 75.9 and 40.7%, respectively refers to the commercial potential as a protein concentrate. Further, diafiltration and/or utltraflltration of PPC Enz can increase its protein content.

To date, the sole technique that has been performed on a pilot scale has been the exploration of several different resins for chromatographic separation. Straetkvern et al. (1999) explored the use of two resins that varied according to ligand concentration. This study confirmed the use of expanded bed adsorption to fractionate the PAT fraction from PFJ while retaining its functionality (Straetkvern ei al. , 1999). However, the adsorbent material for this method is expensive for general processing of waste streams. The present example constitutes the first study on the pilot plant scale up for the recovery of potato proteins from by-products (PFJ and potato pulp) using (NH ₄ ) ₂ S0 ₄ and commercial multi-enzymatic system, Depol 670L.

EXAMPLE V - COMPARATIVE COMPOSITIONAL ANALYSIS OF POTATO PROTEIN

ISOLATES

The protein isolates prepared in Example IV were further characterized.

Protein content. Nitrogen content was determined using Leco® TruSpec N (Leco Corporation, Michigan, USA). Prior to nitrogen determination, samples were freeze dried and stored at -80°C. Nitrogen content was multiplied by a factor of 6.25 to determine the total crude protein content. Based on the protein content, the yield was calculated as: A) Protein Recovery Yield (%) = (Total protein content isolate / Total protein content PFJ) x 100

B) Extract Recovery Yield (g/ kg potato) = (Mass of recovered extract/ Mass of initial potato) x 100

Carbohydrate content. Sugar content of the PPC UF, PPI AS, and PPC Enz was determined using phenol-sulfuric acid colorimetric assay. 400 μΙ_ of potato protein suspensions (10 mg/ ml_) were suspended with 10 μΙ_ of 80% phenol and 1 ml_ of sulphuric acid. Mixtures were vigorously vortexed and incubated at 25°C for 15 min. The absorbance at 480 nm were measured using a Beckmann Coulter spectrophotometer DU 800 (Beckmann Clouter, Fuerllerton, CA).

The molecular weight distribution of the carbohydrate present in the PPI and PPCs were analyzed by high-performance size exclusion chromatography using a Waters HPLC system Model 25P (Waters Corporation, Maine, USA) equipped with a refractive index detector. Three columns were connected in series at 25°C (TSK G3000 PWXL, TSK G4000 PWXL, and TSK G5000 PWXL, Tosoh Bioscience, Montgomeryville, PA). An isocractic elution at a flow rate of 0.4 mL/min using 5.84 g/L sodium chloride was employed. A standard curve was developed using dextrans with known molecular weights of 50, 150, 270, 410, and 670 kDa.

Structural characterization of potato protein isolates via Sodium Dodecyl Sulphate Polyacrylamide Gel Electrophoresis (SDS-PAGE). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed according to the method of Laemmli (1970) using a Protean II electrophoresis apparatus (Bio-Rad Laboratories, Hercules, CA, USA), and Ready Gel precast gels (Any kDa™) polyacrylamide Tris-HCI. Protein samples of 5mg/mL were dispersed in 1 mL of Laemmli sample buffer (Bio-Rad) with the addition of 2% SDS and 5% β-mercaptoethanol and vortexed. The dispersions were heated at 100°C for 5 min and centrifuged before loading. After centrifugation for 5 min at 1 000g, varying volumes of sample solution were loaded in each well to yield 10, 15, and 20 pg protein and gels run at constant voltage (200 V) in Tris-glycine buffer containing 0.1 % SDS. Gels were stained with Coomassie Brillant Blue™ R-250. A Precision Plus Protein Dual Xtra™ Standards (Biorad) was used as a molecular marker, ranging in molecular weights from 2 to 250 kDa.

Structural characterization of potato protein isolates via size exclusion high performance liquid chromatography. HPLC size exclusion chromatography was conducted as previously described by Achouri et al. (2010). Briefly, protein samples were dissolved in 2.12 g/L phosphate buffer (pH 7.8) at a concentration of 5 mg/mL. An Agilent Bio SEC-3 column (7.8 x 300 mm) connected to an Agilent- 1200 Series HPLC system (Agilent Technologies, Mississauga, ON, Canada) was used. The protein solution (50 μΙ_) was loaded on the column and eluted with 2.12 g/L phosphate buffer containing 8.77 g/L NaCI (pH 7.8) at a flow rate of 1 mL/min. The elution was monitored at 280 nm. Mixed gel filtration standards comprising thyroglobulin (670 kDa), γ-globulin (158 kDa), ovalbumin (44 kDa), myoglobin (17 kDa) and vitamin B12 (1.35 kDa) from Bio-Rad Laboratories (Mississauga, ON, Canada) were used to estimate the molecular masses of the samples.

Structural characterization of potato protein isolates via Differential Scanning Calorimetry. Calorimetric measurements were taken using a 2910 modulated differential scanning calorimeter (TA Instruments Inc., New Castle, DE, USA). The instrument was calibrated using indium as a standard. Potato protein samples (10 mg) were dispersed in D ₂ 0- phosphate buffer prepared at pH 7. Aliquots (20 L) of the solutions were hermetically sealed in aluminum DSC pans and heated at a scan rate of 10°C min under helium through the range of 20 to 1 10°C. A pan filled with 20 pL D ₂ 0-phosphate buffer prepared at pH 7 was used as a reference. Data were analyzed using the Universal Analysis software from TA Instruments. Transition temperatures (To, onset temperature of denaturation; Td, maximum temperature of denaturation) and enthalpies (ΔΗ; areas below the endothermic curves in joules per gram of dry weight) were measured. All samples were run in triplicate.

Structural characterization of potato protein isolates via Fourier transform infrared spectroscopy. For FTIR analysis, potato protein dispersions were prepared by dissolving 10 mg protein in D ₂ 0-phosphate buffer prepared at pH 7. D ₂ 0 was used instead of H ₂ 0 because of its greater transparency in the infrared amide I' region. Aliquots (10 μΙ) of each sample were held in an IR cell with a 25 pm path length, between two CaF ₂ windows and infrared spectra were recorded at 25°C with an Excalibur™ FTIR spectrometer (system FTS 3000) equipped with a deuterated triglycine sulfate (DTGS) detector (Bio-rad Laboratories, Cambridge, MA). The spectrometer was purged with dry air for 10 min before recording the spectrum. A total of 256 scans were averaged at 4 cm ^"1 resolution. The signal-to-noise ratio was > 20000: 1 . Deconvolution was performed using the Digilab Merlin™ FTIR software (version 3.4) with a half-bandwidth of 13 cm ^"1 and an enhancement factor of 2.4 ( auppinen, Moffatt, Mantsch, & Cameron, 1981). The spectra were baseline-corrected between 1750 and 1595 cm ^"1 and normalized by dividing the absorbance value at each wavenumber in this range by the integrated area over this range, with the use of OMNIC software (Nicolet, Thermo Electron Cooperation). Band assignment was assigned according to (Kong & Yu, 2007) where bands at 1637± 3.0 and 1675 ± 5.0 cm-1 as β-sheets, 1645 ± 4.0 cm ^"1 as random coil, 1653 ± 4.0 cm-1 as a-helix, and 1671 ± 3.0 cm ^"1 and 1689 ± 2.0 cm-1 as β- turns. According to Pots et al. (1998) the band at 1618 cm ^"1 was assigned to intermolecular β-sheet which is associated with aggregation.

Molecular weight distribution of protein isolates and concentrate. In order to understand the relative proportions of each potato protein fraction, SDS-PAGE gel has been analyzed using the Alpha View software and the results are outlined in Table 8. Protein samples obtained by three extraction techniques showed similar proportions of PAT (21.55 - 25.57%), with PPI AS having the highest content, followed by PPC Enz. All three types of PPI and PPC were composed of a relatively lower proportion of Pis ranging from 21-25 kDa, whereas higher proportions were obtained for both 15-20 kDa and <15 kDa. PPC UF, in particular had a high proportion (24.51 %) of high molecular weight proteins when compared to the other two techniques.

A previous study outlined the protein profile of (NH ₄ ) ₂ S0 ₄ at 60% saturation on laboratory scale where a high proportion of PAT was extracted 36.1 % (Waglay, arboune, & Alii, 2014), as compared to 24.5% for the pilot scale process. A similar distribution was obtained for both Pis and high molecular weight proteins of 51 .5/ 1 1.2% (Waglay, Karboune, & Alii, 2014) and 63.4/1 1 .1 % for laboratory and pilot plant processes, respectively.

Depol 670L multi-enzymatic extraction on laboratory scale resulted in a protein concentrate with higher proportion of PAT (49.78%), lower Pis (27.02%), and similar proportion of high molecular weight proteins (13%; data not shown). Whereas, pilot-scale resulted in proportions of 24.53% PAT, 62.1 % Pis, and 13.4% high molecular weight protein (Table 8).

To estimate the molecular weight distribution, size exclusion high performance liquid chromatography (HPLC) was performed for PPC UF, PPI AS, and PPC Enz, and the results are shown in Figure 8. HPLC results for all potato protein samples matched well with the results obtained by SDS-PAGE (Table 8) where extraction processes resulted in relatively similar PAT profile, with large variations in the Pis being recovered. PPIs and PPC showed a wide variation between 1 .35 and 17 kDa indicative of the Pis. As well as a good distribution around 44 kDa indicative of PAT. In its native form, PAT is largely present in its dimer form of 80 kDa which is likely the peak between 44 kDa and 158 kDa. All extracts contained high molecular weight proteins, indicative of the peaks above 158 kDa. Furthermore, despite the diafiltration steps (molecular weight cut off of 5 kDa), a few peaks were obtained at <1 .35 kDa, which is indicative of peptides that have potentially interacted with polyphenols present in the by-products which would render them insoluble (Straetkvern et al., 1999). Due to the similarity between both protein isolates PPC UF and PPI AS, the thermal denaturation and functional properties will compare protein isolate PPI AS with protein concentrate PPC Enz.

Thermal and structural denaturation of protein isolate and concentrate The thermal denaturation of protein isolate (PPI AS) and concentrate (PPC Enz) were investigated. As shown in Table 8, the structural changes are not related to a single protein, but rather a complex mixture. Moreover, several competing interactions such as with the present polyaccharides and hydrolysed saccharides could also provide additional changes to the conformation of the protein (Boye et al. , 1996).

Figure 9 shows the heat flow profiles collected from differential scanning calorimetry of the concentrate and isolate. As shown on Figure 9, both exhibited a similar differential scanning calorimetry profile, with the decline around 70-75°C. Contrarily purified PAT examined by Pots et al. (1998) showed a sharp decline between 50-60°C. Whereas, the most abundantly known Pis found in potato (serine Pis) exhibited a denaturation temperature between 62- 69°C (Pouvreau et al. , 2005). The differences in denaturation temperature could be due to the presence of Pis, which would have a stabilizing effect on the available PAT. A previous study conducted by Koppleman er a/. (2002), found that aggregated PAT in the presence of Pis have a stabilizing effect limiting the fully irreversible denaturation which was examined in their absence. Table 9 shows the denaturation temperature (T _d ) and the calorimetric enthalpies (ΔΗ). PPC Enz resulted in the highest T _d of 75.86°C, followed by PPI AS with T _d of 72.81 °C. The differences in T _d could be due to the presence of carbohydrates with varying sizes. The presence of these sugars has been shown to effect the denaturation temperature (Boye et al., 1996). PPC Enz resulted in the higher proportion of polysaccharide content (data not shown) when compared to PPI AS. Expectedly, PPC Enz would contain smaller sugar components. The multi-enzymatic system Depol 670L is found to contain several glycosyl-hydrolases activities which cleave the polysaccharide network surrounding the proteins within the potato cell wall rendering smaller saccharides. Similar to these results, varying sizes of sugars have been found to have a different stabilizing effect on bovine serum albumin by increasing the denaturation temperature as the sugar size decreases (Boye ef al. , 1996).

Similar to Pots ef al. (1998), a different profile before and after denaturation can be seen for PPC Enz and PPI AS indicating a difference in conformational structure following denaturation. The higher enthalpy of 0.85 J/g was obtained for PPC Enz, whereas the lower 0.38 J/g for PPI AS (Table 9). These difference could be explained by the conformational changes that have taken place prior to denaturation either between protein-protein or protein- polysaccharide interactions. Indeed, that aggregation and disrupting the protein's hydrophobic interactions are an exothermic process that can result in a lower enthalpy (Boye et al., 1996).

Table 9. Thermodynamic Data from Differential Scanning Calorimetry Profile of Potato Protein Extracted on Pilot Plant Scale by Ammonium Sulphate, and Multi-Enzymatic System (Depol 670L).

aAmmonium sulphate performed at 60% saturation, abbreviated PPI AS

bMulti-enzymatic system used was Depol 670L, abbreviated PPC Enz

As part of the study of the effect of the extraction methods on the stability of protein isolate and concentrate, their secondary structural components were assessed from 25-95°C using FTIR (Figure 10). Both isolate and concentrate exhibited varying conformational profiles at 25°C (Figure 10); PPC Enz possessed a prominent aggregation peak (1618 cm ^"1 , Figure 10B) prior to thermal denaturation. Possible explanations, would be that the addition of the enzymes led to slight pH variation during extraction, where the pH was not controlled. This could lead to ideal conditions below PAT's isoelectric point 4.9 where the potato polysaccharides will carry an opposite charge to the protein molecule forming a complex or aggregative phase separation (Doublier et al. , 2000). This aggregation band was minimally shown for PPI AS as proteins extracted in this manner have been structurally comparable to native potato protein (van Koningsveld, et al., 2001).

In summary, the C=0 stretching vibrations are responsible for the absorption of energy by the amide backbone. These vibrations are influenced by hydrogen bonding and are found in the FTIR spectrum Amide I region from 1600-1700 cm ^"1 . Commonly, a second-derivative procedure is used to separate overlapping structural components found in this region (Matheus et al., 2006; Haris, 2013). FTIR spectra (Figure 10A and 10B) demonstrate with all extracting agents a significant effect in the Amide I maximum shifting to higher frequencies following heating. For both isolate and concentrate, at 25°C, a maximum was exhibited at 1638 cm ^"1 , pertaining to predominately β-sheet, followed by a peak at 1654 cm ^"1 , which is associated with o-helical structures. Indeed, native PAT has been shown to be composed of 45% β-strand, 33% a-helix, and 15% random coil (Pots et al., 1998). In additional potato serine Pis, the most abundant in the tuber, have been studied to belong to (3— 11 protein subclass, which is composed primarily of of β-sheet or β-turns (Pouvreau et al. , 2005). In both cases the thermal denaturation curves (Figure 10A and 10B) show that as temperatures rise above 55°C, conformational changes of the protein mixture's secondary structure occurs. Above 55°C a pronounced aggregation peak results (1618 cm ^"1 ) and a drop in peak separation ranging from 1650-1630 cm ^"1 . This is in agreement with previous findings of Pots et al. (1998) and van Koningsveld et al. (2001) where PAT begin to denature at 50- 55°C, leading to breakdown in alpha-helical (1654 cm ^"1 ) and beta-stranded regions (1634 cm ^" ) and increased occurrence of aggregation (1684 or 1618 cm ^"1 ). Despite the difference in thermal denaturation temperature determined for DSC ranging from 72.81 -75.86°C and FTIR results show break down of the secondary structure at around 55°C, The literature shows that native PAT denatures at 55°C (Pots et al , 1998) and most Pis denature at about 70°C (van Koningsveld er a/. , 2001). Therefore the results of the DSC measurements may be more closely related to the Pis present in the mixture. Contrarily, the FTIR results may be more closely related to the PAT present in the mixture as shown by the maxmimum peak in the β- strand region (1634 cm ^"1 ) as well as, the poor peak separation following heat treatments at 55°C. In addition, the higher DSC measurements could be attributed to the predominant conformation of β-sheet structure in all protein extracts, which is often associated with a higher T _d (Shevkani et al. , 2015).

The thermal transition curves were developed by plotting the intensities of frequencies versus temperature, the intensities of frequencies consisted of: the second derivative spectrums for intensity of increasing (1618 cm ^"1 , predominantly aggregation), and the most prominent structure for both PAT and Pis, the second derivative spectrum for intensity of decreasing (1634 cm ^"1 , predominantly β-strand). The cross section of these curves could be computed as the T _m (FTIR) (Matheus et al., 2006). The FTIR spectrums of the extracting agents, PPI AS (Figure 10C) results in T _m (FTIR) for 1634cm ^"1 at 49.6°C. Finally, PPC Enz a greater conservation is exhibited for the β-sheet region Tm(FTIR) 54.6°C (Figure 10D). Based on the protein profile (Table 8) the PAT to Pis ratio range from 0.395-0.403. A previous study conducted, demonstrated that PAT in the presence of Pis aggregate together upon heat treatments (Koppelman et al. , 2002). Other studies have further demonstrated the stabilizing effect of sugars which have been shown to shift the transition temperature by ~1.5°C (Boye & Alii, 2000). To date, no study compares the structural changes of the complex mixture of potato protein isolates based on different extracting agents. Moreover, the analysis was done on a complex mixture of proteins with varying salt and sugar concentration, where many interactions result. Therefore these results are more representative on a large industrial scale to examine the denaturation conditions of complex protein mixture for their potential application in food systems. EXAMPLE VI - COMPARATIVE FUNCTIONAL ASSESSMENT OF PILOT PLANT POTATO

PROTEIN ISOLATES

Emulsifying activity. Emulsifying activity index (EAI) and emulsion stability index (ESI) were determined by the turbidimetric method of (Cameron et al. , 1991). 1.5 mL of corn oil was added to 4.5 mL of 0.5% (w/v) potato protein solution prepared in 2.12 g/L phosphate buffer (pH 7), after which the mixture was homogenized at 20 000 rpm at room temperature for 1 min with a PT 2100 Polytron homogenizer (Kinematica AG, Littau-luzern, Switzerland). 250 pL of the emulsion was taken out from the bottom at different times (0 and 15 min) and diluted with 50 mL of 0.1 % sodium dodecyl sulfate solution. The absorbance of the diluted emulsion was, then, determined at 500 nm with a Cary 300 Bio, UV-Visible Spectrophotometer (Varian Canada Inc., Quebec, Canada). All measurements were assessed in triplicate. The Emulsifying activity index (EAI) and Emulsifying stability index (ESI) were calculated using the following equations:

where, c is the initial protein concentration in PPI or PPC solution (g protein/mL), I is the optical path (0.01 m), φ is the oil volume fraction used to form the emulsion, dilution factor is 200, t is 15 min, and A ₀ and A ₁₅ are the absorbance of the diluted emulsions at 0 and 15 min, respectively.

Emulsifying properties. Table 10 presents the emulsifying properties of both extracting techniques PPI AS and PPC Enz. The emulsifying activity index (EAI) is expressed both per gram of extract as well as gram of protein in the extract. When examining, EAI by gram extract we can see that PPC Enz with 13.4 m ² / g extract was lower when compared to PPI AS. As previously shown with the secondary structure characterization (Example V), PPC Enz, resulted in small aggregation peaks (1618 cm ^"1 ) at 25°C, prior to thermal treatments. This correlated well with EAI (m ² / g extract) which show lower results obtained, depicting that aggregation leads to a loss of protein at the oil-water interface thereby reducing the EAI. However when taking into account the gram of protein per gram powder, the EAI for PPC Enz was significantly higher (33.5 m ² / g protein) than for the isolate. These differences could be attributed to the varying percentage of protein found in the isolate (Table 8). A similar trend was observed for sweet potato protein, almond protein, wheat gluten, and acidic subunits of soy 1 1 S globulin, where it was found that the EAI decreased with increasing protein concentration.

Table 10. Emulsifying Properties of potato protein isolates extracted on pilot plant by ultrafiltration, ammonium sulphate saturation, and multi-enzymatic system Depol 670L.

aEmulsifying activity index.

bEmulsifying stability index.

'Ammonium sulphate performed at 60% saturation, abbreviated to PPI AS.

g Multi-enzymatic system used was Depol 670L, abbreviated PPC Enz

hStandard deviation taken of three triplicates

The results obtained could be due to the ease in diffusion of the protein to the oil-water interface at low concentrations, leading to increased development of new oil droplets, thereby increasing EAI. Conversely, at higher protein concentrations diffusion is less likely to occur due to the activation-energy barrier for the protein to adsorb on the interface. In addition for PPC Enz, the presence of small sugars may have increased the protein's solubility, therefore enhancing the protein's participation to adsorb at the oil-water interface. As well, for both PPI AS and PPC Enz a larger proportion of Pis < 15 kDa is present (Table 8), which could lead to increase flexibility of these small proteins at the interface, leading to higher EAI. Interestingly a positive correlation is seen with the thermal properties (T _d and ΔΗ) determined by DSC and the EAI, indicative of the effect of the structural composition. Where the conformational changes of the protein tertiary and quaternary structure allow for the protein to easily absorb to the oil-water interface.

Contrarily to the high EAI obtained for low protein concentration, high emulsifying stability index (ESI) values were obtained at high protein concentrations. The ESI was higher for PPI AS (18.0 min), as compared to PPC Enz (16.7 min; Table 10). At high protein concentrations the proteins at the interface of the oil-water droplets will be increased thus decreasing the droplet flocculation, creaming, and coalescence through electrostatic repulsion. Whereas in the case of PPC Enz with a lower protein concentration (Table 8) the electrostatic repulsion among protein molecules is to low which leads to droplet flocculation and in turn low ESI.

Based on the results, and as compared to other isolates, potato protein have better EAI values of approximately 13.5 m ² /g powder and ESI about 17.8 min whereas soy protein isolates EAI is 10.9 m ² / g powder and ESI 0.8 min. On the other hand for 0.4% sweet potato protein solutions were found to have an EAI of 50 m ² /g and ESI of 65 min. Therefore potato protein possess superior emulsifying properties to soy protein, but seem inferior to sweet potato proteins.

Foaming ability. The procedure proposed by Waniska and Kinsella (1979) with some modifications was used for measuring foaming properties. The protein was dispersed in 2.12 g/L phosphate buffer (pH 7) at concentration of 0.5% (w/v) with stirring for 10 min at room temperature (25 °C). The protein solution (15 mL) was then injected into the sparging chamber of a water-jacketed glass condenser via the septum-stoppered inlet. Nitrogen gas was sparged into the protein solution until the foam chamber (55 mL) was filled with foam, while simultaneously maintaining the volume of liquid in the sparging chamber by addition of protein solution. The required time to form 55 mL of foam, and the volume of protein solution added were recorded. After 5 min, the volume of liquid drained from the foam was also noted. Several parameters were used to predict foaming properties as follows: G, (%), which is the percent of gas entrapped in the foam; FE (%), which is the percent foam expansion and R ₅ (%), which represents the percent of liquid retained in the foam after 5 min. The following equations were used to determine the foaming parameters:

fill Fifrtfls ^r V|_

&E - ¾2a _s lie R _& - ¾ K it* where, G, is percentage of gas entrapped in 55 mL of foam, FR is gas flow rate (mL/ min nitrogen), T _f is time to fill the column with foam, V ₀ is the initial volume of sample in the jacketed condenser (15 mL), V, is the volume of liquid injected, V _d is the volume of liquid drained from the foam after 5 min, V _r is the volume of liquid retained in the foam after 5 min, R ₅ is the percentage of liquid retained from the foam after 5 min, FE is the percent of foam expansion, and the total volume of the jacketed glass condenser is 70 mL.

Table 1 1 outlines the foaming properties obtained for PPI AS and PPC Enz. Elevated G, and FE are correlated to increased foam capacity and expansion, whereas increased R ₅ is related to foam stability. PPC Enz resulted in the highest G, and FE which could be related to high foam capacity and expansion. This higher foam capacity and expansion for PPC Enz could be due to the increased presence of sugars as shown by phenol-sulfuric acid (data not shown). The presence of sugars within the suspension leads to increase solubility of the protein, therefore rendering the protein more available to participate in foaming. This phenomenon was also encountered by Partsia and Kiosseoglou (2001), who described that the presence of complexing agent carboxymethyl cellulose with potato proteins decreases surface tension at the air and water interface. This allows for more air to be incorporated as well as enables the polysaccharides to help with foam stability by decreasing the occurrence of drainage (Partsia & Kiosseoglou, 2001). Another explanation for the differences of foaming characteristics amoung PPI and PPC is the proportion of PAT and Pis. Ralet and Gueguen (2001) found that PAT was able to form very stable and resistant foams, conversely, Pis 16- 25 kDa formed very unstable foams which broke rapidly (Ralet & Gueguen, 2001). PPI AS resulted in a high relative proportion of PAT (25.57%) and Pis 16-25 kDa (31.69%), comparatively to PPC Enz 24.53 and 22.6%, for PAT and Pis, respectively. Therefore possible explanations for the lower foaming expansion and stability for PPI AS could be due to the higher presence of PI (16-25 kDa), leading to foam breakage, despite the relatively high proportion of PAT.

Table 1 1. Foaming Characteristics of potato protein isolates extracted on pilot plant by ultrafiltration, ammonium sulphate saturation, and multi-enzymatic system Depol 670L

cPergentage of gas entrapped in 55 mL of foam.

dPercentage of liquid retained from the foam after 5 min.

ePercent of foam expansion

Like EAI, foaming properties are also governed by secondary conformation of the proteins, where the ease in the protein unfolding revealing its hydrophobic nature at the air-water interface will help in the foam development. A positive correlation was seen between thermal properties (T _d and ΔΗ) and foaming capacity, where PPC Enz showed the highest foam capacity (high G, and FE) and high thermal properties (T _d and ΔΗ).

Based on this study when compared to other protein isolates, potato protein have weaker foam expansion (approximately 392%) and better foam stability (about 58.3%) as compared to soy protein (foam expansion and stability of 532% and 41 %, respectively). While the invention has been described in connection with specific embodiments thereof, it will be understood that the scope of the claims should not be limited by the preferred embodiments set forth in the examples, but should be given the broadest interpretation consistent with the description as a whole.

REFERENCES

Achouri, A., Boye, J. I., Belanger, D., Chiron, T., Yaylayan, V. A., & Yeboah, F. K. (2010). Functional and molecular properties of calcium precipitated soy glycinin and the effect of glycation with κ-carrageenan. Food Research International, 43, 1494-1504.

Boye, J. I., Ismail, A. A., & Alii, I. (1996). Effects of physicochemical factors on the secondary structure of β-lactoglobulin. Journal of Diary Research, 63, 97-109.

Boye, J. I., & Alii, I. (2000). Thermal denaturation of mixtures of a-lactalbumin and β- lactoglobulin: a differential scanning calorimetric study. Food Research International, 33, 673-682.

Cameron, D. R., Weber, . E., Idziak, E. S., Neufeld, R. J., & Cooper, D. G. (1991). Determination of Interfacial Areas in Emulsions Using Turbidimetric Droplet Size Data: Correction of the Formula for Emulsifying Activity Index. Journal of Agricultural and Food Chemistry, 39, 655-659.

Doublier, J.-L, Gamier, C, Renard, D., & Sanchez, C. (2000). Protein-polysaccharide interactions. Current Opinion in Colloid & Interface Science, 5, 202-214.

Haris, P. I. (2013). Probing protein-protein interaction in biomembranes using Fourier transform infrared spectroscopy. Biochimica et Biophysica Acta, 1828, 2265-2271 .

Kirsten, W. J., & Hesselius, G. U. (1983). Rapid, automatic, high capacity Dumas determination of nitrogen. Microchemical Journal , 529-547.

Kong, J., & Yu, S. (2007). Fourier Transform Infrared Spectroscopic Analysis of Protein Secondary Structures. Acta Biochimica et Biophysica Sinica, 39(8), 549-559.

Koppelman, S. J., van Koningsveld, G. A., Knulst, A. C, Gruppen, H., Pigmans, I. G., & de Jongh, H. H. (2002). Effect of heat-induced aggregation on the IgE binding of patatin (Sol 1 1) is dominated by other potato proteins. Journal of Agricultural and Food Chemistry, 50, 1562- 1568.

Laemmli, U. (1970). Cleavage of Structural Proteins during the Assembly of the Head of Bacteriophage T4. Nature , 680-685.

Matheus, S., Friess, W., & Mahler, H.-C. (2006). FTIR and nDSC as Analytical Tools for High-Concentration Protein Formulations. Pharmaceutical Research, 23, 1350-1363. Partsia, Z., & Kiosseoglou, V. (2001). Foaming properties of potato proteins recoveres by complexation with carboxymethylcellulose. Colloids and Surfaces B: Biointerfaces, 21 , 69-74.

Pots, A. M., De Jongh, H. H., Gruppen, H., Hamer, R. J., & Voragen, A. G. (1998). Heat- Induced Conformational Changes of Patatin, The Major Potato Tuber Protein. European Journal of Biochemistry, 252, 66-72.

Pouvreau L, Gruppen H, van Koningsveld G, van den Broek LA, Voragen AG. Conformational stability of the potato serine protease inhibitor group. J Agric Food Chem. 2005 Apr 20;53(8):3191 -6.

Pouvreau, L., Gruppen, H., van Koningsveld, G., van den Broek, L. A., & Voragen, A. G. (2005). Tentative assignment of the potato serine protease inhibitor group as beta-ll proteins based on their spectroscopic characteristics. Journal of Agricultural and Food Chemistry, 52(25), 7704-7710.

Ralet, M.-C, & Gueguen, J. (2001). Foaming Properties of Potato Raw Proteins and Isolated Fractions. Lebensmittle-Weissenschaft Und Technologie, 34, 266-269.

Seo S, L'Hocine L, Karboune S. Allergenicity of potato proteins and of their conjugates with galactose, galactooligosaccharides, and galactan in native, heated, and digested forms. J Agric Food Chem. 2014 Apr 23;62(16):3591 -8.

Shevkani, K., Singh, N., Kaur, A., & Rana, J. C. (2015). Structural and functional characterization of kidney bean and field pea protein isolates: A comparative study. Food Hydrocolloids, 43, 679-689.

Strand, L. L., Rechtoris, C, & Mussell, H. (1976). Polygalacturonase Release Cell-Wall- Bound Proteins. Plant Physiology , 722-725.

Straetkvern, K. O., Schwarz, J. G., Wiesenborn, D. P., Zafirakos, E., & Lihme, A. (1999). Expanded Bed Adsorption for Recover of Patatin from Crude Potato Juice. Bioseparation, 7, 333-345.

Thomassen, L. V., & Meyer, A. S. (2010). Statistically designed optimisation of enzyme catalysed starch removal from potato pulp. Enzyme and Microbial Technology , 297-303. van Koningsveld, G. A., Gruppen, H., de Jongh, H. H., Wijngaards, G., van Boekel, M. A., Walstra, P., et al. (2001). Effects of pH and Heat Treatments on the Structure and Solubility of Potato Proteins in Different Preparations. Journal of Agricultural Food Chemistry , 4889- 4897.

Waglay, A., Karboune, S., & Alii, I. (2014). Potato protein isolates: Recovery and characterization of their properties. Food Chemistry, 142, 373-382. Waniska, R. D., & Kinsella, J. E. (1979). Foaming Properties of Proteins: Evaluation of a Column Aeration Apparatus using Ovalbumin. Journal of Food Science, 44, 1398-141 1 .

Wu S, Murphy PA, Johnson LA, Reuber MA, Fratzke AR. Simplified process for soybean glycinin and beta-conglycinin fractionation. J Agric Food Chem. 2000 Jul;48(7):2702-8.