PROCESS FOR RECOVERING PROTEINS FROM WHEY PROTEIN CONTAINING FEEDSTOCKS
FIELD OF THE INVENTION
The present invention relates to processes for recovering acidic peptide fractions, such as CMP (caseino-macropeptide), and β-lactoglobulin from whey protein containing feedstocks using anion exchangers.
BACKGROUND OF THE INVENTION
Cation and anion exchange processes have been used to make various purified protein or peptide products from milk raw materials such as skim milk and whey since the late 1970's.
Whey protein isolate (WPI), is produced as a result of whey proteins being adsorbed from the milk raw material by the ion exchanger and then being recovered from the ion exchanger by desorption (elution) after first washing away the treated milk raw material. The latter still contains the lactose, minerals and fat content of the original milk raw material, so the WPI is typically about 90-94% protein. Microfiltration(MF) is also used to produce WPI. The MF membrane retains the fat content of the raw material but passes the whey protein, lactose and minerals. Ultrafiltration(UF) is then used to separate the lactose and minerals from the whey protein. WPI produced by MF and UF is thus high in protein content and, like ion exchange WPI, very low in fat.
With cation exchangers, a pH < 5 is usually used to recover the maximum amount of protein as WPI (cation WPI). With anion exchangers a pH > 5, usually > 6, is used to produce the WPI (anion WPI).
Surprisingly, we have found that solutions of the WPI produced from the anion exchanger are not heat stable in the acidic region of pH 3.5-4.0. This
is the preferred pH range to avoid the sourness that can occur at pH values below 3.5 for protein fortification of beverages like fruit drinks. Unstable WPI solutions are unacceptable as they give a cloudy appearance or form a sediment.
We have further surprisingly discovered that if acidic peptides and proteins are absent from the anion WPI then it is heat stable under such acidic conditions and can be used for a wider range of applications.
One such group of acidic peptides is CMP (caseinomacropeptide, also known as GMP (glycomacropeptide) and CDP (casein derived peptide)), which is a complex mixture of macropeptides found in sweet wheys (rennet and cheese wheys), derived from kappa-casein by the action of the enzymes chymosin and/or pepsin. CMP is rich in sialic acid and has a number of potential therapeutic uses, as well as having functional properties which make it very useful as an ingredient in food compositions. Thus, a process to remove CMP from whey to produce a heat and acid stable WPI also produces an isolated CMP fraction. One important utility of CMP is as a nutritional component for use in the diets of persons suffering from phenylketonuria as it has a very low phenylalanine content. In addition, sialic acid rich components are present in human milk at a level of about 3-5 times higher than in bovine milk and are considered to act as an infection protection factor for infants. They are also important for the development of the brain. Therefore it would be useful to be able to produce sialic acid rich food products on an industrial scale.
A number of prior art processes are known for isolating sialic acid rich CMP. US 5,270,462 describes a process for manufacturing sialic acids-containing composition from sweet wheys by adsorbing the major whey proteins on a cation exchanger at a pH of 2-5 and the sialic acids, such as CMP, are selectively recovered as an exchanger-passed solution. Further processing gives a product having 2.9-4.8% sialic acid.
US 5,216,129 describes the preparation of CMP from a whey protein concentrate by heat denaturation and precipitation of the whey proteins.
The supernatant is concentrated by ultrafiltration and residual soluble proteins precipitated by the addition of ethanol to the retentate. Further processing gave a CMP product containing 10.9% sialic acid. Such a process
suffers the disadvantage that the whey proteins are denatured, thereby destroying their solubility and functional properties and losing much of their economic value.
Saito et al in Animal Science and Technology, 65, 624-630, 1994 describes the use of affinity chromatography to separate CMP into asialo- and sialo- CMP components. The latter was found to contain about 10% sialic acid. Such a procedure required the prior preparation of CMP by a separate procedure.
Tanimoto et at in Bioscience, Biotechnology and Biochemistry, 56, 140-141, 1992 describes a process in which a crude CMP powder was first prepared and purified by anion-exchange chromatography on Q-Sepharose. A solution of the crude CMP was loaded onto the Q-Sepharose column at pH 7.5 to adsorb the CMP and then eluted with a linear gradient of sodium chloride. The fractions containing CMP were combined to give a purified CMP with 7.6% sialic acid. Such a process is possible only with CMP that has already been separated from whey proteins and would not be suitable for large scale commercial use.
Kawakami et al in Milchwissenschaft, 47, 688-693, 1992 similarly describes the use of anion-exchange chromatography of CMP that had previously been isolated from whey. In this case though they collected separately the CMP components detected in the gradient elution as seven separate peaks. These fractions had sialic acid (NeuAc) contents ranging from 0.4 to 18.9%. Their anion exchange process however is suitable only for small scale laboratory work and again, as with the preceding two publications, it requires a crude CMP to be prepared first.
Processes for isolating CMP from rennet or cheese whey or whey concentrates using anion exchangers are described in GB 2188526 and GB 2251858. At a pH of less than about 5 most of the other whey proteins are positively charged and do not bind to anion exchangers.
The process described in GB 2188526 involves contacting whey protein containing material with an anion exchanger at a pH of 4 to 6, more particularly at a pH of 4.8 to 5.0, and then eluting the bound proteinaceous material, which, in the case of sweet wheys, is mainly CMP.
The process described in GB 2251858 differs from the process described in GB 2188526 in that the anion exchanger is contacted with the milk raw material at a pH of 4 or below instead of a pH of 4 to 6. In the applicants' experience this produces a CMP sub-fraction which is highly glycosylated, particularly acidic and still negatively charged at a pH of 4 or below.
WO 98/ 14071 also describes a method for purifying CMP, which involves contacting a CMP containing feedstock with an anion exchanger under conditions which adsorb the CMP, el ting the CMP, removing the impurities therefrom in one or more of a number of alternative ways, and recovering the purified CMP. The examples disclose that when an adsorption pH of 4.9 was used, a high yield of CMP was obtained and that it had a sialic acid content of 5.4% after further purification. A lower pH of 4.2 resulted in a decreased yield of CMP (45%) and increased sialic acid content (12.7%) after further purification. In a further example the total CMP was adsorbed initially but then eluted in two stages. Eluate 1 at pH 3.2 gave a CMP sub-fraction with only 2.6% sialic acid, while eluate 2 using salt gave a CMP product with 17.4% sialic acid after further purification.
However, none of the above references are concerned with producing a β- lactoglobulin enriched WPI that is acid and heat stable. In particular, processes for producing purified WPI, generally make no mention of CMP.
One exception is WO 95/ 19714 which indicates that it is necessary to first remove CMP, if present as in sweet wheys, for the efficient processing of the whey by anion exchangers. It is suggested that it be removed by preliminary treatment of the whey at pH 5 by anion exchanger as first described in GB 2188526 cited above. However, the processing referred to in WO 95/ 19714 involved the separation of β-lactoglobulin from α-lactalbumin and it was this separation which is rendered more efficient by the prior removal of CMP. This had been previously demonstrated in J. Dairy Research 52,167-181, 1985.
US 5434250 describes a process for preparing an α-lactalbumin enriched
WPC by the selective adsorption of β-lactoglobulin. CMP is mentioned as remaining partly with the α-lactalbumin in the exchanger passed solution. It
is removed therefrom by ultrafiltration technology to further enrich the WPC in α-lactalbumin. No mention is made of the recovery of β-lactoglobulin isolate fraction from the anion exchanger or its protein composition.
The above disclosures teach that sialic acid rich CMP may be isolated from whey. However none of the cited processes provide a simple efficient process for industrial scale production of CMP. In addition, no prior art process teaches the production of a β-lactoglobulin enriched WPI deplete in CMP which is acid and heat stable.
With the above background in mind, it is an object of the present invention to provide a process for the preparation of acid and heat stable β- lactoglobulin-enriched whey protein isolates and/ or CMP-enriched isolates using anion exchange and, optionally, microfiltration, which will at least provide the public with a useful choice.
SUMMARY OF THE INVENTION
Accordingly in a first aspect the present invention provides a process for producing a CMP isolate and an acid and heat stable β-lactoglobulin enriched WPI from a feedstock containing whey proteins including CMP and β-lactoglobulin, said process comprising the steps:
(a) contacting the feedstock with an anion exchanger under conditions in which CMP and β-lactoglobulin are both adsorbed by the anion exchanger;
(b) eluting the β-lactoglobulin and the CMP from the anion exchanger and contacting the eluate with a second anion exchanger under conditions whereby CMP is selectively adsorbed;
(c) collecting the β-lactoglobulin containing flow-through from the second anion exchanger; and
(d) eluting CMP from the second anion exchanger.
Preferably the feedstock is selected from the group consisting of sweet whey, UF retentate derived sweet whey, or reconstituted WPC (whey protein concentrate) derived from sweet whey. These feedstocks may optionally be modified so as to reduce their ionic strength by, for example, demineralisation or dilution with water. These feedstocks will also contain α-lactalbumin and BSA and other whey proteins as would be appreciated by a skilled person.
Preferably, in step (a) the feedstock is contacted with the anion exchanger at a pH of 5.5-8.5 and more preferably at a pH of about 6-7.5.
Preferably, in step (c) the eluate is contacted with the second anion exchanger at a pH of about 4-5.5 and more preferably at a pH of about 4.5- 5.
The whey proteins are eluted from the anion exchanger(s) using the following elution conditions:
In a batch process a salt concentration of 0-100 mM may be used and the pH lowered to 1.5-3 by the addition of acid.
In a column process the anion exchanger may be washed with 20-100 mM acid containing 0-100 mM salt.
Alternatively for step (d) a 500 mM salt solution may be used at any pH.
Alternatively, this process may be reversed, i.e. the whey protein containing feedstock may be contacted with a first anion exchanger at a pH of about 4.5-5 to adsorb acidic peptides and the flow-through contacted with a second anion exchanger at a pH of about 6-7.5 to adsorb both α-lactalbumin and β-lactoglobulin. The α-lactalbumin and β-lactoglobulin enriched WPI, which is acid and heat stable, is then eluted from the second anion exchanger, whilst the acidic peptides (including CMP) may be eluted from the first anion exchanger under conditions set out above.
Alternatively, in this reverse process CMP may be eluted at pH 2.5-4 from the first anion exchanger which elutes aglyco-CMP, and this eluate
combined with the WPI eluate of the second anion exchanger to form a β- lactoglobulin enriched WPI containing aglyco-CMP.
In an alternative embodiment, the second ion exchanger may be replaced with a cation exchanger wherein the flow-through from the first anion exchanger is contacted with said cation exchanger of a pH of 3-4.5 to adsorb both of α-lactalbumin and β-lactoglobulin. A α-lactalbumin and β- lactoglobulin enriched WPI, which is acid and heat stable, is then eluted from the cation exchanger. The aglyco-CMP eluted at pH 2.5-4 from the anion exchanger is combined with the WPI eluate of the cation exchanger to form an β-lactoglobulin enriched WPI containing aglyco-CMP which is acid and heat stable.
According to a second aspect the present invention provides a process for producing a CMP isolate and an acid and heat stable β-lactoglobulin enriched WPI from a feedstock containing whey proteins including CMP and β-lactoglobulin, said process comprising the steps:
(a) microfiltering (MF) the feedstock to produce a permeate (WPI stream);
(b) contacting the MF permeate of step (a) with an anion exchanger under conditions whereby CMP is selectively adsorbed;
(c) collecting and further processing the β-lactoglobulin flow-through from the ion exchanger to produce a β-lactoglobulin enriched WPI; and
(d) eluting CMP from the ion exchanger.
Preferably the feedstock is selected from the group consisting of sweet whey,
UF retentate derived sweet whey, or reconstituted WPC (whey protein concentrate) derived from sweet whey. These feedstocks may optionally be modified so as to reduce their ionic strength by, for example, demineralisation or dilution with water.
Preferably, in step (b) the MF permeate is contacted with the anion exchanger at a pH of about 4-5.5. More preferably at about pH 4.5-5.
The CMP is eluted at step (d) from the anion exchanger. In a batch process this may conveniently be achieved by using a combination of low pH and low salt concentration, e.g. pH 2 and 50 mM NaCl or a higher salt concentration under neutral conditions, e.g. pH ≥6 with 500 mM NaCl, or points in between.
Alternatively, the aforementioned process may be reversed, i.e., the whey protein containing feedstock may be first contacted with an anion exchanger at a pH of about 4-5.5, preferably about 4.5-5 to adsorb acidic peptides and the flow-through processed by microfiltration (MF) to produce a β- lactoglobulin enriched WPI which is acid and heat stable. The acidic peptides (including CMP) may be eluted from the anion exchanger under the conditions set out above.
In a third embodiment, the present invention provides a process of recovering CMP and β-lactoglobulin from a feedstock containing whey proteins including CMP and β-lactoglobulin, the process comprising the following steps:
(a) contacting the feedstock with an anion exchanger under conditions in which CMP and β-lactoglobulin are both adsorbed by the anion exchanger;
(b) eluting the β-lactoglobulin, optionally together with a fraction of the
CMP, from the anion exchanger; and
(c) eluting CMP, or the fraction of the CMP not eluted in step (b), from the anion exchanger.
Preferably, the feedstock is selected from the group consisting of sweet whey, UF retentate derived from sweet whey, or reconstituted WPC (whey protein concentrate) derived from sweet whey. These feedstocks may optionally be modified so as to reduce their ionic strength by, for example, demineralisation or dilution with water.
Preferably, in step (a) the feedstock is contacted with the anion exchanger at a pH of from about 5 to about 8.5. More preferably at a pH of about 6-7.5.
Preferably, in step (a) the adsorption is carried out under conditions so as to achieve a loading of protein on the anion exchanger of at least 50%, more preferably at least 75%, of the protein adsorption capacity of the anion exchanger.
In one preferred embodiment, step (b) comprises eluting β-lactoglobulin and substantially no CMP. Preferably the β-lactoglobulin is eluted under conditions of pH about 4 to about 5, and a salt concentration of from 0 to about 100 mM, to produce a first, β-lactoglobulin-containing, eluate.
In another preferred embodiment, step (b) comprises eluting β-lactoglobulin and a fraction of the CMP from the anion exchanger. Preferably the β- lactoglobulin and fraction of CMP are eluted under conditions of pH of less than 4, more preferably at a pH of from about 2.5 to less than 4, and a salt concentration of from 0 to about 50 mM. In this embodiment, step (b) will elute from the exchanger a fraction of less glycosylated or non-glycosylated CMP (aglyco-CMP) together with the β-lactoglobulin, to produce a first eluate containing β-lactoglobulin and a fraction of the CMP; and step (c) will produce a CMP-containing eluate enriched in more highly glycosylated CMP (glyco-CMP) rich in sialic acid.
In a fourth embodiment, the present invention provides a process for producing a sialic acid rich glyco-CMP isolate and a heat and acid stable β- lactoglobulin enriched WPI containing aglyco-CMP from a feedstock containing whey protein including β-lactoglobulin and CMP, the process comprising the steps:
(a) contacting the feedstock with a anion exchanger under conditions in which sialic acid rich CMP (glyco-CMP) is selectively adsorbed;
(b) collecting and further processing the β-lactoglobulin contained in the flow-through which further comprises non-sialic acid bearing CMP
(aglyco-CMP) to produce a β-lactoglobulin enriched WPI containing aglyco-CMP; and
(c) eluting the sialic acid rich glyco-CMP of step (a) from the ion exchanger.
Preferably the feedstock is selected from the group consisting of sweet whey, UF retentate derived sweet whey, or reconstituted WPC (whey protein concentrate) derived from sweet whey. These feedstocks may optionally be modified so as to reduce their ionic strength by, for example, demineralisation or dilution with water.
Preferably the whey protein containing feedstock is contacted with an anion exchanger at pH 2.5-4.5, preferably pH 3-4, to selectively adsorb the sialic acid rich glyco-CMP. The flow-through may be then treated by known processes (anion exchange or MF) to produce a WPI. This β-lactoglobulin enriched WPI will be acid and heat stable and will also contain the non-sialic acid bearing CMP, i.e. will comprise a β-lactoglobulin enriched WPI containing aglyco-CMP.
The process of this embodiment may be carried out in reverse order, i.e. wherein a WPI produced by MF or by anion exchange treatment of a whey protein containing solution including β-lactoglobulin and CMP, produced by known processes as would be appreciated by a person of skill in the art, is used as the feedstock.
Preferably the WPI feedstock is contacted with the anion exchanger at pH
2.5-4.5 preferably 3-4 to selectively adsorb glyco-CMP and the flow-through collected and further processed by UF to produce an acid and heat stable β- lactoglobulin enriched WPI containing aglyco-CMP. Glyco-CMP may be eluted from the anion exchanger as described above.
In a fifth embodiment the present invention provides a process for producing a sialic acid rich glyco-CMP isolate and an acid and heat stable β- lactoglobulin enriched WPI containing aglyco-CMP from a whey protein containing feedstock including CMP and β-lactoglobulin comprising the steps:
(a) contacting an anion exchanger with an excess of feedstock under conditions whereby sialic acid rich glyco-CMP is selectively adsorbed;
(b) further processing the flow-through by known processes to produce a WPI; and
(c) eluting the sialic acid rich glyco-CMP of step (a) from the anion exchanger.
Preferably the feedstock is selected from the group consisting of sweet whey,
UF retentate derived sweet whey, or reconstituted WPC (whey protein concentrate) derived from sweet whey. These feedstocks may optionally be modified so as to reduce their ionic strength by, for example, demineralisation or dilution with water.
Preferably the anion exchanger of step (a) is overloaded with respect to CMP so that non-sialic acid bearing aglyco-CMP which may initially bind to the exchanger is displaced by the more acidic sialic acid-rich glyco-CMP.
Preferably the whey protein containing feedstock is contacted with the anion exchanger of step (a) at a pH of about 4-6, preferably about 4.2-5.5.
The breakthrough solution at step (b) may be further processed by anion exchange or MF to produce a WPI by known processes known to a person of skill in the art.
This process may be reversed, i.e. a WPI produced by anion exchange or MF may be overloaded with respect to CMP onto an anion exchanger (at pH 4-6, preferably pH 4.2-5.5), whereby sialic acid rich glyco-CMP is selectively adsorbed onto the anion exchanger and β-lactoglobulin containing non sialic acid bearing aglyco-CMP is collected in the flow-through and further processed to produce a β-lactoglobulin enriched WPI. The β-lactoglobulin enriched WPI so produced will be acid and heat stable and will further comprise non sialic acid bearing aglyco-CMP.
The sialic acid rich glyco-CMP eluted from the first anion exchanger in either the fourth or fifth embodiment may be combined with the flow-through from
the second anion exchanger in each case to make an α-lactalbumin/ sialic acid enriched product. Or, if the steps in these two embodiments are carried out in the reverse order, sialic acid rich glyco-CMP eluted from the second anion exchanger may be combined with the flow-through from the first anion exchanger used to make the WPI to produce an α-lactalbumin/ sialic acid enriched product.
In all of the above embodiments, the step of eluting CMP from the anion exchanger preferably takes place under conditions of pH lower than those used to elute β-lactoglobulin, and/or using an eluent with a higher salt concentration than that used to elute β-lactoglobulin.
Preferably, the eluates containing β-lactoglobulin and CMP are recovered separately, to produce a β-lactoglobulin-enriched WPI; a β-lactoglobulin enriched WPI containing aglyco-CMP; a CMP isolate (comprising aglyco-CMP and glyco-CMP); and a glyco-CMP isolate, respectively by known methods such as ultrafiltration and spray drying. The β-lactoglobulin enriched WPI's can be neutralized prior to ultrafiltration and spray drying or optionally, they can be ultrafiltered and spray dried without neutralization at a pH in the region of 3.0-3.5 to make a WPI that is ready to use in an acid beverage.
The de-proteinised feedstock (flow-through) from an ion exchange step in which both β-lactoglobulin and CMP are adsorbed may optionally be further processed by ultrafiltration, anion exchange or cation exchange procedures known in the art to recover additional products.
In a sixth embodiment there is provided a process for producing glyco-CMP from a whey protein containing feedstock comprising:
(a) contacting an anion exchanger with an excess of feedstock under conditions whereby sialic acid rich glyco-CMP is selectively bound; and
(b) eluting the sialic acid rich glyco-CMP of step (a) from the anion exchanger.
Preferably the feedstock is overloaded onto the anion exchanger at step (a) at a pH of 4-6, preferably at a pH of 4.2-5.5.
The sialic acid rich glyco-CMP bound to the anion exchanger is eluted from the anion exchanger as discussed above. Particularly, in a batch process this may conveniently be achieved by using a combination of low pH and low salt concentration, e.g. pH 2 and 50 mM NaCl or a higher salt concentration under neutral conditions, e.g. pH ≥6 with 500 mM NaCl, or points in between.
In a further embodiment, the present invention provides a process of producing acid/heat stable WPI from any whey protein containing feedstock comprising the steps of
(a) contacting the feedstock with an anion exchanger under conditions whereby the major whey proteins β-lactoglobulin, CMP and optionally α-lactalbumin are adsorbed;
(b) eluting whey proteins from the anion exchanger;
(c) subjecting the eluted protein solution to mild hydrolysis conditions whereby the CMP is desialylated;
(d) further processing the solution of step (c) to produce a WPI.
The present invention further contemplates a process for producing a β- lactoglobulin enriched WPI from acid whey. Acid wheys do not contain CMP but they do contain acidic peptides and minor proteins which are highly phosphorylated. Acid wheys may therefore be similarly processed as herein described to produce β-lactoglobulin enriched WPI and a further isolate enriched in phosphopeptides/ proteins. The more highly phosphorylated peptides bind preferentially to anion exchangers, and at lower pH levels like the glyco-CMP does since both the phosphate and sialylate groups do not lose their negative charge until pH < 3. (The pKas of phosphoric acid and sialic acid both lie between 2 and 3.) The phosphopeptides/proteins are useful as calcium carriers for food ingredients.
Preferably, the anion exchanger(s) used in the processes of the present invention diethylaminoethyl (DEAE) and quarternary amino (QA) exchangers.
Where processes do not involve overloading of feedstock it is preferred that these anion exchangers are of the industrially useful cellulose type comprising a water-insoluble, hydrophilic, water swellable, hydroxy(C2 ~ C ) alkylated and cross-linked regenerated cellulose derivatised with quaternary amino (QA) groups, preferably in granular or beaded form.
Preferably, the level of substitution of the QA groups on the anion exchanger(s) is 1.4 milliequivalents per dry gram of anion exchanger (meq/g) or greater.
More preferably, the level of substitution of QA groups is from about 1.4 to about 2.5 meq/g, more preferably from about 1.5 to about 2.5 meq/g, and most preferably from about 1.7 meq/g to about 2.5 meq/g.
Preferably, the cellulose is hydroxypropylated cross-linked regenerated cellulose.
For the embodiments which include overloading of the anion exchanger with feedstock, the preferred anion exchanger is QMA Spherosil™ or like ion exchanger.
In a further aspect, the present invention produces a β-lactoglobulin enriched WPI product which is heat and acid stable (WPI*) .
In one embodiment such a product may further comprise the aglyco fraction of CMP, and thereby comprise a β-lactoglobulin enriched WPI containing aglyco-CMP (WPI**).
In a further aspect, the present invention provides a β-lactoglobulin enriched WPI (WPI*) obtained by or obtainable by a process as defined above.
In a further aspect, the present invention provides a β-lactoglobulin enriched
WPI containing aglyco-CMP (WPI**) obtained by or obtainable by a process as defined above.
In a further aspect, the present invention provides a CMP isolate obtained or obtainable by a process as defined above.
In a further aspect, the present invention provides a glyco-CMP isolate obtained or obtainable by a process as defined above.
In a yet further aspect, the present invention provides foodstuffs comprising the β-lactoglobulin enriched WPI (WPI* or WPI**) of the invention, including acid beverages.
In a still further embodiment, the present invention provides foodstuffs comprising the CMP or glyco-CMP isolates of the invention, including infant formulas and food formulations for patients suffering from phenylketonuria.
While the present invention is broadly as defined above, it is not limited thereto and also includes embodiments of which the following description provides examples.
BRIEF DESCRIPTION OF THE DRAWINGS
The present invention will now be described in more detail with reference to the accompanying drawings, in which
Figures 1 to 7 show schematically the processes of the main embodiments of the present invention; and
Figure 8 shows HPLC analyses of UF retentate prepared from cheese whey, and of first and second eluates obtained using a process of the invention, showing the individual whey proteins in each sample.
Figure 9 shows the acid/heat instability arising when glyco-CMP is added back to acid/heat stable anion WPI (WPI*)
Figure 10 shows the displacement of initially adsorbed aglyco-CMP (A and B variants) by glyco-CMP as a column of Macro-Prep™ High Q is overloaded with WPI.
DESCRIPTION OF THE INVENTION
As defined above, the present invention provides new processes useful for recovering separately both β-lactoglobulin and CMP from a single whey protein-containing feedstock.
The applicants have found that up to four products may be made by the processes of the present invention which are enriched in either β- lactoglobulin, CMP (aglyco-CMP and glyco-CMP or just mainly glyco-CMP) or both from a feedstock containing these proteins /peptides. In particular, the products may comprise a β-lactoglobulin enriched WPI (WPI*); a β- lactoglobulin enriched WPI containing aglyco-CMP (WPI**); a CMP isolate; and a glyco-CMP isolate.
"WPI" (whey protein isolate) is a term used loosely in the industry for whey protein products isolated using ion exchange or microfiltration. It is not well defined but has come to mean a product with < 1 % fat and greater than about 90% protein. In all cases β-lactoglobulin is the major protein present, while α-lactalbumin may vary from very little to an amount giving the same ratio of α-lactalbumin/ β-lactoglobulin as found in the original feedstock. Immunoglobulins and CMP may or may not be present depending on whether cation exchange, anion exchange and MF is used.
"Isolate." In cases where β-lactoglobulin is not the major protein present in the eluate stream from the ion exchanger we have simply referred to the product as an "isolate," e.g. CMP isolate.
"β-lactoglobulin enriched WPI" as used herein means a WPI with a % β- lactoglobulin content greater than that found in the whey protein containing feedstock relative to acidic peptides, i.e. peptides with pi < about 4.5. In the case of sweet wheys it is relative to CMP, in particular relative to both glyco- and aglyco-CMP, and is labelled WPI* herein. This WPI* contains < about 5% total CMP (In the case of acid wheys it is relative to phosphopeptides.).
"β-Lactoglobulin enriched WPI containing aglyco-CMP" as used herein means a WPI with a % β-lactoglobulin content greater than that found in the whey
protein feedstock relative to glyco-CMP but near normal β-lactoglobulin content relative to aglyco-CMP. In particular, this product contains more than 5% aglyco-CMP and less than 5% Glyco-CMP. It is labelled WPI**.
"CMP isolate" as used herein means an isolate containing CMP as the major protein present. This CMP contains both glyco-CMP and aglyco-CMP in a near normal ratio as found in the whey protein feedstock. It typically has a sialic acid content of 2-5%.
"Glyco-CMP isolate" as used herein means an isolate containing CMP as the major protein present but the CMP in this case has a higher ratio of glyco- CMP to aglyco-CMP than that found in the whey protein feedstock applied to the anion exchanger. It typically has a sialic acid content greater than 5%.
Some of the processes require the anion exchanger to be overloaded. This arises at pH 4-6 when it is required to preferentially bind glyco-CMP for example. "Overloading" as used herein means contacting the anion exchanger with so much feedstock or WPI that the protein binding capacity is taken up with the most tightly binding peptides e.g. glyco-CMP and there is little capacity left over for other peptides or proteins that would normally bind under the conditions being used, e.g. aglyco-CMP at pH 4-5 and/or D- lactoglobulin at pH 5-6. Below pH 4 there is little else to compete with the binding of glyco-CMP so overloading is not necessary below pH 4 to obtain selectivity for it.
The simplest methods of producing a heat/ acid stable WPI by anion exchanger involve removing the acidic peptides and proteins from the process stream either before or after the anion WPI is prepared. To achieve this an additional anion exchange step is used at a pH of about 5. At this pH, the major whey proteins β-lactoglobulin, α-lactalbumin, BSA and immunoglobulins are either neutral or positively charged and do not bind to the anion exchanger. However, acidic peptides and some minor whey proteins whose isoelectric point (IEP) is less than bout 4.5, collectively called acidic peptides herein, remain negatively charged and are selectively adsorbed from the process stream.
Thus, whey is first adjusted to about pH 6-7.5 and passed through a first anion exchanger to adsorb both the β-lactoglobulin and the acidic peptides. The proteins are then eluted and the eluate contacted with a second anion exchanger at about pH 4.5-5 to remove the acidic peptides. The major whey proteins β-lactoglobulin, α-lactalbumin and BSA are then recovered in the flow- through as WPI (Figure la). We have surprisingly found that such a WPI, labelled as WPI*, is heat stable under acidic conditions.
An alternative process is shown in Figure lb, whereby the whey is first adjusted to about pH 4.5-5 and passed through a first anion exchanger to selectively adsorb acidic peptides. The flow-through, containing β- lactoglobulin, α-lactalbumin and BSA, is then adjusted to a pH of about 6- 7.5 and passed through a second anion exchanger whereby the whey proteins are adsorbed onto the exchanger. The CMP isolate and the β- lactoglobulin/ α-lactalbumin enriched WPI may be obtained as separate eluates from the two anion exchangers. Again, the WPI produced by this process, labelled WPI* is heat stable under acid conditions.
WPI is also made commercially by microfiltration (MF). Such a WPI (MF WPI) is similar to anion WPI in that it contains acidic peptides and is not acid/ heat stable. It can be made stable by passing it through an anion exchanger at about pH 4.5-5 to remove acidic peptides (Figure 2a). The flow- through containing whey proteins is then further processed to produce acid/heat stable WPI*. The acidic peptides, including CMP, may be eluted separately.
A variation on this process is shown in Figure 2b, whereby the whey is first treated by anion exchange at about pH 4.5-5 to remove acidic peptides and then the flow-through is further processed by microfiltration to produce a WPI (WPI*) which is acid/heat stable.
A further process involves first adsorbing both β-lactoglobulin and CMP onto an anion exchanger, followed by eluting first the β-lactoglobulin (optionally together with a fraction of the CMP), and then, by lowering the pH and/ or by increasing the salt concentration, eluting the CMP (or the remaining adsorbed fraction thereof) as shown in Figure 3.
This particular process involves the step of contacting the feedstock with an anion exchanger under conditions in which both CMP and β-lactoglobulin will be adsorbed. The adsorption step is followed by first eluting the β- lactoglobulin from the exchanger, optionally together with a fraction of the adsorbed CMP, and then eluting the CMP (or the fraction of CMP remaining adsorbed) from the anion exchanger.
The feedstock is contacted with the anion exchanger under conditions which will bring about adsorption of CMP and β-lactoglobulin to the anion exchanger. To achieve this, it is preferred that the feedstock be contacted with the anion exchanger at a pH of about 5 to about 8.5, more preferably at a pH of about 6 to 7.5 (Figure 3a).
It is expected that a portion of the α-lactalbumin in the whey feedstock will be adsorbed and eluted with the β-lactoglobulin. The exact size of the portion will depend on the ion exchanger and pH used as well as the ratio of feedstock to ion exchanger. The α-lactalbumin will be spread between the β- lactoglobulin isolate and the breakthrough stream except in cases where the anion exchanger is substantially under-loaded or over-loaded, in which case it will be present in either the β-lactoglobulin isolate or breakthrough stream respectively. This is known to those skilled in the art.
The step of contacting the feedstock with the anion exchanger to adsorb the CMP and β-lactoglobulin can be carried out in any convenient manner. Preferred methods are to carry out this step in a stirred bed of anion exchanger or in a column of the anion exchanger. It is generally preferred that the adsorption step be carried out at a temperature of less than about 20°C, more preferably at around 8-15°C, to minimise the growth of mesophilic bacteria. It is also preferred that the contact time of the anion exchanger with the whey protein solution is less than about 2 hours, more preferably less than about 1 hour.
It is also generally preferred that the adsorption step is carried out under conditions so as to achieve a loading of protein on the anion exchanger of at least 50%, more preferably at least 75%, of the protein adsorption capacity of the anion exchanger.
Once adsorption of the CMP and β-lactoglobulin has taken place, the adsorbed proteins are sequentially eluted.
The β-lactoglobulin is eluted first. In one preferred embodiment, elution of β- lactoglobulin is carried out under conditions which, while achieving elution of the β-lactoglobulin (and α-lactalbumin, if adsorbed together with β- lactoglobulin), allow all or substantially all of the CMP to remain adsorbed to the anion exchanger. In this embodiment, it is preferred that the elution is carried out at a pH of from about 4 to about 5, and at a salt concentration of from 0 to about 100 mM. The elution may conveniently be carried out using sodium chloride, although other salts may also be used. This elution produces a first eluate which contains mainly β-lactoglobulin.
This eluate may then if desired be subjected to further processing including one or more of ultrafiltration, diafiltration, evaporation, freeze-drying and spray drying, to produce β-lactoglobulin enriched WPI*.
After the β-lactoglobulin has been eluted, the CMP remaining adsorbed to the anion exchanger is eluted. This may conveniently be achieved by using a combination of low pH and low salt concentration, e.g. pH 2 and 50 mM
NaCl, or a higher salt concentration under neutral conditions, e.g. pH > 6 with 500 mM NaCl, or points in between.
Again, the eluted CMP may if desired be subjected to further processing including one or more of ultrafiltration, diafiltration, evaporation, freeze- drying and spray drying, to produce a CMP isolate. The CMP isolate may be further purified to make it suitable as a dietary supplement suitable for persons suffering from phenylketonuria.
In an alternative, preferred embodiments of the invention, the process is carried out so as to achieve a fractionation of the CMP, that is, to split the CMP between the first and second eluates. CMP is not a single peptide, but is a complex mixture of macropeptide with varying degrees of glycosylation (covalently attached carbohydrate). About 40% of the macropeptide is not glycosylated at all (referred to as aglycomacropeptide, or aglyco-CMP), while the remainder has carbohydrate groups attached at up to five different sites on the peptide chain (referred to as glycomacropeptide or glyco-CMP). Each
of these carbohydrate groups so attached may contain up to two sialic acid units.
In this embodiment of the invention, the adsorption of the CMP and β- lactoglobulin is carried out in the same manner as described above.
However, the first elution step (b) is carried out under conditions which will also elute a fraction of the CMP. This may conveniently be achieved by carrying out the elution at a pH of less than 4, preferably about pH 2.5 to less than 4, and preferably at a salt concentration of from 0 to 50 mM (Figure 3b). Under these conditions, we have found that a fraction of the
CMP, which is mainly aglyco-CMP or less glycosylated CMP, can be eluted together with β-lactoglobulin to produce a β-lactoglobulin enriched WPI containing aglyco-CMP, which is labelled herein as WPI**. The CMP remaining adsorbed to the anion exchanger will mainly be glyco-CMP, which is rich in sialic acid.
This fractionation is believed to be achieved by virtue of the negatively charged sialic acid groups associated with the carbohydrate groups, which lower the isoelectric point (IEP) of the glyco-CMP from near pH 4 towards pH 2. The greater the number of sialic acid groups present on the MP the lower the IEP and the lower the pH, or the higher the salt concentration, required to elute it. Thus, the applicants have found that by manipulating the pH and ionic strength at which elution of β-lactoglobulin is carried out, a first eluate which contains a desired amount of the CMP (from about 20 to about 70% of the CMP) in addition to the β-lactoglobulin, and a second eluate which contains the balance of the CMP (mainly glyco-CMP), can be obtained.
The lower the yield of the CMP in the second eluate, the richer it will generally be in sialic acid. Sialic acids are contained at higher levels in human milk than bovine milk and are considered to act as an infection protection factor for infants. They are also important for the development of the brain.
Thus, in this embodiment of the invention, the first eluate will contain mainly β-lactoglobulin and some CMP (mainly aglyco-CMP), which may be subjected to further processing if desired, including one or more of
ultrafiltration, diafiltration, evaporation, freeze-drying and spray drying, to produce a β-lactoglobulin enriched WPI containing aglyco-CMP (WPI**). Again, the β-lactoglobulin enriched WPI containing aglyco-CMP (WPI**) produced by the process of this invention has been found to be heat stable in the pH 3.5-4.0 region.
Elution of the fraction of CMP remaining bound to the anion exchanger can conveniently be achieved by using a combination of low pH and low salt concentration, e.g. pH 2 and 50mM NaCl, or a higher salt concentration under neutral conditions, e.g. pH > 6 with 500 mM NaCl, or points in between.
Again, the eluted CMP may be further processed as desired, and the resulting product will be an isolate containing mainly glyco-CMP, enriched in sialic acid. Such further processing may include removal of contaminating proteins such as residual β-lactoglobulin, eg by cation exchange, to further enrich the sialic acid content.
In a further embodiment, the process of the present invention uses a whey protein containing feedstock at pH 2.5-4.5 which is contacted with an anion exchanger to selectively remove the very acidic material from it, e.g. the sialic acid rich glyco-CMP. The glyco-CMP deplete flow-through whey is then contacted with a second exchanger at pH 6-7.5 and a WPI prepared. Alternatively the flow- through whey stream may be used to prepare MF WPI. This WPI** will again contain the non-sialic acid bearing aglyco-CMP (Figure
4a).
The steps of the aforementioned embodiment can be carried in the reverse order, whereby MF WPI, or anion WPI, that is not acid/heat stable is used as a feedstock. The WPI is passed through an anion exchanger at a pH 2.5-4.5 to selectively adsorb the sialic acid rich glyco-CMP. The bulk of the whey protein and non-sialic acid bearing CMP are present in the flow- through. This is recovered as the acid/heat stable β-lactoglobulin enriched WPI containing aglyco-CMP, WPI**. The sialic acid rich glyco-CMP isolate is eluted from the anion exchanger and recovered (Figure 4b).
In yet another embodiment, a whey protein containing solution is contacted with an anion exchanger at about pH 5 (pH 4-6) to selectively bind the sialic acid rich glyco-CMP by overloading the column such that even if non-sialic acid bearing CMP is bound initially it is eventually displaced by the more tightly binding very acidic peptides e.g. sialic acid rich glyco-CMP (Figure
5a). The steps of this process may be reversed as shown schematically in Figure 5b.
A process to produce a glyco-CMP isolate only is set out in Figure 6, whereby an excess of whey protein containing feedstock is contacted with an anion exchanger at pH 4-6 (preferably pH 4.5-5), whereby sialic-acid rich glyco- CMP is selectively bound to the exchanger, which may then be eluted therefrom.
The feedstock used in the processes of the present invention can be any solution containing both CMP and β-lactoglobulin. CMP is found only in sweet wheys, so suitable feedstocks include sweet whey, UF retentate derived from sweet whey, and reconstituted WPC derived from sweet whey.
The applicants have found surprisingly that the β-lactoglobulin enriched
WPIs, WPI* and WPI**, produced by the processes of the present invention are heat stable in the acidic region of pH 3.5 - 4.0. This contrasts with WPI products produced by anion exchange using a conventional, non-selective elution to give a total WPI product comprising β-lactoglobulin, CMP and other minor components including the very acidic peptides and proteins.
Such WPIs suffer the disadvantage of not being heat stable, especially in the acidic region of pH 3.5 - 4.0. This gives solutions of the WPIs a cloudy appearance or forms a sediment which is unacceptable in certain applications such as protein fortification of acidic beverages such as fruit drinks.
Anion exchangers suitable for use in the processes of the present invention include diethylaminoethyl (DEAE) and quaternary amino (QA) exchangers.
Where processes do not involve overloading of feedstock it is preferred that these anion exchanger are of the industrially useful cellulosic type comprising a water-insoluble, hydrophilic, water swellable, hydroxy (C2-C )
alkylated and cross-linked regenerated cellulose, derivatised with quaternary amino (QA) groups, which is preferably in granular or beaded form. Such cellulose matrices and processed for preparing are then described in US patent 4,175,183 (John S Ayers). It is also particularly preferred that the anion exchangers have a substitution level of QA groups of about 1.4 meq/g or higher, more preferably from about 1.5 to about 2.5 meq/g, and most preferably from about 1.7 meq/g to about 2.5 meq/g.
In this specification, the term "QA" or "quaternary amino", when used in the context of ion exchangers, means a functional group selected from a group of the formula -ORi-Z, wherein Ri is a lower alkylene group containing 1 to 3 carbon atoms and optionally substituted with a hydroxyl group, and Z is a quaternized amino group of the formula: -N^Rs ^+OH- or salts thereof, wherein R2, R3 and * are each a lower alkyl group containing 1 to 4 carbon atoms, optionally substituted with a hydroxyl group, or a further group of the formula -Rι-NR2R3R +OH_ or salts thereof wherein Ri, R2, R3 and R-t are as defined above. Examples of suitable QA groups are -CH2CH2N+R2R3R4 Cl~ and -OCH2CHOHCH2N+R2R3R4 CF, wherein R2, R3 and R4 are the same or different and are selected from -CH3, -CH2CH3, -CH2CH2OH, -CH2CHOHCH3, -CH2CH2N+R2R3R4 Cl~ and -CH2CHOHCH2N+R2R3R4 Cl~.
The QA anion exchangers particularly preferred for use in these processes of the present invention and having a substitution level of 1.4 meq/g or greater may be prepared as disclosed in co-pending application NZ 501644 and NZ 505071.
In embodiments which do include overloading of the exchanger with feedstock, QMA Spherosil™ or like anion exchanger is preferred.
The invention will now be described in more detail with reference to the following non-limiting examples.
EXAMPLES
Example 1
QA GibcoCel™ HG2 (1.17 meq/g), an anion exchanger made from granular regenerated cellulose and particularly suited to large scale industrial use, was obtained from Life Technologies Ltd, Auckland, New Zealand. It was suspended in water and then collected on a sintered glass filter where it was washed with 1 M hydrochloric acid, water, 1 M sodium hydroxide and finally de-ionised water. It was then drained of excess water by vacuum filtration. This QA cellulose in its hydroxide form was then further alkylated to raise the density of positively charged QA groups.
The QA GibcoCel™[OH~] (450 g) was made up to a thick slurry by the addition of water (310 mL) and 30% (w/v) aqueous sodium hydroxide (50 mL). The mixture was chilled before adding 60 mL of (3-chloro-2- hydroxypropyl)trimethylammonium chloride (60 wt. % solution in water). These ingredients were mixed as a slurry for 17 hours at room temperature followed by 2 hours at 60° C. The QA cellulose product was collected on a filter and washed with water, 1 M hydrochloric acid and water before removing the excess water by vacuum filtration.
A small sample of the product was analyzed to determine its substitution levels of QA groups. About 5 g of the moist product was converted to its hydroxide form by washing with 1M sodium hydroxide followed by demineralized water. The sample was then titrated in 1 M sodium chloride with 1.00 M hydrochloric acid to an end-point of pH 4. After titration the sample was collected on a dry tared sintered-glass filter, washed with water and dried overnight at 105° C. The substitution level was calculated as the small ion exchange capacity (S.I.C.) and found to be 1.8 milli-equivalents per dry gram (meq/g), i.e. S.I.C.=V/wt where V = volume in mL of 1.00 M HC1, and wt=dry weight of the sample (g).
Example 2 (comparative) WPI with CMP by standard anion exchange (single elution)
The moist QA cellulose (213 g, equivalent to 300 mL settled volume) from Example 1 was placed in an 800 mL reaction vessel fitted with a screen and outlet tap at the bottom, and an overhead stirrer. To this was added 200 g
of freshly prepared UF retentate (15.3% protein, 21.5% total solids, pH 6.3) prepared from cheese whey, followed by 400 mL of water. This was stirred for 10 minutes and then the pH was adjusted to 7.0 with 10% sodium hydroxide while stirring for a further 40 minutes. The vessel was drained to the surface of the QA cellulose which was then washed with water (360 mL). The filtrate and washings were combined to give 820 g of flow-through solution. This was found to have remaining in it only 24% of the original protein giving an adsorption capacity of 78 g/L for the QA cellulose. (In a similar experiment where 50% more retentate and water were used the adsorption capacity only increased to 90 g/L, indicating that this is about the maximum capacity obtainable).
Further water was added to the vessel to give a total volume, cellulose plus water, of 450 mL and 50 mL of 4 M sodium chloride was added to give a final concentration of approximately 0.4 M. This was stirred for 1 hour without any pH adjustment and the vessel was then drained and washed with 0.4 M sodium chloride (360 mL). The protein solution and washings were combined to give 547 g of eluate. This was analysed for protein (TN x 6.38) and by HPLC to determine the concentrations of the individual proteins; CMP, proteose-peptone PP5, α-lactalbumin, BSA, β-lactoglobulin and immunoglobulins using the method published in J. Chromatography A, 878 (2000), 183-196. Samples of the starting retentate and the flow-through solutions were treated similarly and used to calculate the results shown in Tables 1 and 2.
Table 1 Distribution of Protein, % of retentate
Protein* CMP PP5 α-Lac BSA β-Lg
Retentate 100 100 100 100 100 100
Flow-through 26 5 26 39 57 7 Single eluate 73 97 84 61 51 97
Mass balance, % 99 102 110 100 108 104
Table 2
Protein Composition of Eluate (% of Total Proteint)
CMP PP5 α-Lac BSA β-Lg
Single eluate 21 3.3 9.4 1.0 55
tβased on TN x 6.38
The eluate was concentrated by ultrafiltration, diafiltered and freeze-dried. The resulting WPI powder was found not to be heat stable under acid conditions as described in Example 5.
Sialic acid was shown to be present in this WPI powder and amounted to
1.6% (w/w) as determined using the method published in J. Biological
Chemistry 234 (1959), 1971-1975.
Example 3
WPI by selective elution atpH 4 (Figure 3 a)
This was similar to Example 2 except that the elution was carried out in two stages, a first elution at pH 4 and a second elution with 0.4 M sodium chloride.
First Elution
After adsorbing whey protein from the retentate at pH 7.0, draining and washing the QA cellulose as described in Example 2, further water was added to the vessel to give a total volume, cellulose plus water, of 450 mL. This was stirred and the pH adjusted to 4.0. After 30 minutes the solution of protein desorbed from the QA cellulose was drained from the vessel and the QA cellulose washed with 10 mM sodium chloride at pH 4 (120 mL). Further 10 mM sodium chloride was added to the vessel to give a total volume of 450 mL and stirred whilst maintaining the pH at 4.0. After 30 minutes the vessel was once again drained and the QA cellulose washed with 10 mM sodium chloride (240 mL). All the washings and protein solutions were combined to give a first eluate (705 g) . A large sample of this first eluate was neutralized and concentrated by ultrafiltration using a
Vivaflow 200 module with a 5000 MWCO membrane (Vivascience) . The retentate was then freeze-dried and the WPI* powder analysed for heat stability under acidic conditions as described in Example 5.
Second Elution
A second protein fraction was obtained by eluting further protein from the QA cellulose with 0.4 M sodium chloride. Water was added to the vessel to once again bring the total volume up to the 450 mL mark and then 50 mL of 4 M sodium chloride was added to give a final concentration of approximately 0.4 M. This was stirred for 1 hour without any pH adjustment and the vessel was then drained and washed with 0.4 M sodium chloride (360 mL). The protein solution and washings were combined to give a second eluate (547 g). A sample of this CMP containing second eluate was dialysed, freeze-dried and analysed for sialic acid as described in Example 2.
Samples of the retentate, flow-through solution, first eluate and second eluate were analysed for protein (TN x 6.38) and by HPLC as described in Example 2. The results are shown in Tables 3 and 4.
Table 3
Distribution of Protein, % of retentate
Protein* CMP PP5 α-Lac BSA β-Lg
Retentate 100 100 100 100 100 100
Flow-through 24 6 26 40 57 7
First eluate, WPP * 49 4 28 57 43 85
Second eluate 21 84 45 3 3 6
Mass balance, % 94 94 99 100 103 98
Table 4 Protein Composition of Eluates (% of Total Proteint)
CMP PP5 α-Lac BSA β-Lg
Retentate 16.8 3.1 11.9 2.3 43
First eluate, WPI* 1.5 1.7 14.0 2.0 75
Second eluate 66 6.5 1.5 0.3 12
tβased on TN x 6.38
Greater than 90% of the CMP and β-lactoglobulin were adsorbed from the diluted retentate by the anion exchanger with lesser efficiencies for the other proteins, especially the immunoglobulins, greater than 80 % of which remained in the flow- through. The immunoglobulins made up 31% of the total protein in the flow-through which could be concentrated and dried to give an immunoglobulin enriched WPC.
By means of this sequential elution two whey protein isolates were obtained;
i) a WPI (WPI*) containing mainly β-lg (75% of total protein). This was found to have improved heat stability under acid conditions compared with the WPI produced by a single elution (Example 2), as shown in Example 5.
ii) a further isolate containing mainly CMP (66% of total protein) and found to have a sialic acid content of 4.4% of powder weight. This isolate is thus enriched in sialic acid compared with the WPI produced by the process of Example 2.
Example 4
WPI with aglyco-CMP by selective elution atpH 3.4 (Figure 3b) This was the same as Example 3 except that the first elution was carried out at pH 3.4 instead of pH 4.0. The second elution, to desorb the protein still remaining on the QA cellulose, could have been carried out as in Example 3 using 0.4 M sodium chloride but alternative conditions using minimal salt and low pH were used to illustrate this option as follows.
Second Elution
At the end of the first elution water was again added to the vessel to give a total volume (QA cellulose plus water) of 450 mL. A small aliquot (6.8 mL) of 4 M sodium chloride was then added to give a nominal concentration of 60 M in the vessel and the pH was adjusted to 2.0 by the slow addition of 10% hydrochloric acid while stirring. After a desorption time of one hour the
QA cellulose was drained and washed as in Example 3.
Samples of the retentate, flow-through, first and second eluates were analysed as in Example 3. The results are shown in Tables 5 and 6. A sample of the second eluate was also dialysed, freeze-dried, analysed for sialic acid content and found to have a sialic acid content of 9% based on powder weight compared with only 4.4% for the second eluate in Example 3.
Table 5 Distribution of Protein, % of retentate
Protein* CMP PP5 α-Lac BSA β-Lg
Retentate 100 100 100 100 100 100
Flow-through 25 6 25 37 50 7
First eluate, WPI** 62 63 64 55 47 84
Second eluate 9 27 3 3 1 6
Mass balance, % 96 96 92 95 98 97
Table 6 Protein Composition of Eluates (% of Total Proteint)
CMP PP5 α-Lac BSA β-Lg
First eluate, WPI** 18 3.3 11.0 1.8 61
Second eluate 50 1.2 4.5 0.2 29
tBased on TN x 6.38
With the use of the lower pH of 3.4 for the first elution a large proportion (63%) of the CMP was recovered along with β-lactoglobulin in the first eluate. The WPI** present in the first eluate thus still has 18% CMP in it. But HPLC analysis showed this CMP to be the aglyco components. This can be seen from Figure 8 showing the reverse phase HPLC chromatograms of the starting whey retentate, first and second eluates. The peaks eluting between three and five minutes are attributed to the glyco-CMP components, while those between five and seven minutes belong to the aglyco-CMP components (J. Chromatography A. 878 (2000), 183-196). These aglyco-CMP components lack sialic acid.
The second eluate contained only 27% of the original CMP but it was the bulk (70%) of the more highly glycosylated and sialylated CMP present in the starting retentate, as can be seen in Figure 8.
By means of this sequential elution, with the first elution carried out at a pH of less than 4 so that the aglyco-CMP is desorbed with the β-lactoglobulin (and α-lactalbumin), a higher yield can be obtained of WPI which is stable to heat under acid conditions. The second elution gave an isolate containing mainly glyco-CMP and was thus further enriched in sialic acid.
Similar products were obtained by carrying out this first elution at about pH 4 still, by the addition of sodium chloride to raise the ionic strength. This will be understood by those skilled in the art of chromatography. However the separation of aglyco- and glyco-CMP was not as efficient.
Example 5
Heat stability of WPIs under acidic conditions The WPI powders from Examples 2 (single eluate) and Examples 3 and 4
(first eluates) were assessed for their heat stabilities under acidic conditions using the following method.
A 3% solution of each of the WPI powders was prepared. Each solution was acidified using 10% phosphoric acid to pH end-points of 4.0, 3.8, 3.6, 3.4 and 3.2. As each end-point was reached a 15 L sample for testing was removed before lowering the pH of the remaining solution to the next point.
All 15 samples were heated for 20 minutes in a water bath at 80°C and then cooled to room temperature in a second water bath. The appearance of sediment in some of the samples was noted. The optical densities at 610 nm of the remaining samples were measured and are shown in Table 7. The WPI products (first eluates) from Examples 3 and 4 did not give rise to sediments and were thus heat stable at pH 3.6-4.0 whereas the WPI from Example 2 was not.
Table 7 OD at 610 nm of 3% WPI solutions after heat treatment
t protein precipitated during heating and settled out.
Table 7 also shows the improved stability of WPIs obtained by anion exchange similar to Example 2 except that the whey solution used was one from which most of the CMP had been removed first (Example 6), or one from which mainly glyco-CMP had been removed (Example 7).
The compositions of the five WPI products tested are shown in Table 8. Values are given for the glyco and aglyco subfractions of CMP as only a reduction in the former is important to obtain stability as shown by the composition of the WPI from Example 4. (Examples 6d and 10 further confirmed this.)
Table 8 Protein Composition of WPIs (% of Total Protein)
Example 6 a) Removal of CMP from UF retentate from cheese whey (Figure lb)
A sample of retentate, prepared by the ultrafiltration of cheese whey on a commercial scale, was obtained from industry, diluted with an equal weight of water and adjusted in pH to 4.8 with 10% hydrochloric acid.
A 750 mL sample of this was added to a beaker containing 320 g (450 mL) of moist QA cellulose from Example 1 and stirred for 40 minutes while maintaining the pH at 4.8 with 10% sodium hydroxide. At the end of the mixing period the beaker contents were transferred to the 800 mL reaction vessel (described in Example 2).
The CMP reduced retentate was drained from the anion exchanger and the latter washed with water. The filtrate and washings were combined to give 1052 g of flow- through retentate reduced in CMP content. WPI was recovered from this as described in part c) below.
b) Recovery of aglyco- and glyco-CMP fractions from the anion exchanger Elution of the adsorbed CMP from the QA cellulose from part a) was carried out in two stages while it was still in the reaction vessel.
i) Desorption in 20 mM NaCl at pH 3.3.
Water was added to the anion exchanger to give a total volume, cellulose plus water, of 800 mL Aqueous sodium chloride (4 mL of 4 M) was added to make the solution approximately 20 mM, and the pH was lowered to 3.3 with 10% hydrochloric acid. After stirring for 1 hour the desorbed CMP solution was drained from the anion exchanger which was then washed with further 20 mM sodium chloride to give 1006 g of protein solution. This was neutralized to pH 6-7 and concentrated by ultrafiltration. The retentate was analysed by HPLC. The protein was found to be 85% CMP with 74% of this being aglyco-CMP. A sample was freeze dried for sialic acid analysis.
ii) Desorption in 200 mM NaCl at pH 3.0.
The second elution of the QA cellulose was carried out in the same manner as for the first except that 40 mL of 4 M sodium chloride was used to make the solution approximately 0.2 M and the pH was lowered to 3.0. The combined filtrate and washings with 0.2 M salt, were neutralized, ultrafiltered and diafiltered.
The retentate was analysed by HPLC. The protein was found to be 90% CMP of which 80% was glyco-CMP. A sample was freeze dried for sialic acid analysis.
Samples of the freeze dried powders from parts b) above and c) below above gave the following results when analysed for sialic acid.
Preparation of WPI*
The flow-through retentate from part a) above was adjusted from pH 4.8 to 7.5 by the addition of 10% sodium hydroxide, added to a fresh lot of QA cellulose (320 g, 450 mL) and mixed for 40 minutes while maintaining the pH at 7.5. The anion exchanger was again collected in the 800 mL reaction vessel and washed with water. The filtrate and washings were combined to give 1257 g of whey protein depleted flow-through. Analysis of this solution by HPLC showed that 96% of the β-lactoglobulin and 61% of the α-lactalbumin had been adsorbed by the QA cellulose. No attempt was made to separate these two proteins.
Further water was added to the vessel to give a total volume, cellulose plus water, of 800 mL. This was stirred, 10 mL of 4 M sodium chloride added to give a final concentration of about 50 mM and the pH adjusted to 3 with 10% hydrochloric acid. After stirring for 1 hour the vessel was drained and the anion exchanger washed with water. The filtrate and washings were combined to give 1016 g of WPI* solution (eluate). When analysed by HPLC the protein in the eluate was found to have a CMP content of only 3%, or more importantly a glyco-CMP content of only 0.8% (Table 8).
The eluate was neutralized to pH 6.7, concentrated by ultrafiltration and diafiltered. A sample of the final retentate was freeze dried. A further sample of the retentate, when diluted with water to a 3% protein concentration, and pH adjusted to 3.2-4.0, was found to have improved heat stability compared with WPI from Example 2. See
Table 7 and also section d) below.
d) Destabilization of Anion WPP by glyco-CMP
The WPI retentate from part c), and the glyco-CMP retentate from part b) were recombined in various proportions and diluted with water so as to give 30 mL solutions of 3% total protein in each case, but having a range of 0 to 12% glyco-CMP present in the total protein. For example the 3% protein solution having 10% of the total protein present as glyco-CMP was made up by combining 6.86 mL of the WPI* retentate (118 mg protein/mL), 6.30 mL of the glyco-CMP retentate
(14.3 mg protein/mL) and 16.84 mL of water. Each solution was then adjusted to pH 3.60 and heated as described in Example 5 to determine the protein stability.
The process was repeated with mixtures of aglyco-CMP and WPI* but covering a much wider range from 0 to 80% aglyco-CMP. The results are shown in Figure 9. (Solutions which formed precipitates, and eventually contained sediments, are shown as having absorbance readings of 2.5 to indicate this sedimentation).
Clearly, it is the glyco-CMP containing fraction which destabilized the protein and not the aglyco-CMP fraction. The 3% protein solution was rendered unstable if it had more than about 4% glyco-CMP but up to about 50% aglyco-CMP could be tolerated without destabilizing the WPI*. Above 50% the instability probably resulted from the small amount (approx. 10%) of glyco-CMP present in the aglyco-CMP fraction.
e) The process of steps a, b and c were repeated. The glyco-CMP product from step b) ii) was kept as a sialic acid enriched product. But the aglyco-CMP eluate from step b) i) was combined with the WPI* eluate from step c) and the combined total eluates were neutralized, concentrated by diafiltration and dried to give an acid /heat stable
product WPI** (Figure 7a). The combined eluate solution was analysed by HPLC and found to have the composition shown in Table 8. The aglyco-CMP content was still about 13%, as for Example 2 WPI, but the glyco-CMP was less than 5%. The overall yield of protein in the WPI** was 72% with a further 6% of starting protein in the glyco-CMP fraction. The major losses were α-lactalbumin and immunoglobulins (7% and 5% of starting protein respectively) in the final flow-through from both anion exchangers (Figure 7a). Immunoglobulins are not adsorbed by anion exchangers and α-lactalbumin is known to be difficult.
In yet another repeat of steps a, b and c the anion exchanger in step c) was replaced with 600 mL of a cation exchanger, SP GibcoCel™, and it was contacted with the flow-through retentate from part a) which had been adjusted to pH 3.5 with 10% HC1 (Figure 7b). The rest of the process was as described in this section e) . The overall yield of protein in the WPI stream increased to 80% as a result of the cation exchanger binding immunoglobulins and more of the α-lactalbumin.
Example 7 a) WPI** by anion exchange from glyco-CMP reduced cheese whey retentate (Figure 5a) This process was the same as example 6 a) except that only half the quantity of QA cellulose was used, i.e. 160 g (225 mL) in order to favour the adsorption of glyco-CMP over aglyco-CMP, and a pH of 4.4 was used instead of 4.8. The treated retentate had a reduced glyco-CMP content and was used to prepare WPI** exactly as described in Example 6 c) for WPI*. The protein composition of the WPI** obtained is shown in Table 8. The improved heat stability of the WPI**, over that from Example 2, is shown in Table 7. The major difference is in the reduction in the amount of glyco-CMP present in the WPI from 8% to about 3% of protein.
b) Recovery of sialic acid rich glyco-CMP from the anion exchanger
The protein adsorbed by the QA cellulose in part a) was recovered from it by a single desorption at pH 3.0 in 200 mM NaCl. Water was added
to the QA cellulose to give a total volume of 450 mL. Aqueous sodium chloride (22.5 mL of 4 M) was added to make the solution approximately 200 mM and the pH was lowered to 3.0 by the addition of 10% hydrochloric acid. After stirring for 1 hour the QA cellulose was drained and washed with 200 mM sodium chloride. The combined filtrate and washings were neutralized, ultrafiltered and diafiltered. HPLC analysis of the retentate showed that 94% of the protein present was CMP with 64% of this being glyco-CMP. A sample of the retentate was also freeze-dried. Analysis of the powder showed it to have a sialic acid content of 8.7%.
c) Preparation of an a-lactalbumin and sialic acid enriched WPC
The residual protein (32%) in the WPI depleted retentate from part a) was found to still contain 68% of the α-lactalbumin present in the starting retentate. This α-lactalbumin made up 40% of the residual protein analysed by HPLC so could be made into a useful WPC. For example, in a separate experiment the sialic acid rich eluate from b) was combined with this WPI depleted flow- through retentate from a). It was then ultrafiltered to produce a WPC enriched in sialic acid and α-lactalbumin, as shown in Table 9, both useful ingredients for an infant formula.
Table 9 Protein Composition (% of protein assayed by HPLC)
% of powder weight.
Example 8 a) WPI** by microfiltration of glyco-CMP reduced cheese whey retentate (Figure 5a)
Glyco-CMP was preferentially adsorbed from cheese whey retentate in much the same manner as for Example 7 a) except that a pH of 5.5
was used. Once again a limited amount of anion exchanger was used to suppress the adsorption of aglyco-CMP as well as α-lactalbumin and β-lactoglobulin which would normally be adsorbed at this pH. Cheese whey retentate (2.25 L) was mixed with water (1.6 L) and adjusted to pH 5.5 with 10% HCl. This was then added to 2.25 L (1.6 kg) of QA cellulose (Example 1) and mixed for 40 minutes whilst maintaining the pH at 5.5. The QA cellulose was then drained and washed with water. The treated retentate and wash water were combined, neutralized and further diluted to a protein concentration of 2%. This solution was microfiltered to produce a WPI containing permeate using a Terra- Pak MFS1 system (0.1 micron membrane). A sample of the permeate was then concentrated by ultrafiltration in the laboratory using a Millipore Prep-Scale ultrafilter with 3 kD membrane. The final retentate (WPI** solution) was diluted to give a 3% protein solution which was subjected to the heat stability test as described in Example 5.
A sample of the original cheese whey retentate was similarly diluted and processed by microfiltration and ultrafiltration to prepare a standard MF WPI for comparison.
The results of the heat stability tests and the composition of the two WPIs are shown in Tables 10 and 11 respectively.
Table 10 OD at 610 nm of 3% WPI solutions after heat treatment
t protein precipitated during heating and settled out.
Table 11 Protein Composition (% of protein assayed by HPLC)
b) Recovery of sialic acid rich glyco-CMP from the anion exchanger
The acidic peptides adsorbed by the QA cellulose in part a) were recovered from it as described in Example 7 b). The protein composition of the product is shown in Table 11 and it was found to have a sialic acid content of 5.2%.
Example 9
Removal of very acidic peptides from anion WPI (Figure 5b) Stage 1 Preparation of anion WPI
The moist QA cellulose (213 g, equivalent to 300 mL settled volume) from Example 1 was mixed with 240 g of cheese whey UF retentate and 360 mL of water as described in Example 2 except that the pH was adjusted to 7.4. After mixing for 40 minutes, the retentate was drained and the bed of QA cellulose washed with water. The treated retentate (combined filtrate and washings) was set aside. (Its protein composition is shown in Table 12 and the protein could be recovered as a WPC or further WPI product enriched in α-lactalbumin by cation or anion exchange treatment as required.)
Further water was added to the vessel to give a total volume, cellulose plus water, of 540 mL. This was stirred, 16 mL of 2 M NaCl added to give a NaCl concentration of approximately 60 mM and the pH lowered to 2.5 with 10% HCl. After a total mixing time of 40 minutes the eluted protein solution was drained from the vessel and the QA cellulose bed washed with a minimum amount of water. The eluate and wash water were combined (600 mL) and
the pH raised to about 6 with 10% NaOH, to give a clear WPI solution with a protein concentration of about 3.6%.
A sample of the solution was dialysed and freeze-dried for analysis.
Stage 2 Removal of very acidic peptides from anion WPI A 70 mL fraction of the WPI solution obtained in stage 1 was diluted with an equal quantity of water, to reduce the ionic strength, and the pH lowered to 4.9. This solution, with a conductivity of 5.5 mS/cm, was passed through a 7.5 mL column of Macro-Prep™ High Q anion exchanger (Bio-Rad Laboratories) which had been previously washed with 0.5 M NaCl and water. The column was loaded at a flow rate of 2.5 mL/min. At the end of the loading the column was washed with 25 mL of 0.05 M NaCl. The flow- through and washings were collected, neutralized to pH 6.5, dialysed and freeze-dried to recover the WPI. The column was eluted by washing it with 25 mL of 0.15 M NaCl to desorb the acidic peptides, mainly glyco-CMP, which had been removed from the WPI. This eluate fraction was also neutralized to pH 6.5, dialysed and freeze-dried.
Analyses were carried out as described in Examples 2 and 5 and the results are shown in Tables 12 and 13.
Table 12 Protein Composition (% of protein assayed by HPLC)
Table 13 OD at 610 nm of 3% WPI solutions after heat treatment
tprotein precipitated and settled out.
The anion WPI (Stage 1 product) and the recovered acidic peptides had sialic acid contents of 1.8% and 12.4% respectively.
The, glyco-CMP components made up 74% of the CMP in the very acidic peptides, but only 36% and 17% of the CMP in the initial WPI and flow- through WPI** respectively.
When stage 2 was carried out at pH 5.3 instead of 4.9 similar results were obtained.
Example 10
Removal of very acidic peptides from MF WPI (Figure 5b)
Microfiltered whey was prepared by microfiltering rennet whey protein concentrate as described in Example 8. The MF permeate (containing 2.5% protein) was then acidified to pH 5.3 using 10% HCl and passed through a 7.5 mL column of Macro-Prep™ High Q anion exchanger as described in Example 9. Twenty mL fractions were collected and at the finish the column was washed with 0.02 M NaCl. Every alternate fraction was analysed by HPLC to determine the concentration of total CMP, glyco-CMP, and the A and B variants of the aglyco-CMP. The results, shown in Figure 10, clearly show the displacement of the aglyco-CMP variants A and B from the column by the more tightly binding glyco-CMP. The presence of these components in the flow-through WPI stream, even at elevated concentrations, e.g. fractions
10 and 6 respectively, did not destabilize the WPI. Some of the glyco-CMP components could even be tolerated in the WPI before it was rendered unstable. This was determined by taking a sample from each fraction, shifting it to pH 3.9 and testing for heat stability. Only fractions 19 and 20 started to show an increase in cloudiness. The fractions were combined, neutralized, dialysed and freeze-dried to recover the WPI**. The column was washed with 25 mL of 0.15 M NaCl to desorb the very acidic peptides, mainly glyco-CMP, which had been removed from the WPI. This eluate was also neutralized, dialysed and freeze-dried. The dry weight of the product amounted to about 4% of the starting MF WPI. Samples of the starting WPI, flow-through WPI** and acidic peptides fraction were analysed by HPLC to determine the ratio of individual proteins as described in Example 2. The results are shown in Table 14. The aglyco-CMP components made up 90% of the CMP remaining in the WPI**, whereas the glyco-CMP components made up 76% of the CMP in the eluate.
Table 14 Protein Composition of WPIs (% of total proteins assayed by HPLC)
Samples of the WPI and WPI** powders, before and after passage through the anion exchange column respectively, were assessed for their heat stabilities under acidic conditions using the method described in Example 5. The results are shown in Table 15. The anion exchange treated WPI** was stable at pH 4.0 and had superior clarity at pH 3.8.
Table 15 OD at 610 nm of 3% WPI solutions after heating
fprotein precipitated during heating and settled out.
The starting MF WPI and the eluate powder were also analysed for sialic acid as described in Example 2. The sialic acid content of the starting WPI was 0.4% while that of the eluate powder was 10%.
Example 11
Stabilization of anion WPI by mild hydrolysis atpH 2
Anion WPI was prepared in much the same manner as in Example 9 except that elution of the WPI from the anion exchanger was carried out at pH 2 in 30 mM NaCl, such conditions being useful for the immediate hydrolysis of sialic acid groups from the CMP by heating.
Ultrafiltered cheese whey (240 g, 23% total solids) was diluted with 360 g of water and contacted with 214 mL of QA cellulose anion exchanger (152 g) from Example 1. The whey and QA cellulose mixture was adjusted to pH 7.4 with 10% NaOH and stirred for 40 min. The protein-depleted whey was then drained off and the QA cellulose washed to remove residual whey solids.
Water was then added to give 320 mL of QA cellulose plus water and 2 M NaCl added to make the solution approximately 30 mM in salt. The mixture was adjusted to pH 2 with 10% HCl and stirred for 40 min. The eluted protein solution was drained from the QA cellulose which was washed with water to give 430 g of WPI eluate.
A 100 g sample of the eluted protein solution, still at pH 2, was heated at 90°C for 10 minutes. It was then cooled and neutralised with aqueous NaOH. The remaining (unheated) eluate was also neutralised. A 15 ml sample of each of the heated and unheated eluate were ultrafiltered to 7 ml in a 5000 molecular weight cut-off Centracon tube (Vivaflo). Diafiltration with 7 ml of water was then carried out. The heated WPI solution had a sialic acid content of 0.3% (% of total solids) whereas the unheated WPI had a sialic acid content of 1.2%.
Both WPI solutions were diluted back to 3% protein to test their relative stabilities under acidic conditions as described in Example 5. The WPI that had been heated had an absorbance of less than 0.1 at pH 3.7 but the untreated WPI was stable only up to pH 3.4 as seen in Table 7 for a similar WPI made in Example 2.
Example 12
Removal of the very acidic peptides (glyco-CMP) from WPI at pH 3.5 (Figure 4b) a) WPI by anion exchange
Freshly prepared rennet whey (360 mL, 0.56% protein) at pH 6.6 was passed through a column (20 cm x 1.6 cm D.) packed with 20 g of
QMA Spherosil™ (39 mL) at a flow rate of 2 mL/min. The last of the whey was washed through with 50 mL of 30 mM NaCl. The deproteinated whey and washings were collected. A further 25 mL of
30 mM NaCl was then circulated through the column at 4 mL/min. It was continuously adjusted to pH 2 with 1 M HCl while external to the column until the pH was stable. The solution containing the eluted protein (WPI) was then displaced from the column with water and neutralized to pH 3.5 using 10% NaOH. Further water was added to it to reduce the conductivity to 3-4 mS/cm.
b) Removal of the very acidic peptides (glyco-CMP) from WPI at pH 3.5
The WPI solution at pH 3.5 from part (a) was passed through a second column (12 mL) of QMA Spherosil (6.2 g) at a flow rate of 1 mL/min. (Analysis of the WPI solution emerging from the column at several points showed that the aglyco-CMP components did not bind to the column at this pH at any stage.) The column was washed with water and the combined flow-through WPI solution and washings
neutralized, dialysed and freeze-dried to give 1.52 g of WPI** powder (75% protein yield).
The column was washed with 0.5 M NaCl (30 mL) to recover the adsorbed acidic peptides (glyco-CMP). The eluate was neutralized, dialysed and freeze-dried to give 0.13 g of powder (6% of initial protein). Analysis showed this powder to have 8.9% sialic acid.
Part a) was repeated and the WPI solution neutralized to pH 6.5, dialysed and freeze-dried for use as a comparative WPI sample.
The composition of the WPIs and their heat stabilities are shown in Tables 16 and 17. The decrease in glyco-CMP from 7.6 to 2.4% was the only significant difference resulting in the WPI** being more heat stable. The sialic acid content of the WPI from part a) was 1.2% and dropped to 0.3% after reducing the glyco-CMP content.
Table 16 Protein Composition of WPIs (% of total proteins assayed by HPLC)
Table 17 OD at 610 nm of 3% WPI solutions after heating
tprotein precipitated and settled out.
Example 13 a) Removal of glyco-CMP from rennet whey at pH 4.9 (Figure 5a)
Freshly prepared rennet whey (0.56% protein) was adjusted to pH 4.9 with 10% HCl and 500 mL passed through a 12 mL column of QMA Spherosil (6.2 g) which had been previously washed with 0.5 M NaCl and water. A flow rate of 0.6 mL/min was used and 25 mL fractions collected. HPLC analysis of 20 DL samples showed that the B and A variants of aglyco-CMP broke through the column after 200 and 250 mL had been loaded respectively and very quickly rose to concentrations greater than that present in the starting whey. Some of the glyco-CMP components had still not broken through even after 500 mL of whey had been loaded. After loading, the column was washed with water. All of the column-passed whey fractions and wash water were combined (520 mL) and pH adjusted to 6.6 with 10% NaOH for further processing to recover WPI (part b, below).
The acidic peptides were recovered from the column by passing 50 mL of 0.5 M NaCl through it. This solution was neutralized, dialysed and freeze-dried for further analysis. It was found to have a sialic acid content of 5.6%. Its protein composition is shown in Table 18.
b) WPI** from glyco-CMP reduced rennet whey
The glyco-CMP reduced whey from part a) was passed through a 39 mL column of QMA Spherosil (6.2 g) at a flow rate of 3.5 mL/min. The column was washed with 30 mM NaCl. The combined column-passed whey and washings contained only 15% of the protein in the whey used at the start of part a).
WPI** was recovered from the column by circulating a further 25 mL of 30 mM NaCl through it whilst maintaining the pH at 2 as described in Example 12 a). The desorbed protein solution was neutralized to pH 6.5 dialysed and freeze-dried for analysis. Its enhanced heat stability is shown in Table 17 and protein composition in Table 18.
Table 18 Protein Composition of WPIs (% of total proteins assayed by HPLC)
Example 14 a) Removal of acidic peptides from acid whey WPC (Figure lb)
A sample of ALACEN™ 342 (WPC containing 80% protein produced from mineral acid whey) was obtained from the New Zealand Dairy Board. A 10% (w/w) solution (1000 g) was prepared and adjusted to pH 4.8 with 10% HCl. The acidic peptides were then removed from 750 g of this using 450 mL of QA cellulose as described in Example 6 a).
b) Recovery of the acidic peptides from the QA cellulose
Water was added to the 800 mL reaction vessel containing the QA cellulose with adsorbed acidic peptides to give a total volume, cellulose plus water, of 800 mL. Aqueous sodium chloride (40 mL of 4 M) was added to make the solution approximately 200 mM and the pH was lowered to 3.0 with 10% HCl. After stirring for 1 hour the
solution of desorbed acidic peptides was drained from the QA cellulose and the latter washed with 200 mM sodium chloride. The combined filtrate and washings were neutralized to pH 6-7 and concentrated by ultrafiltration and diafiltration. The retentate was analysed by HPLC and a sample freeze-dried for phosphate analysis using the method as published in J. Biological Chemistry 235, 769-775, 1960. The result is shown in Table 20.
The HPLC analysis showed that the protein present in this eluate was composed mainly of PP5 (33%) plus the very hydrophilic peptides with a retention time of 3-7 minutes (as for CMP from sweet whey). About 80% of these hydrophilic peptides present in the starting WPC solution appeared in this eluate.
c) Preparation of WPI*
An acid/heat stable WPI (WPI*) was recovered from the combined flow-through retentate and column washings (1006 g) from part a) above using a fresh lot of QA cellulose as described in Example 6 c). The results of the acid/ heat stability test on this WPI* are shown in Table 19 along with a WPI made from the same WPC by anion exchange but without the prior removal of the acidic peptides. The preparation of this WPI is described in section d) below.
d) Preparation of WPI containing acidic peptides (comparative) WPI containing the acidic peptides was prepared from mineral acid
WPC using the same method as described in part c) above except that the feedstream to the anion exchanger was 1200 g of a solution of ALACEN 342 reconstituted at 6.25% solids (5% protein). The WPI was eluted from the QA cellulose in 50 mM sodium chloride at pH 2.5.
Table 19 OD at 610 nm of 3% WPI solutions after heat treatment
t protein precipitated during heating and settled out.
Table 20 Phosphate content, % of powder weight
Example 15 a) Removal of acidic peptides from anion WPI obtained from acid whey (Figure la)
The WPI containing eluate from Example 14 d) was neutralized to pH 6 and concentrated by ultrafiltration and diafiltration. A sample of the final retentate was freeze-dried for phosphate analysis. Half the retentate, i.e. 100 g, was diluted to 460 g with water to make the protein concentration about 3%. The pH of this was lowered to 5.3 with 10% HCl and then it was passed through a 6.5 mL column of Macro-Prep High Q at 2 mL/min. The flow- through containing WPI* was collected. A sample of it was subjected to the acid/heat stability
test of Example 5. The results are shown in Table 19. A further sample of the flow-through was neutralized, freeze-dried and analysed for phosphate content (Table 20).
The acidic peptides were desorbed from the column of Macro-Prep High Q by passing 60 ml of 0.5 M NaCl through it. The eluate was neutralized, dialysed, freeze-dried and analysed for phosphate and the result shown in Table 20.
INDUSTRIAL APPLICATION
The present invention is believed to provide an efficient process of using anion exchange to obtain whey protein containing products from a whey protein-containing feedstocks. In particular WPI's containing mainly β-lactoglobulin or β-lactoglobulin and mainly aglyco-CMP which are heat and acid stable are produced. In addition, a CMP isolate, which may optionally also be enriched in glyco-CMP (and consequently sialic acid) is also provided.
Although the invention has been described with reference to particular embodiments, those persons skilled in the art will appreciate that variations and modifications can be made without departing from the scope of the invention.