WO1996012730A1

WO1996012730A1 - Separation of amino acids and peptides from protein hydrolysates

Info

Publication number: WO1996012730A1
Application number: PCT/NZ1995/000107
Authority: WO
Inventors: Roger Anthony Stanley; Dawn Marie Scott; Elizabeth Emma Doolin
Original assignee: Industrial Research Limited
Priority date: 1994-10-20
Filing date: 1995-10-20
Publication date: 1996-05-02
Also published as: AU3818095A

Abstract

A method for the separation of hydrophobic amino acids and peptides containing aromatic groups from protein hydrolysates by the use of anion exchange media at substantially neutral pH and the regeneration of the media for further use.

Description

Separation of amino acids and peptides from protein hydroiysates

The invention relates to a method of separating amino acids and peptides from protein hydroiysates. In particular the invention relates to the removal of hydrophobic amino acids

and peptides containing aromatic groups from protein hydroiysates by the use of anion exchange media thus improving the taste quality of the hydrolysate.

Background Art

Hydroiysates are produced by enzymatic cleavage of proteins to manufacture products with

lowered allergenicity and improved nutritional availability (for example hydroiysates of casein, whey and soy proteins are produced commercially). Such hydroiysates are desirable products for use with both humans and animals. In particular, such hydroiysates are useful when the subject has digestive problems such as an incapability to break down whole proteins in the digestive tract. However the action of the enzymes during the hydrolysis process causes the release of hydrophobic peptides which are perceived as bitter. As a result, when used in food products, the maximum level of incorporation of such hydroiysates is limited by the bitter taste. Therefore for oral administration, it is desirable that the amount of these compounds in the product be lowered to achieve an acceptable taste.

It is known that bitter hydrophobic peptides can be lowered in concentration or removed by the use of adsorbents such as activated carbon, or synthetic resins such as XAD4 and XAD7 (Rohm and Haas) or HP20 and HP21 (Mitsubishi Kasei Corporation) where the base matrix is composed of a polymer with hydrophobic properties such as styrene-divinyl benzene, acrylate or methacrylate. Indeed a number of studies and much research has been directed to the debittering of protein hydroiysates using hydrophobic adsorbents. United States Patent No

4.075,195 to John F Roland entitled "Debittered Protein Product and Method of Manufacture"

discloses a process for the removal of the bitter fraction of a protein hydrolysate by passing the hydrolysate through a bed of phenolic resin particles. United States Patent No 5,266,685 to John T Garbutt entitled "Non-Bitter Protein Hydroiysates" discloses the treatment of a protein hydrolysate with a hydrophobic adsorptive resin which is functional to remove the

protein hydrolysate bitter taste components. In this US patent there is also disclosure of the perceived undesirability of the use of ion-exchange resins for the debittering process. The use of adsorbents does, however, have several disadvantages in commercial production. The synthetic polymer adsorbents, with the exception of XAD16 (Rohm and Haas) or equivalents, are not currently permitted by the Food and Drug Administration (USA) for use in food applications. In addition, the hydrophobic adsorbents are difficult to regenerate. They require excess alkali and possibly solvents during the regeneration thus adding to the cost of the process.

Japanese Patent 59-159792 to Meiji Confectionary KK entitled "Manufacture of Casein Phosphopeptide" is directed to a new method of manufacture of Casein Phosphopeptide via the hydrolysis of casein by trypsin. The bitter taste, which is caused by the presence of peptides in the trypsin hydrolysate is removed by the use of activated carbon or cation exchange resins. Anion exchange resins are disclosed as being unsuitable for use in the removal of the bitter peptides from the mixture. The patent discloses the reclaiming of the column through the suitable treatment with acid/alkali or by calcination regeneration. This Japanese patent therefore discloses the use of cation exchange resins in connection with the removal of the bitter taste associated solely with trypsin hydroiysates. The disclosure does not teach the applicability of anion-exchange media in connection with the removal of specific amino acids and peptides from a mixture of protein hydroiysates and also does not teach how regeneration of the anion-exchange media can be achieved. No further reference to this regeneration process is made.

It is therefore an object of an invention to provide an improved method for separating hydrophobic amino acids and peptides containing aromatic groups from protein hydroiysates

or at least to provide a viable alternative.

Statement of the Invention

The invention in a first aspect comprises a method for separating hydrophobic amino acids and peptides containing aromatic groups from protein hydroiysates using anion exchange media.

The invention further comprises a method for separating hydrophobic amino acids and peptides containing aromatic groups from a mixture of protein hydroiysates comprising:

(a) using an anion exchange resin at a substantially neutral pH, and thus in a neutral form, to bind and remove said amino acids and peptides from the mixture of protein hydroiysates; and

(b) removing the bound amino acids and peptides from the anion exchange resin by

lowering the pH to convert the neutral resin to the ionised form.

Preferably the hydrophobic peptides separated from the protein hydroiysates contain one or more amino acid unit(s) with an aromatic group. Preferably the anion exchange resin is a hydrophobic base matrix with a weak base ion- exchange functionality.

Preferably the weak base ion-exchange functionality has a pK_a or pK_a's in the pH range of substantially 2.0 to substantially 9.0, more preferably between 2.0 and 6.0.

Preferably the ion-exchange resin is regenerated by removing the bound amino acids and peptides from the ion exchange resin using a combination of low pH and high temperature.

Preferably the pH at which the resin is used to bind the amino acids and peptides is greater than the pH required to completely ionise the weak base functionality. Preferably this pH is between substantially 4.0 to substantially 9.0 and more preferably between 6.5 and 8.0.

Preferably the pH used to remove the bound compounds from the ion exchange resin is sufficiently low to create an ionised form of the weak base functionality. Preferably this is between substantially 1.5 and substantially 6.0 and more preferably between substantially 1.5 and substantially 4.0.

Preferably the temperature used in the regeneration of the ion exchange resin is between substantially 45°C and 100°C, more preferably between substantially 50°C and substantially 100°C, and most preferably between substantially 60°C and substantially 80°C.

Preferably the protein hydroiysates are derived from animal products or plants. Preferably the protein hydroiysates are hydroiysates of casein, whey, or soy protein.

The invention further comprises a process for the separation of hydrophobic amino acids and proteins containing aromatic groups from protein hydroiysates using a hydrophobic anion exchange resin with a weak base functionality comprising the steps of:

(a) equilibrating or regenerating the resin;

(b) adding the protein hydrolysate to the resin at a pH of between substantially 6.5 to substantially 8.0, at a temperature of between substantially 5°C and substantially 20°C, and at a concentration of between substantially 2% to substantially 20% w/v;

(c) mixing the hydrolysate and the resin;

(d) separating the non-bound product from the resin and rinsing with water; and

(e) regenerating the resin with hot water at a temperature of between substantially 60°C and substantially 80°C and at a pH of between substantially 1.5 and substantially 4.0.

Drawings

The attached Figures show the results of some of the Example processes in graphical form.

In the Figures:

Fig. 1 shows the results of Example 1; Fig. 2 shows the results of Example 2; Fig. 3 shows the results of Example 4; Fig. 4 shows the results of Example 6; and Fig. 5 shows the results of Example 7. Detailed Description of the Invention

The invention is directed to the separation of hydrophobic amino acids and peptides containing aromatic groups from protein hydrolysate mixtures using anion exchange media. In the past,

ion exchange resins generally have not been thought to be suitable for use in large scale separation processes for this purpose as it was thought that sufficient selectivity of binding the compounds of concern to the surface of the resins did not exist for practical industrial production purposes.

It has now been found that the use of anion exchangers with a weak base functionality and hydrophobic base matrix at a substantially neutral pH will bind hydrophobic amino acids and peptides containing aromatic groups to the surface of the resin and thus remove them from the hydrolysate mix (an adsorption stage).

The types of amino acids and peptides that can be selectively removed are in general terms the hydrophobic amino acids and peptides containing aromatic groups. Such compounds include phenylalanine, tyrosine, tryptophan and peptides thereof as well as others as are well known in the art. When used in the dairy industry, this separation method can be advantageously used to de-bitter hydroiysates of casein or whey for example. The bitter taste caused by combinations of amino acids having hydrophobic end groups can be removed if

such compounds are separated relatively selectively from the protein hydroiysates. The use of these hydroiysates in food products, if the bitter taste is reduced, will therefore be greatly enhanced. The process of the invention will not remove all the hydrophobic amino acids peptides however, as only those containing aromatic groups are preferentially removed. Hydrophobic amino acids/peptides containing only straight chain molecules are only removed

6

SUBSTITUTE SKΞET ULE 26 to a very limited extent (see Example 7). Hydrophobic peptides which contain units having aliphatic straight chains will be removed from the hydrolysate provided the peptide also

contains units having aromatic groups. The specificity is directed to amino acid units containing aromatic groups and therefore if these are connected to amino acid units containing aliphatic straight chains, the resultant peptide will also be removed from the hydrolysate

mixture. This can clearly be seen in Example 7.

The surveyed and selected preferred anion exchange resins will have

(1) hydrophobic base matrices;

(2) sufficiently high surface area available for sufficient binding of the amino acids and peptides;

(3) charge purity - no quaternary charged groups associated with the resin, that, at neutral . pH, would be carrying a charge which would interfere with the hydrophobic interaction. This may be compared to most commercial weak anion exchange resins which have some quaternary charge associated with them as a side product of manufacture; and

(4) charge density - sufficient density of chargeable groups on the surface of the resin to be effective in the regeneration of the resin on the lowering of the pH.

The ion-exchangers useful in the process of the invention are generally comprised of a weak base group such as a primary, secondary or tertiary amine with a pK, or pK s in the range of pH 2.0 to pH 9.0 that are attached to a particulate matrix with hydrophobic character such

as styrene divinyl benzene. Alternative combinations of weak base groups and particulate matrices as will be known in the art may also be used. Anion exchange resins are subtantially neutral in charge when the pH of the equilibrating solution is above the pK_a of the base ligand. The preferred anion exchange resins are those with weak base ion-exchange groups that are substantially neutral at neutral pH. The more the equilibrating pH is above the pK_a of the weak anion exchange resin, the lower the density of charged groups on the surface of the resin.

Therefore at neutral pH, the preferred resins for this invention are substantially neutral, enhancing the interaction between the hydrophobic amino acids and peptides and the resin. The pH range can be as wide as pH 4.0 to pH 9.0 depending on requirements as will be known in the art. More neutral pH's between substantially 6.5 and 8.0 are however preferred.

The most suitable resins will have all the above properties. Exemplary commercial resins include: Relite A329 from Sybron/Relite, which is the best of the surveyed resins. Others, such as Purolite A 103, Purolite A 100 (Purolite International), Amberlite IRA93SP (Rohm and Haas), Dowex MWA1 (Dow Chemical Company) are also suitable, but are not as effective as Relite A329 because their binding capacities are less. Other suitable resins as will be known in the art may also be used.

The process of the invention includes the ability to regenerate the anion exchange resin. This relies on the ability to alter the surface charge of the resin, and thus the hydrophobicity and strength of adsorption of the hydrophobic substances onto the resin. This is achieved by lowering the pH of the solution to substantially lower than the pK„ of the ion-exchange groups on the resin. This induces charged groups on the resin surface thus releasing the bound substances. The resin used in the process of the present invention can therefore be regenerated by lowering the pH surrounding the resin, preferably at high temperature, causing the amino acids and peptides to unbind from the resin surface thus regenerating the resin for further use (a regeneration step). The pH can be lowered by the use of any suitable acid or buffer (eg HC1,

H₂SO₄ etc) as will be known in the art. Preferably the pH is lowered to between substantially 1.5 to 6.0 although a pH between 1.5 and 4.0 is considered most suitable. Further a temperature of between 45°C and 100°C enhances the regeneration step. This may be achieved by the use of hot water or the like as will again be known in the art. Temperatures between 50°C and 100°C are preferred with temperatures between 60°C and 80°C being considered most suitable. The hydrophobic ion-exchangers with weak base functionality have been substituted for conventional non-ion exchange adsorbents with the amino acids binding to the ion- exchanger in the uncharged form (ie at the substantially neutral pH). A shift to lower pH to

convert the ion-exchanger to the charged form results in the release of the bound amino acids.

It is thought that the acid used in the regeneration step also neutralises the carboxyl groups of the amino acids and peptides giving them a nett positive charge which is repelled by the nett

positive charge of the ion exchange resin. In this way the process of the present invention can be carried out with resins that have an established history and acceptability of use in food processes. Moreover, the resins can be readily regenerated for reuse without the application of strong caustic, acids or toxic solvents, again giving better acceptability in food applications for example. The present invention requires only sufficient acid to alter the charge on the anion exchange resin to result in release of the bound substances. It does not require excessive amounts of acids that may be recommended to clean ion exchange media on an empirical basis. In conventional ion exchange processes anion exchange resins are regenerated by the use of strong alkali. While the method of the present invention operates preferentially at low temperatures for the adsorption stage and at high temperatures for elution (ie the regeneration step), the process will

work, albeit at reduced efficiency, at almost any temperature of liquid water. It may actually be beneficial to move outside the preferred temperatures to fit process requirements and restrictions. For example, there may be advantages in operating the process at temperature extremes to prevent bacterial growth in the system used to operate the process. Such

temperature extremes would be well known to persons skilled in the art.

The use of ion exchange resins in such separation processes has several advantages over the use of adsorbents. Such advantages include:

(i) ion exchange resins do not easily foul with large molecules because the charge on the resin covers most of the surface and only smaller molecules, such as the smaller peptides, will be able to bind in the spaces between the charges on the resin surface;

(ii) more ion exchange resins are permitted for food use applications by the Food and Drug Administration (USA) and their cost is generally subtantially lower than hydrophobic adsorbents;

(iii) by lowering the pH at high temperature, the ion exchange resins can readily be regenerated for further use;

(iv) ion exchange resins operated according to the protocol are relatively specific and will not remove all of the peptides from the hydrolysate mixture; and

(v) the ion exchangers can be operated in the neutral range and not in the hydroxide form eliminating base catalysed amino acid degradation which allows the recovery of the bound amino acids for potential later use. Preferably the process is carried out by first washing the ion-exchanger according to manufacturer's recommendation and then washing in the presence of a salt (e.g. NaCl or the like). The ion-exchanger is then washed with buffer or water to achieve the desired pH for adsorption of amino acids and peptides, which is above the pK_a of the weak base ligands

attached to the resin. The hydrolysate, dissolved in water at the appropriate equilibration pH, is then contacted with the resin in stirred batch mode or in column mode at a ratio of resin

to hydrolysate sufficient to promote adsorption of the desired prr - rtion of hydrophobic amino acids. Such ratios would be well readily determinable to a ; n skilled in this art. The

resin is then drained and washed free of unbound hydrolysate product and the bound molecules are eluted with a regenerant, such as a buffer or acid, at a pH low enough to convert

substantially all the weak base groups on the resin to the charged form. It is preferable that the adsorption step be carried out at low temperature to promote adsorption of the hydrophobic amino acids, and that the regeneration be carried out at high temperature to facilitate elution of the hydrophobic amino acids.

The resins can be washed with water and re-used without further equilibration if the "irocess

hydrolysate has significant buffering capacity to result in an equilibrium at a substantially neutral pH. If this is not the case, sufficient alkali must be added to achieve this.

EXAMPLES

To demonstrate the general principles involved in using the described invention, model bitter substances commonly associated with hydroiysates were used in the following examples. Such substances include L-Tryptophan, an aromatic amino acid with hydrophobic properties, and its derivative L-Tryptophan methyl ester. In the latter the carboxylic acid group is derivatised to the non-charged methyl ester form, eliminating any charge effects of this group.

EXAMPLE 1 - L-TRYPTOPHAN UPTAKE ONTO A RANGE OF SYNTHETIC RESINS

The following experiment was performed to determine the extent of binding of a hydrophobic

amino acid, at neutral pH, to a range of resins having different base matrix properties and varying strengths of ionic functionality. Synthetic polymeric resins based on polystyrene and acrylic are more hydrophobic than resins based on agarose, for example.

PROCEDURE

Binding of L-Tryptophan (L-Trp) to a number of synthetic weak and strong anion exchange resins, a synthetic hydrophobic adsorbent, and a hydrophilic weak anion exchange resin was tested as follows. Solutions of L-Trp (0.5% w/v) were prepared in 0.01M potassium phosphate, 0.5M NaCl buffer, pH 7.4, and mixed in batch mode with samples of each of the resins listed in Table 1 (10 mL/g wet resin) at 30°C for 16 hours.

The amount of non-bound amino acid was determined by OD measurement of the solutions at 280 nm, and from that, the amount of amino acid bound to each resin was calculated. The results are shown in Table 1 and Figure 1. Table 1 : Binding of L-Tryptophan to a series of resins to investigate the effects of charge type and base matrix properties on binding capacity.

Resin Matrix type Resin type L-Tryptophan bound - mg/g wet resin

Diaion HP20 Styrene-DVB Hydrophobic 19.0 adsorbent

Pharmacia DEAE Agarose Weak anion 0.2 Sepharose exchanger

Purolite A860 Acrylic Strong anion 0.0 exchanger

Amberlite IRA958 Acrylic-DVB Strong anion 3.5 exchanger

Purolite A 107 Styrene-DVB Weak anion 23.3 exchanger

Diaion WA30 Styrene-DVB Weak anion 12.1 exchanger

Dowex MWA1 Styrene-DVB Weak anion 10.6 exchanger

Amberlite IRA93SP Styrene-DVB Weak anion 13.6 exchanger

Purolite A 103 Styrene-DVB Weak anion 13.2 exchanger

Purolite A100 Styrene-DVB Weak anion 9.3 exchanger

Purolite A 105 Styrene-DVB Weak anion 9.3 exchanger

Relite A329 Styrene-DVB Weak anion 14.5 exchanger

DISCUSSION

Under these conditions, the hydrophobic adsorbent and the eight synthetic weak anion exchange resins bound the greatest amounts of L-Trp with varying efficiencies. To the contrary, the two synthetic strong anion exchange resins and the hydrophilic agarose resin, bound very minimal amounts of L-Trp.

The results demonstrate the ability of the tested synthetic resins with hydrophobic base matrices and weak anion exchange functionalities to adsorb bitter hydrophobes to the same extent as hydrophobic adsorbents conventionally used for this purpose. Under the conditions of this Example, the strong anionic functional groups on the strong anion exchange resins interfere with the interaction.

EXAMPLE 2 - EFFECT OF PH ON ELUTION OF L-TRYPTOPHAN FROM AN ADSORBENT AND A WEAK ANION EXCHANGE RESIN

The purpose of this experiment was to determine optimal conditions for regeneration of weak anion exchange resins after they have been used to selectively remove bitter components from

a process solution. To do this, the effect of pH was studied on the desorption of a hydrophobic amino acid from a weak anion exchange resin, and was compared to the effect of pH on the regeneration of an adsorbent.

PROCEDURE

Relite A329 was washed with 0.5M NaCl for 1 hour, rinsed with distilled water, and then equilibrated into 0.01M potassium phosphate buffer, pH 7.4. Diaion HP20 was equilibrated into 0.01 M potassium phosphate buffer, pH 7.4. L-Tryptophan (L-Trp, 0.5% w/v in 0.01M potassium phosphate buffer, pH 7.4) was mixed with each resin in batch mode (10 mL L- Trp/g wet resin) at 30°C for 2 hours, and then filtered off. The resins were washed with the equilibration buffer. Samples of both resins, bound with L-Trp, were suspended in water and then titrated to the following pH values using either 0.1M HC1 or 0.1 M NaOH as appropriate:

pH 2, 4, 6, 9, 10. The amounts of L-Trp released from the resins were measured by OD of

the solutions at 270 nm against a standard curve of L-Trp. The results are shown in Table 2 and Figure 2.

Table 2

Resin Final pH ^τ -Tryptophan released g/g wet resin

Relite A329 1.99 0.5

Relite A329 4.06 10.3

Relite A329 6.03 7.5

Relite A329 6.87 5.5

Relite A329 9.08 4.5

Relite A329 10.1 3.6

Diaion HP20 1.99 8.4

Diaion HP20 3.99 8.3

Diaion HP20 6.00 7.3

Diaion HP20 7.47 7.8

Diaion HP20 9.04 7

Diaion HP20 10.01 9

DISCUSSION

It should be noted that 100% recovery of L-Trp was not obtained from either of the two test resins under the conditions used, because in batch processes an equilibrium is formed between bound and free amino acid. The drop in pH had a significant effect on the desorption of L-Trp from the weak anion exchange resin, with the optimal pH for regeneration of this resin being <pH 4. Desorption

of L-Trp from the adsorbent resin, however, was independent of pH, and was less than the maximum amount released from the anion exchange resin.

The results show the advantage of using the ionic functionality of the anion exchange resins at low pH to regenerate the resin and release the adsorbed bitter components.

EXAMPLE 3 - EFFECT OF pH. TEMPERATURE AND SALT CONCENTRATION ON ELUTION OF L-TRYPTOPHAN FROM AN ADSORBENT AND A WEAK ANION EXCHANGE RESIN

To further optimise the conditons for regeneration of weak anion exchange resins, the following experiment was performed.

PROCEDURE

Relite A329 and Diaion HP20 were prepared as outlined in Example 2, and L-Trp was again bound to both resins under the conditions described. Samples of both resins bound with L- Trp, were suspended in water. Half of the resin samples were titrated to pH 4 using 0.1M HC1 and the other half remained at pH 7.4. Resin samples at both pH conditions were mixed at 20°C, 50°C and 70°C for 1 hour. L-Trp released from the resins was removed by filtration and measured by OD at 270 nm. A sample of each resin was suspended in 0.1M NaCl, instead of water, and was mixed at pH 7.4 and 20°C. Again the released L-Trp was measured by OD at 270 nm. The results are shown in Table 3.

16 P. LE 25 Table 3

Resin pH Temperature NaCl added L-Tryptophan °C released mg/g wet resin

Relite A329 4.0 20 No 12.9

Relite A329 4.0 50 No 17.1

Relite A329 4.0 70 No 20.7

Relite A329 7.4 20 No 7.8

Relite A329 7.4 50 No 1 1.2

Relite A329 7.4 70 No 15.6

Diaion HP20 4.0 20 No 9.9

Diaion HP20 4.0 50 No 13.0

Diaion HP20 4.0 70 No 17.2

Diaion HP20 7.4 20 No 10.4

Diaion HP20 7.4 50 No 13.1

Diaion HP20 7.4 70 No 19.0

Relite A329 7.4 20 Yes 8.8

Diaion HP20 7.4 20 Yes 9.8

DISCUSSION

Optimal release of L-Trp from the weak anion exchange resin was achieved with a combination of low pH (pH 4.0) and high temperature (70°C). Release of L-Trp from the adsorbent was significantly improved with increasing temperature, but unaffected by a drop in pH. As in Example 2, a shift in pH to ionise the weak anionic resin, provided a mechanism for release of the hydrophobic amino acid.

The presence of salt at a concentration typically present in milk had very little effect on the release of L-Trp from either resin. EXAMPLE 4 - EFFECT OF TEMPERATURE AND pH ON THE ELUTION OF L-

TRYPTOPHAN METHYL ESTER FROM RELITE A329

PROCEDURE

Relite A329 was treated in the following manner: washed with hot water (70°C), equilibrated in 0.5M NaOH, washed with cold distilled water (20°C), re-equilibrated into 0.5M NaCl and then given a final rinse with distilled water (20°C).

lOmL volumes of L-Tryptophan methyl ester (L-TrpME, 0.5%w/v in 0.01M potassium phosphate buffer, pH 6.5) were added to 1 gram samples of washed resin and mixed for 1 hour at 20°C. The resin samples were filtered to remove unbound L-TrpME and washed with 0.0 IM potassium phosphate buffer, pH 6.5. A number of eluents were added to the resins and shaken for 1 hour at either 20°C or 80°C (see Table 4). L-TrpME released from the resin samples was removed by filtration and measured by OD at 270 nm against a standard curve of L-TrpME. The pHs of the final soltuions were measured. The results are shown in Table 4 and Figure 3.

TABLE 4 - THE EFFECT OF TEMPERATURE AND pH ON THE ELUTION OF L- TRYPTOPHAN METHYL ESTER FROM RELITE A329

Eluent Temperature Final pH L-TrpME released °C mg/g wet resin

0.50M NaOH 20 13.4 5.1

0.05M NaOH 20 12.4 3.8 0.0 IM Phosphate, 20 7.1 9.3 pH 6.5

Distilled water 20 6.6 10.4

0.05M HC1 20 2.9 15.9

0.50 HC1 20 <1.0 16.9

0.50M NaOH 80 13.3 9.4

0.05M NaOH 80 12.5 7.5

0.0 IM Phosphate, 80 6.6 17.4 pH 6.5

Distilled Water 80 6.3 18.0

0.05M HC1 80 3.0 22.6

0.50M HC1 80 <1.0 23.8

DISCUSSION

The results show a dramatic increase in the release of L-TrpME from Relite A329 as the pH was lowered right down to pH 1. As in Example 3, the regenerating properties at low pH were enhanced at high temperature (80°C). Interestingly, strong alkali was ineffective at regenerating the resin.

EXAMPLE 5 - EFFECT OF RESIN POROSITY AND CHARGE DENSITY ON THE BINDING AND ELUTION OF L-TRYPTOPHAN METHYL ESTER

PROCEDURE

The resins used in this example were washed in 0.5M NaOH, rinsed with distilled water, and equilibrated into 0.5M NaCl. Prior to use, they were given a final rinse in distilled water. 0.5%w/v solutions of L-TrpME in 0.01M potassium phosphate buffer, pH 6.5, were added to resin samples (lOmL/g wet resin) and mixed for 1 hour at 20°C. The resin samples were filtered to remove unbound L-TrpME, and were washed with buffer.

Desorption of the amino acid from the resins was performed by mixing the resin samples with either 0.05M HCl or 0.01 M potassium phosphate buffer, pH 6.5. for 1 hour at either 20°C or

80°C (See Table 5). The amounts of L-TrpME bound to, and released from, the resins were

measured by OD at 270nm against a standard curve of L-TrpME. The results are shown in

Table 5.

TABLE 5: BINDING AND ELUTION OF L-TRYPTOPHAN METHYL ESTER

Resin L-TrpME Eluent Temperature L-TrpME bound mg/g of elution released wet resin °C mg/g wet resin

Ionac AFP329 5.7 Phosphate 20 4.5

Ionac AFP329 5.8 Phosphate 80 4.1

Ionac AFP329 3.8 HCl 20 6.8

Ionac AFP329 4.7 HCl 80 6.2

Relite A329 19.8 Phosphate 20 12.4

Relite A329 22.5 Phosphate 80 15.6

Relite A329 20.4 HCl 20 20.2

Relite A329 20.5 HCl 80 20.7

Macronet 32.9 Phosphate 20 0 MN100

Macronet 33.5 Phosphate 80 0.5 MN100

Macronet 35.1 HCl 20 17.5 MNIOO

Macronet 35.5 HCl 80 16.9 MN100

Macronet 33.7 Phosphate 20 0.1

MN150

Macronet 30.6 Phosphate 80 0.1 MN150 Macronet 32.0 HCl 20 23.5 MN150

Macronet 31.7 HCl 80 21.1 MN150

Macronet 36.1 Phosphate 20 0.4 MN200

Macronet 35.9 Phosphate 80 0.1 MN200

Macronet 33.3 HCl 20 12.4 MN200

Macronet 35.1 HCl 80 13.1 MN200

Macronet 30.5 Phosphate 20 0.5

MN250

Macronet 29.2 Phosphate 80 0.2 MN250

Macronet No result HCl 20 No result MN250

Macronet 31.9 HCl 80 15.3

MN250

Amberlite 30.8 Phosphate 20 2.9 XAD16

Amberlite 30.2 Phosphate 80 3.6 XAD16

Amberlite 27.2 HCl 20 16.2 XAD16

Amberlite 29.5 HCl 80 15.8 XAD16

DISCUSSION

This example illustrates two important effects: the importance of resin surface area for binding capacity, and the importance of charge density for desorption during regeneration of the resin. The primary differences between resins Ionac AFP329 and Relite A329 are porosity and surface area. The former resin, with very low surface area, was unable to adsorb L-TrpME, and clearly demonstrated the requirement of the resin to have significant porosity for the process of the present invention. On the other hand, the Purolite Macronet resins have very high surface area and porosity, and thus displayed adsorptive properties for this amino acid,

as good as, or even better than, the conventional adsorbent Amberlite XAD 16.

However, because the Macronet resins have either no surface charge (MN 200 and MN250) or only low densities of charged groups (MN100 and MN150, 0.4-0.6 mmoles/g dry resin), their regeneration properties at low pH were similar to the adsorbent resin, XAD 16. With Relite A329, 100% desorption of L-TrpME was achieved at low pH because of the much greater charge density of this resin (1.2 mmoles/g dry resin). Optimum desorption of the amino acid was achieved with low pH, high temperature and high density of charges on the resin.

EXAMPLE 6 - REMOVAL OF BITTER COMPONENTS FROM CASEIN AND WHEY HYDROLYSATES - COMPARISON OF 2 ADSORBENTS AND 1 WEAK ANION

EXCHANGE RESIN

The aim of this experiment was to investigate whether the bitter components in the casein and whey hydroiysates, ie the hydrophobic amino acids, di and tri peptides, could be as effectively removed from the hydroiysates using a synthetic weak anion exchange resin, as compared to

using conventional hydrophobic adsorbents. PROCEDURE

Relite A329, Diaion HP20 and Diaion HP21 were washed in 0.5M NaOH. rinsed with distilled water, washed with 0.5M NaCl and finally rinsed with distilled water again. Casein hydrolysate and whey hydrolysate ( 10% w/v solutions in water) were mixed with the 3 resins (each at 2 mL hydrolysate/g wet resin, 10 mL/g and 50 mL/g) for 1 hour at 20°C. The non-

bound material was removed from the resins by filtration and taste tested for elimination of bitter taste. OD measurements were performed at 214 nm and 280 nm to monitor levels of amino acids and peptides. The results are shown in Table 6 and Figure 4.

Table 6: COMPARISON OF ADSORBENTS AND A WEAK ANION EXCHANGE RESIN FOR THE REMOVAL OF BITTER COMPONENTS FROM CASEIN AND WHEY

HYDROLYSATES

Resin Solution Wt of Volume OD 280nm OD 214nm Taste applied resin 10%w/v Treated Treated g wet hydrolysate hydrolysate hydrolysate

Diaion Casein 1.0 2 0.312 1.323 Acceptable HP20 hydrolysate

Diaion Casein 1.0 10 0.599 1.998 Acceptable HP20 hydrolysate

Diaion Casein 1.0 50 0.752 2.078 Bitter HP20 hydrolysate

Diaion Whey 1.0 2 0.217 1.268 Acceptable HP20 hydrolysate

Diaion Whey 1.0 10 0.476 1.903 Acceptable HP20 hydrolysate

Diaion Whey 1.0 50 0.563 2.072 Bitter HP20 hydrolysate

Relite Casein 1.0 2 0.322 1.718 Acceptable A329 hydrolysate

Relite Casein 1.0 10 0.543 1.990 Acceptable A329 hydrolysate Relite Casein 1.0 50 0.714 2.088 Bitter

A329 hydrolysate

Relite Whey 1.0 2 0.257 1.679 Acceptable A329 hydrolysate

Relite Whey 1.0 10 0.488 2.007 Acceptable A329 hydrolysate

Relite Whey 1.0 50 0.593 2.036 Bitter A329 hydrolysate

Diaion Casein 1.0 2 0.307 1.322 Acceptable HP21 hydrolysate

Diaion Casein 1.0 10 0.433 1.991 Acceptable HP2 I hydrolysate

Diaion Casein 1.0 50 0.705 2.079 Bitter HP21 hydrolysate

Diaion Whey 1.0 2 0.181 1.128 Acceptable HP21 hydrolysate

Diaion Whey 1.0 10 0.231 1.961 Acceptable HP21 hydrolysate

Diaion Whey 1.0 50 0.413 2.045 Bitter HP21 hydrolysate

None Casein 10 0.710 2.093 Bitter hydrolysate

None Whey 10 0.638 2.070 Bitter hydrolysate

DISCUSSION

According to the taste test, the weak anion exchange resin (Relite A329) was as effective as the 2 adsorbents at reducing the bitter taste of whey and casein hydroiysates. However the capacity of all three resins to remove bitter components was exceeded at a ratio of 5g hydrolysate to lg resin. By OD measurement at 280 nm, Relite A 329 was equally as effective as Diaion HP20, but slightly less effective than Diaion HP21, at removing aromatic amino acids from the hydroiysates. EXAMPLE 7 - SPECIFICITY OF AN ADSORBENT AND A WEAK ANION EXCHANGE

RESIN FOR AMINO ACIDS WITH DIFFERENT STRUCTURES

The purpose of the following experiment was to determine the specificity of a weak anion

exchange resin and of an adsorbent resin for amino acids and peptides with different structures. The extent of binding of these compounds to the resins was determined at both

neutral pH and low pH.

The amino acids tryptophan, phenylalanine and tyrosine contain aromatic rings, whilst valine, leucine and glycine contain aliphatic straight chains. All of these amino acids, and peptides containing these, except glycine, have hydrophobic properties and hence contribute to the bitter taste of hydroiysates.

PROCEDURE

Relite A329 was washed with 0.5M NaOH and then rinsed with distilled water. The pH of

the resin was titrated to pH 7.0 using 0.5M HCl, and then rinsed again with distilled water. Amberlite XAD 16 was used as supplied.

A number of amino acids and peptides were mixed with these two resins using lOmL of 0.5% w/v solutions in 0.01M potassium phosphate, pH 6.5 buffer, and 1 gram samples of resin (Table 7). The samples were shaken at 20°C for 1 hour, and amounts of non-bound material were measured by OD at the optimum UV wavelength for each solution. The pH of the solutions were then lowered to pH 2 using approximately 10 mL of 0.1 M HCl. Again the samples were shaken at 20°C for 1 hour. The amounts of non-bound material were re- measured following the drop in pH. Spectral shifts in optimum UV wavelengths for each of the solutions was taken account of. The proportions of bound amino acids and peptides were

calculated for both pH conditions, and are shown in Table 7 and in Figure 5.

Table 7 : Binding of amino acids and peptides with different structures

Resin Solution applied % Bound % Bound pH 6.5 pH <2

Relite A329 Trp-Leu 76 40

Relite A329 Val-Leu

Relite A329 Phe-Leu 63

Relite A329 Tyrosine 79 61

Relite A329 Leucine

Relite A329 Tryptophan 38 22

Relite A329 L-TrpME 57 14

Relite A329 Glycine

Amberlite XAD 16 Trp-Leu 100 100

Amberlite XAD 16 Val-Leu

Amberlite XAD 16 Phe-Leu 77 49

Amberlite XAD 16 Tyrosine 84 56

Amberlite XAD 16 Leucine 15

Amberlite XAD 16 Tryptophan 73 50

Amberlite XAD 16 L-TrpME 74 73

Amberlite XAD 16 Glycine 12

Trp Tryptophan

Leu Leucine

Val Valine

Phe Phenylalanine

TrpME Tryptophan methyl ester

26 su DCTITI ITP - < ir-r-T- >*» I n r- r» \

U I I I U I C OΠCC I ΠU LC ^U DISCUSSION

Amino acids and peptides containing only straight chain molecules did not bind to either of the two resins tested, showing that the base matrix of these resins is specific to compounds with aromatic properties. As in previous examples with L-Trp and L-Trp ME, the release of amino acids with aromatic structures, with drop in pH, was more dramatic with the weak anion exchange resin than with the adsorbent. This is interpreted as an effect of increased ionic functionality on the surface of the anion exchange resins induced by the addition of acid.

By way of explanation and for general clarity the following table (Table 8) describes the various adsorbents and resins that have been used in the previous Examples. The table details the name of the manufacturer, the resin type, the trade name of the resin, and the matrix type.

Table 8

Resin Manufacturer Resin Type Matrix Type

Diaion HP20 Mitsubishi Kasei Hydrophobic Styrene DVB Corp adsorbent

Diaion HP21 Mitsubishi Kasei Hydrophobic Styrene DVB Corp adsorbent

Amberlite XAD 16 Rohm and Haas Hydrophobic Styrene DVB adsorbent

Purolite A860 Purolite Strong anion Acrylic International exchanger

Amberlite IRA958 Rohm and Haas Strong anion Acrylic DVB exchanger

DEAE Sepharose Pharmacia Weak anion Agarose CL-6B exchanger

Purolite A100 Purolite Weak anion Styrene DVB International exchanger Purolite A 103 Purolite Weak anion Styrene DVB International exchanger

Purolite A 105 Purolite Weak anion Styrene DVB International exchanger

Purolite A 107 Purolite Weak anion Styrene DVB International exchanger

Amberlite Rohm and Haas Weak anion Styrene DVB IRA93SP exchanger

Diaion WA30 Mitsubishi Kasei Weak anion Styrene DVB Corp exchanger

Dowex MWA1 Dow Chemical Weak anion Styrene DVB Company exchanger

Relite A329 Sybron Relite Weak anion Styrene DVB exchanger

Ionac AFP329 Sybron Chemicals Weak anion Styrene DVB Inc exchanger

Macronet MN-100 Purolite Weak anion Styrene DVB International exchanger

Macronet MN-150 Purolite Weak anion Styrene DVB International exchanger

Macronet MN-200 Purolite Hydrophobic Styrene DVB International adsorbent

Macronet MN-250 Purolite Hydrophobic Styrene DVB International adsorbent

DVB = divinyl benzene

The foregoing describes preferred forms of the invention and it is to be understood that the scope of the invention is not to be limited to the specific forms described. Modifications and variations as will be obvious to a person skilled in the art may be made to the forms of the invention as described without departing from the spirit or scope of the invention as defined in the attached claims.

Claims

1. A method for separating hydrophobic amino acids and peptides containing aromatic groups from protein hydroiysates using anion exchange media.

2. The method of claim 1 wherein the anion exchange media is an anion exchange resin.

3. The method of claim 1 or claim 2 wherein the anion exchange media has a hydrophobic base matrix with a weak base ion-exchange functionality.

4. The method of claim 3 wherein the weak base ion-exchange functionality has a pK_a or pK_a's in the pH range of substantially 2.0 to substantially 9.0.

5. The method of claim 3 wherein the weak base ion-exchange functionality has a pI , or pK_j's in the pH range of substantially 2.0 to substantially 6.0.

6. The method of any one of claims 2 to 5 wherein the pH at which the anion exchange media is used to bind the amino acids and peptides is greater than the pH required to ionise the weak base functionality of the media.

7. The method of claim 6 wherein the pH at which the anion exchange media is used to bind the amino acids and the peptides is between substantially 4.0 and substantially 9.0. 8. The method of claim 6 wherein the pH is between substantially 6.5 and substantially

8.0.

9. The method of any one of the previous claims wherein the anion exchange media is regenerated for further use.

10. The method of claim 9 wherein the anion exchange media is regenerated by removing the bound amino acids and peptides from the anion exchange media using a low pH.

1 1. The method of claim 10 wherein the pH used to regenerate the anion exchange media is sufficiently low to create an ionised form of the media's weak base functionality.

12. The method of claim 10 wherein the pH used is between substantially 1.5 and substantially 6.0.

13. The method of claim 10 wherein the pH used is between substantially 1.5 and substantially 4.0.

14. The method of any one of claims 10 to 13 wherein the regeneration of the media uses a combination of low pH and high temperature.

15. The method of claim 14 wherein the temperature used is between substantially 45°C

and substantially 100°C.

16. The method of claim 14 wherein the temperature used is between substantially 50°C and substantially 100°C.

17. The method of claim 14 wherein the temperature is between substantially 60°C and

substantially 80°C.

18. A method of separating hydrophobic amino acids and peptides containing aromatic groups from a mixture of protein hydroiysates comprising:

(a) using an anion exchange resin at a substantially neutral pH, and thus in a substantially neutral form, to bind and remove said amino acids and peptides from said mixture or protein hydroiysates; and

(b) removing the bound amino acids and peptides from the anion exchange resin

by lowering the pH to convert the substantially neutral resin to the substantially ionised form.

19. The method of claim 18 wherein the pH used in step (a) is between substantially 4.0 and substantially 9.0.

20. The method of claim 18 wherein the pH used in step (b) is between substantially 1.5

and substantially 6.0.

21. A method for the separation of hydrophobic amino acids and proteins containing aromatic groups from protein hydroiysates using a hydrophobic anion exchange resin with a weak base functionality comprising the steps of:

31

SUBSTITUTE SHEET RULE 25 (a) equilibrating or regenerating the resin;

(b) adding the protein hydrolysate to the resin at a pH of between substantially 6.5 to substantially 8.0. at a temperature of between substantially 5°C and substantially 20°C, and at a concentration of between substantially 2% to substantially 20% w/v;

(c) mixing the hydrolysate and the resin;

(d) separating the non-bound product from the resin and rinsing with water; and

22. The method of any one of the preceding claims wherein the protein hydroiysates are derived from animal products or plants.

23. The method of any one of the preceding claims wherein the protein hydroiysates are hydroiysates of casein, whey, or soy protein.

24. A method of separating hydrophobic amino acids and peptides containing aromatic groups from protein hydrolysate substantially as herein described with particular reference to the Examples and attached Figures.

25. Protein hydrolysate when formed by the method of any one of the previous claims.

26. A food product containing protein hydrolysate when formed by the method of any one of claims 1 to 20.