WO1983003673A1

WO1983003673A1 - Method for determining the content of collagen in a sample containing connective tissue

Info

Publication number: WO1983003673A1
Application number: PCT/DK1983/000042
Authority: WO
Inventors: N. Foss Electric A/S; Ole-Christian Bjarno
Original assignee: Ole-Christian Bjarno
Priority date: 1982-04-16
Filing date: 1983-04-14
Publication date: 1983-10-27
Also published as: EP0105898A1

Abstract

The content of collagen in a sample containing connective tissues such as a sample of meat or a meat product is determined by treating the sample in comminuted form in an aqueous solution at a temperature of from about 100<o>C to about 200<o>C, in particular about 140<o>C, and at a pH between 4.5 and 6.5, in particular about 5.0, until the collagen in the sample is substantially quantitatively dissolved, removing precipitated components from the resulting suspension, subjecting the solution thereby obtained to protein determination, in particular by IR transmission spectrophotometry, and determining the content of connective tissue in the sample by correlating the protein determination value so obtained to collagen content values determined by a reference method. Alternatively, the protein determination may be performed by the Kjeldahl method. The buffer may be a tartaric acid/tartrate buffer. The method has the advantage of being considerably faster than known reference methods such as hydroxy proline determination.

Description

Method for Determining the Content of Collagen in a Sample Containing Connective Tissue

Meat consists of a number of various proteins, muscle proteins con- sisting of myosin and actin , sarcoplasma proteins, i . e. , the proteins which are dissolved in meat juice, and blood proteins, i . e. , plasma blood proteins and haemoglobin . Finally, there are connective tissue proteins consisting of collagen as the most important component and elastin . Apart from these, foreign proteins may be present in the meat products, primarily soy protein and casein .

For some time, there has been a desire to characterize meat products on the basis of their content of muscle protein and content of other proteins than muscle protein, the other proteins being considered as having less nutritive value. From a nutritional point of view, the amino acid composition of collagen is less advantageous than the amino acid composition of muscle protein . Since at present there is no easily applicable method for analyzing the collagen content, few countries have found it convenient to legislate about the collagen content in meat products although such legislation is being prepared in the EEC and will come into force once the method has been developed.

The standard method of determining collagen utilizes the fact that the amino acid hydroxyproline is found in large quantities in collagen, whereas it is only found in small quantities in muscle protein . In collagen, 12.5% of the total amount of amino acid is hydroxyprolin .

For this reason, the collagen is hydrolyzed, i. e. , decomposed into its amino acids, and the hydroxyproline is measured by means of a colour reaction . As hydroxyproline averages 12.5% of the total collagen con¬ tent, the total collagen content is found by multiplying the hydr¬ oxyproline content by 8.

To determine the hydroxyproline a 5-10 g sample is hydrolyzed for 16 hours at 110°C in hydrochloric acid containing tin chloride. The re-

O _t , V/ leased hydroxyproline is oxidized with chloroamine-T, and the oxida¬ tion product is reacted with 4-dimethyIaminobenzaldehyde resulting in a red colour which is then measured on a spectrophotometer at 560 nm.

However, the standard method is subject to uncertainty. Hydroxy¬ proline, multiplied by 8 so as to define the collagen content, is also found in varying amounts in the other meat proteins so that, in varying degrees, it is erroneously ascribed to the collagen value. The factor 8 is also a decided approximation as it ranges from 8.85 to 6.67, depending on the type of connective tissue (J . Sci . Food Agric. 1981 , 32, 539-546) . The same source further mentions that inter¬ ference from other amino acids, especially when measuring tissue with a low collagen content, may result in collagen values overestimated with 15-20%.

Over the years a great deal of work has been done to optimize and speed up the standard method of determining - collagen . Thus, Dr. Arneth from the Bundesanstalt fur Fleischforschung in Kulmbach has suggested the use of perchloric acid instead of hydrochloric acid for hydrolysis so as to reduce the total amount of time needed for ana- lysis to 4 hours. This method, however, has not been widely accept¬ ed, i . a . , because it is unhealthy to use perchloric acid (Deutsche Lebensmittel-Ruπdschau 1979, 75, 14-17) .

Another proposal to speed up the standard method suggests the use of partial hydrolysis so that only part of the hydroxyproline is re- leased from the collagen . This method is maintained to result in rela¬ tively precise measurements with a time for analysis of 1 1/2 hours. However, this method has not been adopted either (Stern et a , U . S. Patent 4,014,646 (1978) ) .

Instead of using hydroxyproline to indicate the content of connective tissue it has also been attempted to use special substances to indicate muscle proteins . Thus, it has been attempted to use creatϊπin (J . Sci . Food Agric. 1981 , 32, 1033) and, at present, experiments with methyl histidine are being conducted in the Netherlands (Dr. Ols- mann, TNO, the Netherlands) . A recent comprehensive review by David J . Etheriπgton & Trevor J . Simms (J . Sci . Food Agric. 1981 , 32, 539-546) concerning the de¬ termination of collagen concludes that, in practice, only the method comprising 16 hours of hydrolysis followed by determination of hydr- oxyprolin via a colorimetrical reaction is used although a number of other methods, i . a . ELISA and Radioimmunoassay, may be used. Further, it is concluded that if a new and more cogent legislation is introduced which requires that the collagen content of meat products be known with certainty it will increase the need for routine methods which are simple, reliable and sufficiently fast to avoid unnecessary delays in the production process .

Collagen has the quality of being transformed into water-soluble gelatine at high temperatures. This is a quality known, e.g . , from the preparation of brawn . Brawn consists of meat and connective tissue from pigs' heads, and the ability of collagen to be transformed into gelatine when boiled and then to set when cooled to form the well-known gelatinous structure is utilized.

This quality of collagen has been the basis of a number of investiga¬ tions into the possibility of separating collagen from muscle protein . Thus, Striegel, (Chemiker-Zeitung, 14th April, 1917) boiled the collagen-containing sample in water for 4-5 hours at reflux tempera¬ ture, added tartaric acid to the mixture which was then further boiled for about 30 minutes, neutralized the solution with NaOH to precipitate albumin, precipitated meat protein fractions by adding zinc sulphate or copper sulphate, washed the mixture with hot water and filtered, and determined the nitrogen content by means of the Kjel- dahl method . However, the method was characterized as not too precise. Lindner (Zeitschrift Fur Lebensmitteluntersuchung Und Forschung, 90^1950, 345) used a modified version of Striegel's method where the boiling was performed for only 1 hour and the neutralisa¬ tion was performed with calcium carbonate to a pH of 6.5. Kotter (Arch . Lebensmittelhygiene 1958, 9, 243) compared the method of Striegel and Lindner with several other methods and concluded that the method was so much less accu rate than the standard method that it could not be used for reliable testing. After Kotter, no work has been reported on collagen determination according to this strategy. Instead, the possiblility of determining muscle protein has been the object of several investigations .

Thus, Langner (Die Fleischwirtschaft Nr. 10/1970, 1391 ) used a series of filterings and washings with water and other solvents to separate muscle protein from collagen and determined the muscle protein by the Kjeldahl method. In this connection, he found that a buffer solution of 0. 1 molar disodium hydrogen phosphate is optimal and protective for the obtainment of the muscle protein . However, the determination is extremely complicated as it requires a series of washings and manipulations of the fϊltered-off muscle protein to make it possible to determine the muscle protein by means of a Kjeldahl analysis.

The present invention relates to new methods for determining the content of collagen in a sample containing connective tissue such as a sample of a meat product. The methods of the invention utilize the ' same strategy as Striegel-Lindner-Kotter, that is, separation of collagen from muscle protein and determination of the collagen . The methods of the invention involve various considerable improvements in comparison to the known methods and are simple, reliable and accu¬ rate.

An important aspect of the invention relates to a method for determin¬ ing the content of collagen in a sample containing connective tissue, comprising treating the sample in comminuted form in an aqueous solution at a temperature of from about 100°C to about 200°C, until the collagen in the sample is substantially quantitatively dissolved, removing precipitated components from the resulting suspension,

, subjecting the solution thereby obtained to protein determination by

I R spectrophotometry, and determining the content of collagen in the sample by correlating the protein determination value so obtained to collagen content values determined by a reference method . Compared to the known methods, this method is much simpler and faster to perform, and the error arising from the interference from non-protein nitrogen in the Kjeldahl determination is obviated .

I n the present specification and claims, the term "substantially quan- titatively dissolved" is intended to indicate that the collagen is dis¬ solved or extracted from the remaining proteins etc. to such an extent that prolonged treatment will not result in any substantial additional extraction of collagen . In practice, a complete separation between collagen and muscle proteins etc. will usually not be obtained .

The I R spectrophotometry is preferably I R transmission spectropho- tometry at a wavelength in the infrared range or the near-infrared range, that is, in a range from about 3-10 μm and up to the over¬ tones in infrared (1 -3 μm) .

The correlation of the content of collagen in the sample to collagen content values determined by a reference method (such as the standard- method) may be performed by regression analysis or any other suitable correlation method and should be adapted to the particular type of product, the particular type of determination, and the particular type of instrument used. I n the examples which follow, examples of regres- sion equations suitable for the conditions reported in the examples and adaptable to most meat products are given . As is explained in the . examples, it is often preferred that the total protein content of the sample be included in the regression .

The comminution of the sample may be performed in any suitable method, e. g . by means of a meat grinder. The comminution should be performed to such an extent that the subsequent treatment in the aqueous solution becomes as effective as possible.

It is preferred that the aqueous solution has a pH between 4.5 and 6.5, preferably between 5 and 6 as this range is around the isoelec- trie point of proteins where the proteins are least soluble. Advan¬ tageously, a pH in the stated range is obtained by means of a buffer system such as a system comprising tartaric acid and a tartrate such as potassium sodium tartrate. Compared to a phosphate buffer system,

O F the buffer system comprising tartaric acid and a tartrate is particu¬ larly valuable in that it shows little I R absorption in the protein determination . «

It has been found that a pH of between 5 and 6 produces the great- est efficiency in the separation (extraction) of the collagen from muscle protein . This conforms with the fact that the isoelectric point of myosin is 5.3 and the isoelectric point of myogen js 6.2. Thus, the greatest specificity, that is, the greatest degree of solubility for collagen versus the smallest degree of solubility for myosin and yo- gen occurs at a pH between 5 and 6.

However, an optimal separation of muscle protein from collagen is not the only factor to be taken into consideration in the selection of the pH . Another factor is interference from foreign proteins, e.g. soy protein, blood protein , casein and egg protein . In practice, meat samples will often contain about 2 to 3% of such foreign protein . According to the invention, the extraction is suitably- performed at a pH at which interference from foreign protein such as soy protein is minimized. This occurs at a pH of about 5.0 which is therefore the generally preferred pH . A suitable buffer for adjusting the pH to this value comprises 0.01 M of tartaric acid and 0,01 M of tartrate such as potassium sodium tartrate; this buffer combination has the added advantage of optimal buffer capacity of the symmetrical buffer.

The advantages with respect to separation of collagen from muscle protein obtained by using a pH in the range of 4.5 to 6.5 and the resulting improved accuracy of the collagen determination are also obtained when the protein determination is performed by other means than the I R spectrophotometry, e. g. , when the protein determination is performed by a Kjeldahl method. Therefore, another method of the invention is the aspect expressed in claims 10-18.

The treatment in the aqueous solution is preferably performed at a temperature above 100°C, such as in the range of 110-200°C, in particular in the range of 120-180°C, such as 120-160°C, especially a temperature of about 140°C.

( OM S The increased temperature may be obtained by performing the treat¬ ment in a closed vessel in an autoclave, or in an oil bath, or in a block thermostate. The increase in the temperature will result in a dramatic reduction of the time necessary for obtaining a substantially quantitative dissolution of the collagen, and will also result in a much better control of the reaction compared to the known treatments lasting several hours, resulting in increased reproducibility. These beneficial effects are also illustrated in the examples . Another aspect of the invention, therefore, is a method utilizing separation at such increased temperature. This aspect is claimed in claims 20-30.

Due to their simplicity, the methods of the invention permit the use of a larger quantity of sample, e.g. 22 g, compared to the 5 g sample of the standard method. Such larger sample size would be unrealistic in the standard method as it would involve a large consumption of reagent. The larger sample size contributes to greater accuracy.

Thus, the present invention relates to methods for determining the content of collagen in products containing connective tissue, in parti¬ cular meat products, which methods show decided advantages over the standard method in that they are performed in much shorter time and may be performed without the use of expensive or unhealthy chemicals.

Also, the methods of the invention may improve the accuracy of the collagen determination , both compared to the standard method and compared to the previous attempts at measuring collagen .

A number of factors contribute to the improvement of the accuracy of the measurement:

1 . By adjusting the pH to near the isoelectric point of meat proteins, the separation is optimized .

2. By optimizing the pH, the interference from foreign proteins (soy protein , casein , egg protein and blood protein) is minimized.

f 3 RE

. OM 3. The error arising from incomplete separation is reduced by cor¬ recting the measured results with the total protein content.

4. When using I R spectrophotometric determination instead of Kjeldahl determination, the error arising from non-protein nitrogen is obviated.

5. By reducing the number of separation stages from previously 3-5 to 1 , the recovery error is considerably reduced.

6. The greater accuracy of the separation method of the invention is also ascribable to the larger sample quantity which is rendered possible by the present invention .

EXAMPLE 1

Samples were prepared consisting of beef and pork with a content of connective tissue of 0-50% in the case of beef and 0-100% in the case of pork to form a mixed sample test series . The samples were ad ix- tures of pu re meat and pure connective tissue in specified amounts, and the content of collagen in these starting materials had been determined by the standard method.

The samples were carefully homogenized in a meat grinder (Mδhle Boy FW 70 N) until a uniform colour had been obtained.

22 g (± 0. 1 g) batches were weighed out and placed in a screw- capped autoclave bottle Duran-Jena GL 45. 100 ml (± 0. 1 ml) of a reagent (buffer) containing 0.01 M of tartaric acid p. a. and 0.01 M of potassium sodium tartrate p. a . was added whereby the pH of the mixture was set at 5.0.

A 10 liter autoclave (CERTOCLAV type CV 11/1600 W) had been set at 140°C in advance and brought to the boil. The cap of the autoclave bottle was screwed on and the bottle was placed in the autoclave which was then closed. The autoclave was allowed to stand for 15 minutes upon reaching 140°C after which it was switched off and cooled until the pressure dropped to atmospheric pressure. Imme¬ diately after the drop of pressure to atmospheric pressure the bottle was removed, shaken briefly and further cooled by standing for 5 minutes .

The sample was filtered by means of a plastic funnel with a folded filter Machery-Nagel MN 875 1/4 18.5 cm 0. The protein content of the filtrate was measured by I R transmission spectrophotometry on a "Super-Scan", type 10600, from A/S N . Foss Electric, Hillerød, Denmark, without an attached rinsing system. (The rinsing system is not necessary in connection with the present determinations as the liquid to be measu red is a solution , not an emulsion . The avoidance of the rinsing operation increases the capacity of the instrument. )

OMPI ^{v ra} The "Super-Scan" is an I R spectrophotometer described e.g . in the "Super-Scan" Instruction Manual, Issue 3 GB, October 1981 , issued by A/S N . Foss Electric. This instrument is a ^' single beam single cuvette instrument which is principally constructed to determine the content of fat, protein, carbohydrates and water in meat products and which measures protein at a wavelength of 6.50 μm.

Based on the protein determination values (determined as pure pro¬ tein) , the collagen content was assessed by single linear regression (best line fit) .

Also, protein determination was performed on some of the samples by means of a Kjeldahl method (Kjel-Foss apparatus from A/S N . Foss Electric) , and based upon the values determined, the collagen values were assessed by single linear regression (best line fit) .

The results of these determinations were compared with the standard method as performed on the mixtures.

The data and results of the experiments appear from Table 1 , wherein KFA designates Kjel-Foss determination, and I RT designates infrared transmission determination .

Table 1

Amount Heat Residual variation between

Num¬ of treatment Compo¬ calculated and measured ber sample °C/mi- sition values, % collagen nutes KFA I RT Standard

6 11 g 130/30 Beef with 0.52 0.26 0.32 0-50% con¬ nective tissue

6 22 g 140/15 Beef with 0.25 0-50% con¬ nective tissue

6 11 g 130/30 Pork with 0.30 0.35 0.05 0-50% con¬ nective tissue

6 22 g 140/15 Pork with 0.15 0-50% con- • nective tissue

5 22 g 130/30 Pork with 0.20 0.36 0-100% con¬ nective tissue

Average: 0.41 0.24 0.24

It will be noted that the residual variation of the infrared transmis¬ sion measurement was of the same order as that of the standard method. The total time for analysis was 35 minutes, a considerable improvement compared to the standard method which takes 18 hours to perform.

The Kjeldahl determination was considerably improved compared to the Kotter determination : The reproducibility was 5.8% relative, and the correlation was 0.98, as compared to a reproducibility of 24.7% relative and a correlation of 0.95 in the Kotter determination, which was performed at a pH of 2.7.

I EXAMPLE 2

The method of the invention was tested against a number of actual meat products to evaluate the test results that would occur in prac¬ tice. The results of this testing appear from Table 2

The collagen determination was performed as protein determination by infrared transmission spectrophotometry by means of a "Super-Scan" instrument in the same manner as described in Example 1 .

The sample preparation was performed as described in Example 1. The samples were 22 g samples. The autoclave was allowed to stand for 15 minutes at 140°C, whereafter it was cooled until the pressure had decreased to atmospheric pressure. Immediately thereafter, the auto¬ clave bottles were taken out, shaken briefly, and cooled further by standing for 5 minutes. The filtration and measurement were performed as described in Example 1 .

By means of the "Super-Scan" instrument, also % fat (% F) , % total protein (% P) , and % carbohydrate (% C) were determined on the products .

The standard method employed in this example was the one usually employed by the Danish Meat Research Institute which closely re- sembles the International Standard Method. All determinations were carried out as determinations in duplicate.

In order to determine the regression equation of collagen, a multiple regression analysis was employed so that the error arising from incomplete separation was reduced by correcting the test result with the total protein content. Thus, the collagen content was calculated on the basis of the following formula stated in Table 2. However, in this particular instance, the "Super-Scan" had not been calibrated as normally for measuring protein . In a "Super-Scan" instrument correctly calibrated for protein determination, the figures in the equations would be somewhat lower:

% collagen = 0.288 x P_eχtract - 0.045 P^, * 0.20.

-£TJ

O

_4\ . W " Λ Table 2

% Colla- % Collagen Resi-

Types of product % F % P % C gen, IRT Standard duals

Beef

Brisket 28.6 15.7 0 2.1 2.5 -0.4

Neck 15.7 18.7 0 1.6 1.8 -0.2

Trimmings 9.9 20.9 0 3.0 3.2 -0.2

Rumpsteak 14.0 20.6 0 1.8 2.0 -0.2

Rump 12.1 20.8 0 1.9 2.1 -0.2

Ground beef 19.4 18.9 0 2.3 2.6 -0.3

Shoulder 4.3 22.3 0 1.5 1.5 0

Thick flank 2.4 21.7 0 1.2 1.1 ⁺0.1

Outside round 8.5 21.9 0 1.6 1.5 ^♦0.1

Brisket 32.0 15.0 0 1.9 1.8 -0.1

Pork

Tenderloin 4.1 23.6 0 0.8 0.8 0

Collar 14.4 18.3 0 1.4 1.4 0

Inside round 1.6 23.1 0 1.3 1.0 ⁺0.3

Flank 26.9 15.2 0 1.5 1.4 ⁺0.1

Shoulder 9.2 19.2 0 1.0 1.2 -0.2

Ham 4.9 21.1 0 1.1 1.3 -0.2

Shoulder 12.1 17.6 0 1.4 1.4 0

Outside round 5.7 21.1 0 1.4 1.1 ^♦0.3

Outside round 9.2 21.2 0 1.5 1.1 ⁺0.4

Ground pork 22.2 17.2 0 2.0 2.2 -0.2

Salami (1) 41.5 11.4 4.8 1.9 2.4 -0.5

Black pudding (1) 22.2 11.2 35.6 1.4 1.5 -0.1

Hamburger (I) 17.2 16.1 1.3 1.9 2.2 -0.3

Liver paste (1) 14.0 10.0 13.0 1.4 1.3 ⁺0.1

Liver paste (II) 17.7 10.3 9.2 1.7 2.0 -0.3

Liver paste (111) 22.45 9.2 7.5 1.6 1.6 0

Smoked mett- wurst (IV) 26.9 10.9 8.8 2.3 2.2 ⁺0.1

Sausages (V) 23.2 11.7 9.7 2.2 1.7 ⁺0.5

Mettwurst (VI) 22.2 11.0 7.0 1.6 1.1 -0.5

Hamburger (III) 15.8 18.2 4.3 2.4 1.9 ⁺0.5

Hamburger (VII) 20.6 18.7 0.9 2.0 1.9 -0.1

Hamburger (VIII) 16.7 15.0 6.4 2.2 1.7 ⁺0.5

Residual standard deviation 0.28

Reproducibility standard deviation 0.055 0.13

- 0047xP ^* 0.20

Regression equation . y = o.3ooxP_extract

Note: The Roman numerals refer to different suppliers of the product. It appears from Table 2 that the residual standard deviation of the determinations performed according to the invention was on level with that for the standard method, and that the reproducibility of the method of the invention is higher than is the case with the standard method.

^

Claims

CLAIMS .

1 . A method for determining the content of collagen in a sample containing connective tissue, comprising treating the sample in commi¬ nuted form in an aqueous solution at a temperature of from about 100°C to about 200°C, until the collagen in the sample is substantially quantitatively dissolved, removing precipitated components from the resulting suspension, subjecting the solution thereby obtained to protein determination by I R transmission spectrophotometry, and determining the content of connective tissue in the sample by corre- lating the protein determination value so obtained to collagen content values determined by a reference method.

2. A method according to claim 1 wherein the aqueous solution has a pH between 4.5 and 6.5.

3. A method accordig to claim 2 wherein the aqueous solution has a pH between 5 and 6

4. A method according to claim 3 wherein the aqueous solution has a pH of about 5.0.

5. A method according to any of claims 2-4 wherein the aqueous solution is buffered with a tartaric acid/tartrate buffer.

6. A method according to claims 4 and 5 wherein the aqueous solution is 0.01M with respect to tartaric acid and 0.01M with respect to tartrate.

7. A method according to any of the preceding claims wherein the treatment in the aqueous solution is performed at a temperature in the range of 120-180°C.

8. A method according to claim 7 wherein the temperature is in the range of 120-160°C.

9. A method according to claim 8 wherein the temperature is about 140°C.

10. A method for determining the content of connective tissue in a sample containing connective tissue, comprising treating the sample in comminuted form in an aqueous solution having a pH between 4.5 and 6.5 at a temperature from about 100°C to about 200°C, until the collagen in the sample is substantially quantitatively dissolved, re¬ moving precipitated components from the resulting suspension, sub¬ jecting the solution thereby obtained to protein determination, and determining the content of connective tissue in the sample by cor¬ relating the protein determination value so obtained to connective tissue content values determined by a reference method.

11 . A method according to claim 10 wherein the aqueous solution has a pH between 5 and 6.

12. A method according to claim 11 wherein the aqueous solution has a pH of about 5.0.

13. A method according to any of claims 10-12 wherein the aqueous solution is buffered with a tartaric acid/tartrate buffer.

14. A method according to claims 12 and 13 wherein the aqueous solution is 0.01M with respect to tartaric acid and 0.01M with respect to tartrate.

15. A method according to any of claims 10-14 wherein the treatment in the aqueous solution is performed at a temperature in the range of 120-180°C.

16. A method according to claim 15 wherein the temperature is in the range of 120-160°C.

17. A method according to claim 16 wherein the temperature is about 140°C.

18. A method according to any of claims 10-17 wherein the protein determination is performed by the Kjeldahl method.

19. A method according to any of claims 10-17 wherein the protein determination is performed by I R transmission spectrophotometry.

20. A method for determining the content of connective tissue in a sample containing connective tissue, comprising treating the sample in comminuted form in an aqueous solution having a temperature from about 110°C to about 200°C, until the collagen in the sample is sub¬ stantially quantitatively dissolved, removing precipitated components from the resulting suspension, subjecting the solution thereby ob¬ tained to protein determination, and determining the content of con- nective tissue in the sample by correlating the protein determination value so obtained to connective tissue content values determined by a reference method.

21 . A method according to claim 20 wherein the treatment in the aqueous solution is performed at a temperature in the range of 120- 180°C.

22. A method according to claim 21 wherein the temperature is in the range of 120-160°C.

23. A method according to claim 22 wherein the temperature is about

140°C.

24. A method according to any of claims 20-23 wherein the aqueous solution has a pH between 4.5 and 6.5.

25. A method accordig to claim 24 wherein the aqueous solution has a pH between 5 and 6.

26. A method according to claim 25 wherein the aqueous solution has a pH of about 5.0.

27. A method according to any of claims 24-26 wherein the aqueous solution is buffered with a tartaric acid/tartrate buffer.

OM v.-i

^* '

28. A method according to claims 26 and 27 wherein the aqueous solution is 0.01M with respect to tartaric acid and 0.01M with respect to tartrate.

29. A method according to any of claims 20-28 wherein the protein determination is performed by the Kjeldahl method.

30. A method according to any of claims 20-28 wherein the protein determination is performed by I R transmission spectrophotometry.

31 . A method according to any of the preceding claims wherein the sample is a sample of at least 10 g, preferably a sample of at least 15 g, in particular a sample of at least 20 g, such as a sample of about 22 g.

^■ ξ_

^ -φ?