MEASUREMENT OF COLOUR PARAMETERS AS AN INDEX OF THE HYDROXYMETHYLFURFURAL CONTENT IN THE UHT STERILISED MILK DURING ITS STORAGE

Dorota Cais-Sokolińska, Jan Pikul, Romualda Danków

ABSTRACT

The purpose of the investigations was to assess colour parameters and the content of hydroxymethylfurfural as indicators of the course of the Maillard’s reaction in UHT sterilised milk and to ascertain the extent of their mutual correlations. The UHT sterilised milk was stored for 24 weeks at temperatures of 4, 8 and 20°C. Colour L*, a* and b* parameters were measured instrumentally and, on their basis, the value of milk colour saturation C* as well as the difference in relation to the model of the ideal whiteness DE were calculated. The amount of total and free hydroxymethylfurfural (HMF) in milk was determined. From among all the examined milk samples, the milk stored at the lowest of the applied temperatures was characterised by the lowest values of the b* parameter, of the C* colour saturation and the difference in relation to the model of ideal whiteness DE. The concentration of the total and free HMF in milk intensified

Key words: UHT milk, process identification, HMF, colour parameters..

INTRODUCTION

One of the objectives of the milk preservation method with the assistance of the high temperature treatment for a short period of time is to obtain the required degree of destruction of microorganisms and enzyme inactivation and, simultaneously, to cause its least possible unfavourable physical-chemical and sensory changes and, first and foremost, to maintain its nutritive value. The appropriately selected temperature and heating time parameters should assist in the achievement of the sterilisation effect which would guarantee health safety as expressed by the corresponding effect of microbiological changes (index B) higher than 1 but lower than 2. They will allow obtaining milk of guaranteed microbiological quality and ensuring absence of clear changes in the nutritive value in which only small quantities of products of the Maillard’s reaction would develop. In the case of hydroxymethylfurfural (HMF), this amount should not exceed 10 µmol·dm^-3 [7]. The effect value of chemical changes (index C) of the milk heated in such conditions will not exceed 1, which corresponds to thiamine losses below 3% and lysine losses lower than 1% [1, 18, 19]. Chemical changes initiated during the process of milk heating can intensify in the course of its subsequent storage [5, 12, 17]. In this group of changes, Maillard’s reactions between lactose carbonyl groups and e-amine lysine residues play the most important role [2, 13, 20]. During the initial stage of the reaction, a condensation of both compounds occurs, which is followed by the development of intermediate products, such as hydroxymethylfurfural, aldehydes, ketones, aldols etc., which, in turn, can undergo further reactions or, in the final stage, undergo the process of condensation or polymerisation. This results in the development of colourful melanoids, which are responsible for the browning of heated milk [11, 14, 15, 16]. Consequently, the intensity of changes associated with the non-enzymatic browning will constitute a decisive factor affecting the degree of colour changes of the heated milk. The Maillard’s reaction initiated during the process of heating may be reinforced and the colour of milk may deteriorate further in the course of the subsequent storage of treated milk depending on temperature conditions.

The objective of the undertaken investigations was to determine interrelationships between colour parameters and the content of hydroxymethylfurfural as the indicator of the course of the Maillard’s reaction in the UHT sterilised milk during its subsequent storage at various temperatures.

MATERIALS AND METHODS

The experimental material was ordinary UHT sterilised milk of 2% fat content manufactured in commercial conditions. The examined milk was manufactured in a process of heating raw milk of the highest hygiene and cytological quality and showing no signs of false chemical composition [6]. From the time of its milking until its processing, the raw milk was stored at refrigerated temperature of 6°C for not longer than 36 h. The objectivisation of the UHT sterilisation temperature-time parameters of the examined milk was carried out on the basis of the Kessler’s diagram [1, 10]. In the result of the applied milk sterilisation parameters, the following indices were obtained: effect of microbiological changes index B = 1.68 and effect of chemical changes index C = 0.56. The calculated values of B and C indices take into consideration not only the time of maintaining milk at the sterilisation temperature but also the time of heating this milk up to this temperature and, then, the time of its coo ling down. The obtained milk was packaged aseptically in airtight packages made from a laminate on a cardboard base of 1 dm³ volume. The experimental milk was then stored in the above-described packages at refrigeration condition at the temperatures of 4±1°C and 8±1°C as well as at the room temperature of 20± 1°C. At each of the above-mentioned temperatures, milk was stored for 24 weeks. Trials were conducted every 4 weeks. The experiment was based on 6 production batches and 4 milk samples were collected from each batch (n = 24).

The colour evaluation of the analysed milk was performed on the basis of readings of results of instrumental measurements. The colour was measured reflectionally using the Hitachi U-3000 spectrophotometer with a 1 cm thickness of the layer and the C source of light. The evaluation of colour in reflected light was performed with the assistance of the CIELAB colour system L*, a* and b* recommended by the Commission Internationale de l`Eclairage, where L* designates brightness, while a* and b* are chromaticity coordinates [4]. The temperature of the assessed sample ranged from 22 to 24°C. On the basis of the measured colour parameters, its saturation C* was calculated employing, for this purpose, formula [8]:

C* = [a² + b²]^0.5

In addition, the value of the colour difference DE was calculated between the colour of each sample and the model of ideal whiteness for which, using formula [8], the following assumptions were made: L*=100, a*=0 and b*=0:

DE=[( DL)² + (Da)² + (Db)²]^0.5

Using the method put forward by Kenney and Basset [9], the content of total and free hydroxymethylfurfural (HMF) was determined in the examined milk. The content of HMF in milk was determined by heating its appropriate quantity at the temperature of 100°C for the period of 60 min following its acidification with an 0.3 N solution of oxalic acid. When determining free HMF, the heating procedure was eliminated in order not to allow the conversion of intermediate products responsible for browning, e.g. 1-amine-1-desoxy-2-ketoze to HMF. The mixture of milk prepared in this way, supplemented with 40% trichloroacetic acid (TCA), was filtrated through the Whatman N^o42 blotting paper. The filtrate was treated with an 0.05 M solution of thiobarbituric acid (TBA) and, next, the mixture was incubated for 35 min at the temperature of 40°C. The absorbance was measured at 443 nm wavelength against a control sample using, for this purpose, a Novaspec II Pharmacia Biotech spectrophotomet er.

total HMF (µmol·dm^-3)= (A-0.055) × 87.5)

free HMF (µmol·dm^-3 )= (A-0.015 × 81.0)

Results of measurements and assays, obtained in the course of performed experiments, were then subjected to statistical analysis using the Excel calculation sheet of the EAV program, ELSQ. The obtained results were employed as the basis for the determination of the standard error (±SE), linear and polynomial regressions of different degree employing the least squares method, the square of correlation coefficient and verification of hypotheses at the set level of significance p = 0.05 [3].

RESULTS AND DISCUSSION

On the basis of the performed investigations, it was found that values of the b* parameter of the examined milk colour altered depending on the applied temperature conditions during its storage (Fig.1). When milk was stored in refrigerated conditions at temperatures ranging from 3 to 5°C, no significant differences were observed between the value of the b* parameter measured at the beginning of storage and the value measured after 24 weeks of storage (R²=0.0008). Also in the 16^th week of milk storage, no significant differences could be found in the b* value parameters between milk stored at the temperature of 4±1°C and 8±1°C. However, after this time, it was observed that the higher was the milk storage temperature, the higher was the value of the b* parameter. The observed change in the milk colour, resulting from the increasing value of this parameter towards yellow colour, intensified together with the passage of storag e time. After 24 weeks of milk storage at the temperature of 20±1°C, the value of the b* parameter was over twice higher in comparison with the value measured directly before its production. A high value of the correlation coefficient square of the parameter b* value increment was calculated for the milk stored at the range of temperatures from 19 to 21°C, depending on the time of its storage (R² = 0.9242).

Fig. 1. Changes in the colour parameter b* (±SE) of the UHT sterilised milk stored at different temperatures A-D; a-c – mean values followed by different capital letters for the same storage times and small letters for the same storage temperatures are significantly different at p=0.05 temp. 4°C temp. 8°C temp. 20°C

The period of 24 weeks of milk storage at different temperatures exerted a significant influence on the intensity value of the C* milk colour (Fig. 2). The above values remained unchanged only during the storage at the lowest of the applied temperatures, i.e. 4±1°C (R² = 0.0004). The storage of milk at higher, albeit still refrigerated, storage temperature, i.e. 8±1°C, caused that from the 16^th week onward, a statistically significant increase in the value of this index was observed. In the 20^th week of storage, the value of the milk colour saturation C* stored at the temperature of 7-9°C was 8.83, whereas directly after processing - this value was 6.08. After 24 weeks of storage, milk stored at the temperature of 20±1°C was characterised by a more saturated colour than that stored at the temperature of 8±1°C. From among the analysed ranges of storage temperatures, the milk colour saturation value stored at the temperature of 20±1°C increased proportionally in relation to the lengthening of the storage time reaching, after 24 weeks, the value of 13.65 (R² = 0.9300).

Fig. 2. Changes in the colour saturation C* (±SE) of the UHT sterilised milk stored at different temperatures A-D; a-c – mean values followed by different capital letters for the same storage times and small letters for the same storage temperatures are significantly different at p=0.05 temp. 4°C temp. 8°C temp. 20°C

Colour changes taking place in milk during its storage at different temperature conditions were also assessed on the basis of calculated milk sample DE values in relation to the model of the ideal whiteness (Fig. 3). As the statistical analysis revealed, storage time did not have a significant influence on changes of this parameter, when the milk was stored at the temperature of 4±1°C. At higher storage temperatures, i.e. 8±1°C, these changes were significant beginning from the 4^th week of storage. After 24 weeks of storage, the difference between the colour of the examined milk and the whiteness model increased significantly (by 17%) in relation to the value directly after the termination of processing. Towards the end of the storage period, milk stored at room temperature was characterised by the highest value of this parameter (DE = 20.54). This milk was the least white from among all the analysed samples. It should be stressed that any increase in the value of the milk parameter DE, in any of the milk storage intervals, had a linear and uniform nature, as evidenced by high values of the square of the correlation coefficient (R² = 0.8906 in 4±1°C, R² = 0.9115 in 8±1°C, R² = 0.8825 in 20±1°C).

Fig. 3. Changes in the colour diferent DE (±SE) of the UHT sterilised milk stored at different temperatures A-D; a-c – mean values followed by different capital letters for the same storage times and small letters for the same storage temperatures are significantly different at p=0.05 temp. 4°C temp. 8°C temp. 20°C

The analysis of results obtained in the course of these studies confirmed a significant impact of the storage time and temperature on changes in the concentration of the total hydroxymethylfurfural (HMF) in the examined milk (Fig. 4). As the time of milk storage increased, the content of the total HMF also increased significantly. It is worth emphasising that no significant differences were observed in the concentration of the total HMF between the milk stored at the temperature of 4±1°C and that stored at 8±1°C. After 24 weeks of storage of the examined milk at the range of refrigerated temperatures, 1.7 times more HMF was determined than directly after production. This increase was more than twofold at room temperature. After each month of storage, approximately 20% more total HMF was determined in the milk stored at room temperature than in that stored at refrigerated conditions. The polynomial regression equations, determined following a fluctua ting distribution of measuring data, indicate a high degree of dependence of the total HMF content on the time of milk storage (0.9217<R²>0.9626) in all of the examined ranges of storage temperatures.

Fig. 4. Changes in the total HMF (±SE) of the UHT sterilised milk stored at different temperatures temp. 4°C temp. 8°C temp. 20°C

The development of the Maillard’s reaction products, depending on the conditions of milk storage, was also confirmed analysing the content of the free HMF (Fig. 5). At the end of the production process, the concentration of the free HMF in milk was found to be at the level of 0.625 µmol·dm^-3, whereas after 24 weeks of storage, this value increased 2 times. It should be noted here that during milk storage at room temperature, i.e. 20±1°C, the free HMF increments were the highest of all the values determined in this experiment. At the end of the storage period, 1.314 µmol·dm^-3 of the free HMF was determined in this milk. The highest increase of the free HMF in milk occurred in the 16^th week of its storage at temperatures of 8±1°C and 20±1°C. The concentration of the free HMF increased in the 16^th week of storage 1.87 times at the temperature of 8±1°C and 2.06 t imes - at the temperature of 20±1°C in comparison with the value determined immediately after the end of the milk production process. The free HMF increments in the milk stored at the lowest of the temperatures applied in this trial were best correlated with the storage time (R² = 0.9955).

Fig. 5. Changes in the free HMF (±SE) of the UHT sterilised milk stored at different temperatures temp. 4°C temp. 8°C temp. 20°C

The linear function values of the b* parameter, colour saturation C* and colour difference DE calculated from the total HMF determined in milk as the controlled variable indicate a varying degree of relationships between the analysed factors (Table 1). Out of all the analysed factors, the evaluation of the colour difference value DE appears to be the best indicator of the accumulation of the Maillard’s reaction products in milk as evidenced by the values of the equation determining the trend line of this factor with the assistance of the method of the least squares as well as the values of the determination coefficient. The high value of the correlation coefficient confirms a strong interrelationship between the value of colour difference DE and the concentration of the total HMF in the same milk (0.86<r>0.91). The course of the reaction of the non-enzymatic browning dete rmined on the basis of changes in the total HMF concentration is least represented by changes in the b* parameter of milk.

When analysing values of linear equations and values of squares of correlation coefficients between the determinant of milk colour and the content of the free HMF in it, a varying degree of their mutual interdependence was found (Table 2). It was determined that, from among the analysed factors, the value of milk colour difference DE may be used as the indicator of the intensity of the Maillard’s reaction. The dependence of values of the correlation coefficient of the colour difference DE function on the content of the free HMF was the highest of all those determined in this experiment (0.84<r>0.94). The lower was the temperature of milk storage, the lower was the value of the slope of the line described on the basis of functions of analysed factors and the higher was the value of the point of their intersection.

Summing up the obtained results it can be concluded that:

• storing milk at temperatures ranging from 3 to 5^oC, no significant differences were observed between the values of the b* parameter measured at the beginning of the storage and the value determined after 24 weeks of storage. Also the value of the colour saturation C* of this milk remained unchanged,

• the whiteness of the milk stored at room temperature was further removed from the model than that of the milk stored at refrigerated conditions. The observed difference increased in intensity together with the duration of storage,

• no significant differences were found in the content of the total and free HMF stored at 4±1°C as well as at the temperature of 8±1°C. After 24 weeks of milk storage at room temperature, over 2 time more HMF was determined than directly after its production,

• a strong correlation was found between the value of the colour difference DE and the content of the total and free HMF of the same milk.

1. Biewendt H. G., 1994. Die Bedeutung von Sterilisationswerten für das Ultrahocherhitzen von Milch [The signification of sterylization values for UHT-heating of milk]. Nahrung 38, 233-252 [in German].

2. Boekel M. A. J. S., 1998. Effect of heating on Maillard reactions in milk. Food Chem. 62, 403-414.

3. Brandt S., 1997. Statistical and computational methods in data analysis. Springer Verlag, New York.

4. CIE Colorimetry Committee, 1974. Technical notes: working program on colour differences. J. Opt. Soc. Am. 64, 896-897.

5. Corzo N., Delgado T., Troyano E., Olano A., 1994. Ratio of lactulose to furosine as indicator of quality of commercial milks. J. Food Prot. 57, 737-739.

6. Council Directive UE, 1992. Directive laying down the health rules for the production and placing on the market of raw milk, heat-treated milk and milk-based products. 92/46/EEC Document 392L0046, Off. J. L 268, 1.

7. Fink R., Kessler H.G., 1986. HMF values in heat treated and stored milk. Milchwissenschaft 41, 638-641.

8. Francis F. J., Clydesdale F. M., 1975. Food colorimetry: theory and applications. The AVI Publishing Company, Westport Connecticut.

9. Keeney M., Bassette R., 1959. Detection of intermediate compounds in the early stages of browning reaction in milk products. J. Dairy Sci. 42, 945-959.

10. Kessler H. G., Horak P., 1981. Objektive Beurteilung der UHT-Milcherhitzung durch Normierung bakteriologischer und chemischer Effekte [Objective evaluation of UHT-milk-heating by standarization of bacteriological and chemical effects]. Milchwissenschaft 36, 129-133 [in German].

11. Macdougal D. B., Garnov M., 1998. Relationship between ultraviolet and visible in Maillard reactions and CIELAB colour space and visual appearance. In.: The Maillard reaction in foods and medicine. Ed: J. M. Amnes, Cambridge, UK.

12. Marconi E., Messia M. C., Amine A., Moscone D., Vernazza F., Stocchi F., Palleschi G., 2004. Heat-treated milk differentation by a sensitive lactulose assay. Food Chem. 84, 447-450.

13. Morales F. J., Boekel M. A. J. S., 1988. A study on advanced Maillard reaction in heated casein/sugar solutions: colour formation. Int. Dairy J. 8, 907-915.

14. Morales F. J., Jimez-Perez S., 2001. Free radical scavening capacity of Maillard reaction products as related to colour and fluorescence. Food Chem. 72, 119-125.

15. Morales F. J., Romero C., Jimez-Perez S., 1995. New methodologies for kinetic study of 5-hydroxymethylfurfural formation and reactive lysine blockage in heat-treated milk and model systems. J. Food Prot. 58, 310-315.

16. Morales F. J., Romero C., Jimez-Perez S., 1996. Fluorescence associated with Maillard reaction in milk and milk resembling systems. Food Chem. 57, 423-428.

17. Morales F. J., Romero C., Jimez-Perez S., 1993. Evaluation of heat-induced changes in Spanish commercial milk: hydroxymethylofurfural and available lysine content. J. Food Sci. Tech. 31, 411-418.

18. Morales F. J., Romero C., Jimez-Perez S., 2000. Characterization of industrial processed milk by analysis of heat-induced changes. J. Food Sci. Tech. 35, 193-200.

19. Pellegrino L., De Noni I., Resmini P., 1995. Coupling of lactulose and furosine indices for quality evaluation of sterilized milk. Int. Dairy J. 5, 647-659.

20. Yaylayan V.A., 1997. Classification of the Maillard reaction a conceptual approach. Trends in Food Sci. Tech. 8, 13-18.

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed ‘Discussions’ in each series and hyperlinked to the article.