Analysis of mono- and disaccharides in milk-based formulae by high-performance liquid chromatography with refractive index detection

https://doi.org/10.1016/j.chroma.2004.06.002Get rights and content

Abstract

A simple and reproducible method for the qualitative and quantitative analysis of free mono- and disaccharides (fructose, glucose, galactose, sucrose, lactulose and lactose) in milk-based formulae by high-performance liquid chromatography (HPLC) with refractive index (RI) detection was developed and validated. The method showed good linearity with determination coefficients exceeding 0.99. The limits of detection (DL) in these sugars were 0.17, 0.13, 0.06, 0.16, 0.05 and 0.25 mg/ml, respectively; and the limits of quantification (QL), 0.27, 0.24, 0.20, 0.26, 0.22 and 0.38 mg/ml. The relative standard deviations (R.S.D.s) for repeatability in fructose, sucrose, lactulose and lactose were 0.78, 0.99, 2.91 and 0.46 and the R.S.D.s for reproducibility were 4.8, 6.15, 7.04 and 2.49, respectively. Recoveries in all sugars were between 93 and 113%.

Introduction

The industry has deployed considerable technological resources to bringing the composition of infant formulae closer to that of human milk [1]. Besides, it has developed several milk-based formulae for adults, e.g. for pregnant women [2], [3], [4], [5], [6]. Milk-based formulae can be based on any appropriate blend of proteins, carbohydrates, fats, minerals and vitamins. Milk powders are usually free-flowing agglomerates formed by spray drying, which extend the shelf-life of dried milk from several days to 18–24 months [7], [8].

One of the industry’s main problems is to control the stability of milk-based formulae, because they contain a lot of components that may interact. Milk powders are especially sensitive to Maillard reaction, as they contain a relatively high concentration of lactose and proteins with a high lysine level, besides the high temperature applied during the manufacturing process and their storage for long periods of time [9]. During heat treatment, lactose undergoes the Lobry de Bruyn-Alberda van Eckenstein rearrangement, which gives rise initially to isomeric disaccharides, mainly lactulose. As lactulose is not known to occur naturally in milk and is only formed in heated dairy products [10], [11], [12], [13], [14], it is a good indicator of heat damage in milk products. Changes may occur during the formulae powders’ long periods of storage, even under appropriate storage conditions, and may even be greater than those caused by heat treatment in the production process. The result is, an unacceptable product. Many milk-based formulae contain sugars besides lactose, the evolution of mono- and disaccharides needs to be evaluated. Observation of product stability will help determine whether there are any differences between the same formulae during storage time and the shelf-life.

Many techniques have been developed in order to evaluate sugar fraction during the Maillard reaction. One of the methods is spectrophotometric [15], [16], also in some studies lactulose is identified by enzymatic method [17]. The major disadvantage of these consists in the difficulty to evaluate simultaneously different sugars.

Another method developed to evaluate damage in milk powders is the capillary electrophoresis [9], which consist in monitoring the β-lactoglobuline of the whey protein fraction. In spite of the promising use, for preparation of sample the caseins need to be precipitated with HCl overnight at 4 °C this mean a lot of time in analysis sample. Yet another method is the high-pH anion-exchange separation with pulsed amperometric detection (AEC–PAD) for evaluating monosaccharides as glucose, fructose and disaccharides as lactulose, lactose, sucrose and maltose. Kaine and Wolnik [18] studied sugars in infant formulae by high pH AEC–PAD, Cataldi et al. [19] gave a comprehensive overview of analytical applications in food for carbohydrate analysis by high-pH AEC-PAD.

One of the methods commonly used in sugar analysis is the gas chromatography (GC). Troyano et al. [20], [21] developed a GC method. With this is possible quantifying glucose, galactose, myo-inositol, lactulose, N-acetylglucosamine, N-acetylgalactosamine and other derivatives. GC has been used in the study of milk [20], [22], dried skim milk [23], in model systems containing protein-bonded lactose [24] and in pasteurized milk [25]. Valero et al. [13] determined the intensity of the heat treatment in milk pasteurized for the amount of lactulose formed by GC of the trimethylsilyl derivatives of the free sugar, besides monosaccharides were determined. Also in UHT milk [26] and in milk permeate GC has been used [27]. In spite of GC is a sensitive method for sugar analysis, sample preparation is laborious. Besides in the CG procedure the anomeric composition of α- and β-anomers is obtained which mean more than one area peak for each compound. The procedure is tedious to be used routinely.

Finally, in many studies, HPLC is used for its accuracy, separation abilities and rapidity [28], [29]. It appeared more then 20 years ago, but remains one of the most widely used techniques. HPLC with refractive index (RI) detection is a powerful technique for quantifying various types of carbohydrate compounds. HPLC–RI was used for determining sugars (glucose, fructose and sucrose) in apple juice [30], disaccharides in whey permeate (lactose, galactose and lactulose) [31], oligosaccharides (fructose, glucose, sucrose, maltose and lactose) in plain cereals, sugar coated cereals, canned fruits, canned vegetables, crackers cookies [32]. HPLC–RI has also been used for determining sugars (sucrose, glucose and fructose), in fruit and drink samples [33], sugars in meat products [34], oligosaccharides in lactose–sucrose systems for determining sucrose inversion by invertase [35] and in sugar casein systems [36]. Martins et al. [37] studied the kinetic modelling of amadori N-(1-deoxy-d-fructos-1-yl)-glycine (DPG; intermediate in the early stages of the Maillard reaction) pathways in aqueous model systems, the quantification of d-glucose and d-manose was made by HPLC using an ion-exchange column (ION-300), and sugars were detected by monitoring the refractive index.

Although difficulties of using eluent gradients and relatively poor sensitivity associated with refractometry, HPLC–RI appears to be an economical, simple and fast method for determination of sugars. The aims of this study were to design and to validate an easy HPLC–RI method that separates the free sugar fraction from components such as proteins and other macromolecules that could create interference in the system; and to analyze qualitative and quantitative free mono- and disaccharides in milk-based formulae.

Section snippets

Reagents and standards

The chemicals used for sample preparation were of analytical reagent grade: HPLC-grade, SDS acetonitrile and methanol (Peypin, France), HPLC-grade, Panreac absolute ethanol, Carrez I and Carrez II reagents (Barcelona, Spain), deionised water purified through a Milli-Q system (Millipore, Bedford, MA, USA). The standard sugars (fructose, glucose, galactose, sucrose, lactulose and lactose) came from Sigma (St. Louis, MO, USA), were >99% pure and were stored in a vacuum desiccator, with silica gel

Results and discussion

HPLC–RI detection was used to determine fructose, glucose, galactose, sucrose, lactulose and lactose. Folks and Jordan [38] suggested as an appropriate mobile phase acetonitrile–water in the range 75:25 to 85:15. We experimented with 75:25, 80:20, 85:15 and 90:10 and found that with a 75:25 (v/v) the sugars eluted rapidly and the mobile phase provided better peak symmetry and acceptable separation peaks, except from glucose and galactose which were overlapped. Although this, glucose and

Conclusion

The results of sugar analysis in the experimental formula for pregnant women sample are given in Table 3, containing 13% of fructose, 9% of sucrose, 0.9% of lactulose and 16% of lactose. The infant formula sample analyzed only contain lactose (57.21 ± 0.2%). The establishment of thermal parameters, defined under specific temperature/time conditions, contributes to the classification of heat treated milks. These thermal parameters are mainly employed to identify and optimize processes, assess

Acknowledgements

The authors are grateful to Laboratorios Ordesa S.L. (Sant Boi de Llobregat, Barcelona, Spain) for providing the samples, and Robin Rycroft for correcting the English. Special thanks are due to CONACYT (Mexico) for their grant to J.L. C.-S.

References (42)

  • I.M.P.L.V.O. Ferreira et al.

    Carbohydr. Polym.

    (1998)
  • P. Guesry

    Prev. Med.

    (1998)
  • M. Neuringer

    Am. J. Clin. Nutr. Suppl.

    (2000)
  • N. Özkan et al.

    J. Food Eng.

    (2003)
  • T.K. Kockel et al.

    J. Food Eng.

    (2002)
  • J. De Block et al.

    Int. Dairy J.

    (2003)
  • E. Valero et al.

    Food Chem.

    (2000)
  • W. Baltes

    Food Chem.

    (1982)
  • L.A. Kaine et al.

    J. Chromatogr. A

    (1998)
  • A. Olano et al.

    Food Chem.

    (1989)
  • A. Olano et al.

    Food Chem.

    (1992)
  • J. Belloque et al.

    Food Chem.

    (2001)
  • M. Villamiel et al.

    Food Chem.

    (2002)
  • J.P. Yuan et al.

    Food Chem.

    (1999)
  • S. Vendrell-Pascuas et al.

    J. Chromatogr. A

    (2000)
  • S.I.F.S. Martins et al.

    Carbohydr. Res.

    (2003)
  • C. Agostini et al.

    Acta Paediatr. Suppl.

    (1999)
  • M. Xiang et al.

    Acta Paediatr.

    (1999)
  • B.G. Jeffrey et al.

    Lipids

    (2001)
  • G.R. Andrews

    J. Dairy Res.

    (1986)
  • R. López-Fandiño et al.

    Food Sci. Technol. Int.

    (1999)
  • Cited by (150)

    • Hyphenating temperature gradient elution with refractive index detection through temperature-responsive liquid chromatography

      2022, Analytica Chimica Acta
      Citation Excerpt :

      Such setup offers a non-destructive, concentration-dependent bulk property detector [8,9]. RID is broadly used for quantitative analysis of solutes occurring in the high mg mL−1 range whereby either chromophores or authentic standards are lacking, and isocratic analysis is possible [7,10–17]. The refractive index is, however, next to the influence of eluting solutes, also strongly affected by the mobile phase composition, density, and temperature.

    • Comparison of ultraviolet and refractive index detections in the HPLC analysis of sugars

      2021, Food Chemistry
      Citation Excerpt :

      However, an unlabeled analysis with UV-based detection is adequate for sugar analysis in the micromole range. As indicated by previous studies, RI detection is the gold standard for detecting sugars (Chávez-Servín, Castellote, & López-Sabater, 2004). However, the current study showed that the UV detection of monosaccharides using both isocratic and gradient methods provided comparable results to RI detection in terms of the LOD and LOQ.

    View all citing articles on Scopus

    Presented at the Third Meeting of the Spanish Association of Chromatography and Related Techniques and the European Workshop: 3rd Waste Water Cluster, Aguadulce (Almeria), 19–21 November 2003.

    View full text