Furosine determination

Guerra-Hernandez, E., Corzo, N. and Garcia-Villanova, B. (1999). Maillard reaction evaluation by furosine determination during infant cereal processing, J. Cereal ScL, 29, 171-176. 

It is possible to assess lysine availability using an Escherichia coli lysine auxotroph,44 but further development is required before such a test can compete with furosine determination. 

In heat treated or stored food products several amino acids are not fully available because of derivatization or crosslinking reactions. Since 30 years furosine is known as a useful indicator of early Maillard reaction which is applied in food science, nutrition and medical biochemistry. Recently more sensitive analytical methods for furosine determination are available which have again increased the attractivity of this important indicator. Lately, N -carboxymethyllysine (CML) became available as another marker of special interest, because CML is a more useful indicator of the advanced heat damage by Maillard reaction than furosine. In addition, CML has the advantage to indicate reactions of lysine with ascorbic acid or ketoses such as fructose. Indicators for protein oxidation of sulfur amino acids are methionine sulfoxide and cysteic acid. An established marker for cross-linking reactions is lysinoalanine, which also indicates protein damages due to processing under alkaline conditions. Other markers formed as a consequence of alkaline treatment are D-amino acids. 

The extent of heat-induced changes in protein-rich foods can be measured by determining some early Maillard reaction products (O Brien and Morrissey, 1989). Acid treatment of protein-bound or free A -fructoselysine liberates lysine, with a yield of 50%, and two other amino acids, furosine (20%) and pyridosine (10%) (Figure 13.1). The three products that result from hydrolysis of A -lactuloselysine are formed in the proportions of approximately 5 3 to 4 1 to 2, however the yield of different derivatives is variable. Therefore, in order to use these unique amino acids as indicators of changes in lysine content, the hydrolysis should be carried out in strictly-defined conditions. Furosine is present in various food products in a very wide range of concentrations (Table 13.2). 

Furosine and fluorescent cross-links (mol amino acid/ mol collagen, except pentosidine mmol/mol) determined by HPLC in hydrolyzates of dentin slices exposed to glucose and buffer, pH 7.4, and non-exposed controls (n=2). 

We determined the content of furosine, which is an amino acid derivative formed by acid hydrolysis of fructose-lysine or lactulose-lysine ( (Fig. 1). 

Crude protein and total lysine plus furosine Total nitrogen was determined by macro Kjeldahl digestion and protein estimated as N x 6.25. Total lysine values were obtained from conventional amino acid analyses carried out on 500-mg samples following digestion with 800 ml of 6M HC1 under reflux by use of a Biotronic LC 6000 or Kontron Liquimat III amino acid analyzer. Furosine was determined in the same way using 300 ml 7.8 M HC1 as described in ( 6 ) with an amino acid analyzer ( 7 ). 

The concentration of furosine in some enteral formulae has been determined, ranging from 245 to 441 mg per 100 g protein.40 It seems to decrease on further processing and does not appear to be a useful indicator of quality. Furosine content seems to be stable in enteral formulae on storage for >36 weeks at 4 °C, 24 weeks at 20 °C, and >12 weeks at 30 °C.41 Stability at 55 °C was down to 3-8 weeks, depending on protein content, whereafter furosine content declined. Reactivity of lysine residues with ophthaldialdehyde declined roughly in parallel with furosine formation. 

Thiazolium salt XTT has been used to evaluate the extent of the Maillard reaction in UHT (ultra-high-temperature)-treated milk, where it proved more rapid and convenient than determination of lactulose, HMF, or furosine.451 To establish the nature of the Maillard reaction products (MRP) involved, the interaction of the aminoreductone from lactose and n-butylamine (Amax = 319.5 nm) with MRP was studied. Excellent correlation (r = 0.967, n = 19) was found between the increased absorbance at 319.5 nm and XTT reducibility, as measured at 492 nm, near Alliax for the formazan, the reduction product of XTT. 

Furosine, a marker of the Maillard reaction product, is a valuable indicator of food protein quality. It is a marker for thermal treatment in foodstuffs and is directly related to the loss of lysine availability. IPC was employed to determine furosine content in beverages based on soy milk and cow milk supplemented with soy isoflavones [39]. Furosine was also analyzed in 60 commercial breakfast cereals to assess their protein nutritional values. The higher the protein content in the formulation, the higher the furosine levels [40]. A simple IPC technique that uses 1-octanesulfonic acid as the IPR allowed the selective determination of histamine levels in fermented food [41]. 

Vallejo-Cordoba, B., Mazorra-Manzano, M. A., and Gonzalez-Cordova, A. R, New capillary electrophoresis method for the determination of furosine in dairy products, J. Agric. Food Chem., 52, 5787, 2004. 

The side chains of proteins can undergo post-translational modification in the course of thermal processes. The reaction can also result in the formation of protein cross-links. A known reaction which mainly proceeds in the absence of carbohydrates, for example, is the formation of dehydroalanine from serine, cysteine or serine phosphate by the elimination of H2O, H2S or phosphate. The dehydroalanine can then lead to protein cross-links with the nucleophilic side chains of lysine or cysteine (cf. 1.4.4.11). In the presence of carbohydrates or their degradation products, especially the side chains of lysine and arginine are subject to modification, which is accompanied by a reduction in the nutritional value of the proteins. The structures of important lysine modifications are summarized in Formula 4.95. The best known compounds are the Amadori product -fructoselysine and furosine, which can be formed from the former compound via the intermediate 4-deoxyosone (Formula 4.96). To detect of the extent of heat treatment, e. g., in the case of heat treated milk products, furosine is released by acid hydrolysis of the proteins and quantitatively determined by amino acid analysis. In this process, all the intermediates which lead to furosine are degraded and an unknown portion of already existing furosine is destroyed. Therefore, the hydrolysis must occur under standardized conditions or preferably by using enzymes. Examples showing the concentrations of furosine in food are presented in Table 4.13. 

Henle, T., Zehetner, G., Klostermeyer, H. Fast and sensitive determination of furosine. Z. Lebensm. Unters. Forsch. 200, 235 (1995)... 

In dairy products also hydroxymethylfurfural (HMF) and lactulose are common markers. For the determination of HMF, which also results from the Maillard condensation (18), precursors of browning products in milk are transformed to HMF after addition of oxalic acid and following heating (79). Principally, the HMF value of a milk can be used as an indicator for the heating process, but data from literature offer a wide range for this value which suspect that the HMF determination is insufficiently reproducible between laboratories. In particular, the level of HMF in untreated material, measured during the determination of "total HMF" and subtracted from the levels in treated milks, is a source of variation (20). However a comparison between the furosine and the HMF-method demonstrated the usefulness of the HMF-method as a rapid and simple measure of heat damage caused by the UHT process (77). 

Additionally to the above mentioned and widely used markers, lysinoalanine (LAL) was determined in model experiments and in several commercial products. LAL is formed throughout heat and/or alkali treatment of proteins by nucleophilic reaction of the lysyl-e-amino-group with the activated double bond of dehydroalanine, which is formed by B-elimination of cystine and phosphoserine in the peptide chain. Unlike furosine, LAL crosslinking creates not only a decrease in lysine but mainly of cyst(e)ine availability in the case of alkaline treatment. 

Several methods have been developed for assaying non-enzymatic glycosylation. As far as biological systems are concerned, these have been extensively reviewed by A. J. Furth in 1988 (5). They include both assays on intact proteins after chemical degradation and selective detection of e.g. 5-hydroxymethylfurfural (HMF) and formaldehyde using the thiobarbituric assay (TEA), and assays on protein hydrolysates with or without previous reduction of the protein-bound Amadori compound. In this last case, the analysis is based on the determination of furosine which is specifically formed from lysine Amadori compounds with a yield of approximately 30% (6). The furosine method, originally developed for milk (7), has been the subject of several analytical improvements both for food products (8) and biological materials (9). More recently, another method has been proposed to evaluate the extent of early Maillard reaction in milk products. This method is based on direct measurement of the Amadori product lactuloselysine which is released after complete enzymatic hydrolysis (10). 

The determination oj furosine (LySfu )- This method is based on the fact that lysine blocked in an early Maillard reaction in the form of e-N-deoxyketosyl-lysine yields upon acid hydrolysis a constant proportion of furosine (Fig. 2) (e-N-2-furoylmethyl-L-lysine). The blocked lysine corresponds, then, to 3-1 times furosine. The available lysine value can then be calculated by subtracting the blocked lysine value from the theoretical lysine value (ThL = lysine present in the unheated milk sample). 

The furosine method (FUR) (Fig. 7). In milk samples, this method allows the direct determination of blocked lysine (DL-Lys) as grammes Lys/16 g N or, still more conveniently, as a percentage of the given theoretical lysine value when the nitrogen content of the sample... 

Obviously, available lysine can then be calculated from blocked lysine, taking the theoretical lysine value as 100. In this case, (Fig. 7), the correlation equation with ALV n has no special meaning, since this regression was used to validate the furosine method. In our experience, the furosine method has been very useful for routine control of lysine blockage in industrial milk samples. It should be borne in mind, however, that the method as described can only be used to determine early Maillard damage to lysine (deoxyketosyl form). 

---