- Email: [email protected]
Maillard Reaction Evaluation by Furosine Determination During Infant Cereal Processing
Journal of Cereal Science 29 (1999) 171–176 Article No. jcrs.1998.0226, available online at http://www.idealibrary.com on
Maillard Reaction Evaluation by Furosine Determination During Infant Cereal Processing E. Guerra-Hernandez∗, N. Corzo† and B. Garcia-Villanova∗ ∗Universidad de Granada, Departamento de Nutricio´n y Bromatologı´a, 18012 Granada, Spain; †Instituto de Fermentaciones Industriales (C.S.I.C.), Juan de la Cierva 3, 28006 Madrid, Spain Received 15 September 1997
ABSTRACT The furosine content of infant cereals has been used to evaluate the thermal damage caused to those foods during their manufacture (toasting, amylolysis and drying steps). The mean furosine contents of raw and toasted flours were 13·7 and 13·8 mg/100 g of protein, respectively. The furosine content increased to 47·3 mg/100 g of protein after hydrolysis using alpha-amylase. The amylolysis process raised the reducing sugar content from 0·5% for raw samples to 12·7%. The furosine content of infant cereals rose to average levels of 500 mg/100 g of protein after roller-drying. The infant cereals with the highest levels of reducing sugars, with or without soy flour, exhibited the highest furosine contents. Furosine is a useful indicator to evaluate the progress of the Maillard reaction during the amylolysis and roller-drying stages of infant cereals manufacture. 1999 Academic Press
Keywords: Maillard reaction, furosine, infant cereal processing.
INTRODUCTION The manufacture of baby cereals represents a relatively simple food system where non-enzymic browning involving the Maillard reaction (MR) can occur easily. This is due to the type and concentration of the main reagents involved, the water activity (aw) of the system during drying, and the time-temperature conditions of the thermal process1. Infant cereals must be processed to improve their dispersibility in liquids and their digestibility since 3–4-month old babies have a reduced capacity to digest starch. Baby cereal processing involves sequential toasting, hydrolysis with alpha-amylase (amylolysis) and drying stages2. During the amylolysis process, the : HPLC=high performance liquid chromatography; PTFE=polytetrafluoroethylene.
∗ Corresponding author: E. Guerra-Hernandez. Tel: (958) 243867; Fax: (958) 243869; E-mail: [email protected] 0733–5210/99/020171+06 $30.00/0
reducing sugar level increases, thus facilitating the Maillard reaction during the toasting and drying processes. These treatments improve the sensory characteristics and starch digestibility, but adverse effects can occur. The loss in nutritional quality and potentially in safety is attributed to destruction of essential amino acids, decrease in digestibility, inhibition of proteolytic and glycolytic enzymes, interaction with metal ions, and formation of antinutritional and toxic compounds3,4. The evaluation of the early stages of Maillard reaction, can be achieved by the determination of furosine (e-N-(furoylmethyl L-lysine), an amino acid formed during acid hydrolysis of the Amadori compound fructosyl-lysine, lactulosyl-lysine and maltulosyl-lysine produced by reaction of e-amino groups of lysine with glucose, lactose and maltose5, respectively. This amino acid has been used to measure the early stages of Maillard reaction, not only in foods6,7, but also in biological samples8. In cereal products, furosine determination has been used to monitor the extent of non-enzymic browning in cakes9 and to monitor pasta processing10. Few studies have been carried out regarding 1999 Academic Press
E. Guerra-Hernandez et al.
172
Table I
Composition of different samples used for furosine determination
Samples
Type of cereal
Single flours
Wheat Oat Rice
Mixture of flours (A)
Ingredients
Seven cereals: wheat, rice, barley, rye, oat, corn and millet
Soy
Eight cereals: wheat, rice, barley, rye, oat, corn, millet and sorghum
Soy and honey
Rice, corn
Mixture of flours (B)
Rice, corn
Soy
Seven cereals: wheat, rice, barley, rye, oat, corn and millet
Soy, biscuit, orange and banana powders
Seven cereals: wheat, rice, barley, rye, oat, corn and millet
Soy, orange and banana powders
Rice, corn
Soy, orange and banana powders
the Maillard reaction in infant cereal products, although Carratu et al.11 and Guerra-Herna´ndez and Corzo12 determined furosine in two and 10 commercial baby cereal products, respectively. Since infant cereals are products for infant feeding, it is of interest to know the extent of Maillard reaction during their processing. The objective of this work was to study the early stages of Maillard reaction by furosine determination at the different steps of infant cereals processing as well as the usefulness of furosine as a parameter to control the process.
dried and then stored at −50 °C until they were analysed. Reagents preparation All chemicals were of analytical grade. A furosine standard was prepared by acid hydrolysis of eN-(1-deoxy-D-fructosyl)-L-lysine according to the procedure of Finot et al.13 and its authenticity confirmed by UV and NMR spectroscopy15. A standard stock solution, containing 1·2 mg/ 100 mL of furosine, was used to prepare the working standard solution.
EXPERIMENTAL Samples
Methods
Infant cereal samples were obtained from a Spanish dietetic products company2. The compositions of samples used in this study are shown in Table I. Single and mixed cereal flours (A) were analysed raw (before being processed), after toasting at 150 °C, after hydrolysis of the toasted material with alpha-amylase at 45–50 °C, and after rollerdrying of the amylase-treated material. Mixed cereal flours (B) with fruits were analysed only after roller-drying. Another study was carried out with wheat, rice and oat flour samples at two toasting temperatures (140 °C and 150 °C). Solid samples were stored at −50 °C until they were analysed. Hydrolysed samples were freeze-
Furosine determination Furosine was determined after hydrolysing the sample (0·15 g) with 10·6 HCl (4·5 mL) at 110 °C for 24 h in a Pyrex screw-cap vial with PTFEfaced septa. High-purity helium gas was bubbled through the solution for 2 min12. Sample preparation for HPLC analysis was carried out following the method of Resmini et al.14. The hydrolysate was filtered with a medium-grade paper filter. A portion (0·5 mL) of the filtrate was applied to a Sep-pak C18 cartridge (Millipore), prewetted with methanol (5 mL) and water (10 mL), then eluted with 3 HCl (3 mL) and evaporated under vacuum. The dried sample was dissolved in a mixture (3 mL) of water, acetonitrile and formic
Furosine formation infant cereal processing
acid (95:5:0·2) before HPLC analysis. Samples were analysed in duplicate. Furosine was quantified by ion-pair reversedphase HPLC (Beckman System Gold chromatograph) according to the method of Delgado et al.15, using a Spherisorb ODS2 5-l column (250×4·6 mm, i.d.; Phenomenex) operating at room temperature. The mobile phase consisted of a solution of 5m sodium heptane sulphonate with 20% (v/v) acetonitrile and 0·2% (v/v) formic acid. The elution was isocratic and the flow rate was 1·2 mL/min. The UV detector (Beckman) was set a 280 nm and the injection volume was 50 lL. Calibration of the chromatographic system for furosine determination was made by the external standard method.
Protein and reducing sugars determination Protein determination was carried out by the Kjeldahl method16. Reducing sugars were determined by a titrimetric method17. RESULTS AND DISCUSSION The precision of the entire assay procedure including acid hydrolysis, sample preparation, and RP-HPLC analysis was evaluated for toasted wheat samples (n=7). The relative standard deviation (RSD) was 6·21% obtained on a sample with an average furosine value of 14·3 mg/100 g of protein. The relative standard deviation (RSD) for samples with high levels of furosine was 3·73%12. Single and mixture of cereal (A) samples processing Figure 1 shows the HPLC chromatograms of furosine at different stages of infant oat flour processing. The greatest increase in furosine content was observed during the final roller-drying step. The furosine contents after the different processing steps of the samples studied are shown in Table II.
Raw flours The furosine contents of raw cereal flours ranged from only trace levels (rice flour) through 10·6 mg/ 100 g protein (wheat flour) to 23 mg/100 g of protein (rice-corn flour) (Table II). Furosine in raw flour samples may originate during grain growth, milling or storage. Similar levels of furosine were
173
found in durum wheat semolina in previous studies10,18, but lower values were found in wheat grain. These differences may be due to the milling and storage conditions used for the flour.
Toasted flours Toasting produced changes in furosine content in the cereal flours studied. Toasted wheat, rice and oat flours had increased furosine contents compared with the raw flours, especially rice (7·3 mg/ 100 g of protein). The furosine values for rice–corn and rice–corn–soy flours decreased after toasting, however (Table II). Since the usual temperatures used in commercial toasting processes are close to 140–150 °C, a study of the influence of temperature on Maillard reaction was carried out with other single infant cereal flour samples (Table III). Toasting at 150 °C gave a lower furosine content for wheat flour than toasting at 140 °C, whereas, for rice and oat flours, the higher temperature gave rise to greater levels of furosine. This observation may be explained by the degradation of the Amadori compound under the more severe heat treatment and a low content reducing sugar in flours. Previous studies19,20 of non-enzymic browning have shown that the increase in furosine is linear during the initial phase of heating but that in the later stages of the Maillard reaction furosine content decreases. Alpha-amylase hydrolysed flours As would be expected, the reducing sugar content increased during amylolysis of the toasted cereal flours (Table IV), favouring the Maillard reaction, and thus leading to increased furosine contents in all the samples (Table II), but especially in samples where rice flour was present in high proportion. This was also observed for the product containing eight cereals and honey. A large increase in furosine content was observed for the rice flour sample, which was six times greater than that observed for the toasted sample, but smaller increases were observed for wheat and oats. The reducing sugar content of the amylase-hydrolysed toasted rice flour sample was 22·8% (w/w) and for wheat and oat approximately 8% (w/w). Roller-dried flours Furosine levels in the roller-dried samples indicated a higher level of damage than for the toasted or the toasted and amylase-hydrolysed samples. The furosine content of the roller-dried rice sample (816 mg/100 g of protein) was much
E. Guerra-Hernandez et al.
174
Figure 1 High performance liquid chromatography (HPLC) chromatograms of furosine after different types of processing of infant oat flour products. (a) Raw, (b) toasted, (c) amylase-hydrolysed, (d) roller-dried.
Table II
Furosine contents of infant cereal foods after different types of processing Furosine content (mg/100 g of protein)
Samplesa Wheat 7 cereals–soy 7 cereals–soy–fruits 7 cereals–soy–biscuit–fruits 8 cereals–soy–honey Rice Rice–corn Rice–corn–soy Rice–corn–soy–fruits Oat
Proteine content (%, w/w) b
9·9 11·8b 7·9b 8·9b 13·5b 6·0c 7·7d 10·8d 8·1d 11·3d
Raw
Toasted
Hydrolised with alpha-amylase
10·6 nd
14·3 16·5
30·5 22·1
nd tr 23·1 22·8
13·7 7·3 17·5 14·4
75·4 44·3 68·1 66·0
11·9
13·3
24·7
Roller-dried 176·3 273·9 400·0 358·0 733·2 816·3 540·3 820·2 1026·0 143·0
a
n=2; bN×5·70; cN×5·95; dN×6·25; eprotein expressed in dry matter. tr=traces. nd=not determined.
greater than those for the roller-dried oat (143 mg/ 100 g of protein) and wheat (176 mg/100 g of protein) samples (Table II). Samples containing seven or eight cereals, which contained a high
proportion of wheat, had higher furosine contents after roller-drying than the corresponding single wheat sample; this could be due to the presence of soy in their formulation. The presence of honey
Furosine formation infant cereal processing
Mixture of cereal flours (B) and fruits
Table III Furosine contents of single cereal flour samples produced at different toasting temperatures
Floursa Wheat Rice Oat
Furosine (mg/100 g of protein)
Proteine content (%, w/w) b
10·6 6·4c 12·0d
140 °C
150 °C
15·4 7·1 10·4
11·1 9·0 12·4
a
n=2; bN×5·70; cN×5·95; dN×6·25; eprotein expressed in dry matter.
Table IV
Reducing sugar contents of raw and hydrolysed flour samples Reducing sugarsb,c (%, w/w)
Samplesa
Raw
Hydrolysed
Wheat 7 cereals–soy 8 cereals–soy–honey Rice Rice–corn Rice–corn–soy Oat
0·3 0·6 0·6 0·3 0·4 0·5 0·7
7·9 8·6 15·5 22·8 13·5 12·2 8·5
a
n=2; bsugars expressed in dry matter; creducing sugars expressed calculated as maltose.
in the product containing eight cereals may explain the high level of furosine in that sample. Infant cereal samples with a large proportion of rice had high furosine values; this cereal seems the most susceptible to Maillard browning reaction. The Maillard reaction is maximal at water activity values in the range 0·5–0·721. Thus, during the roller-drying process, intermediate moisture systems are produced and lysine losses increased dramatically. Table V
Biscuits Banana powder Orange powder Soy flour a
In order to study the influence of other ingredients on furosine content, analysis was carried out on three dried samples (containing seven cereals and fruits, with and without biscuit, and a product containing rice, corn, soy and fruits) to which fruit powders and biscuits had been added.
Ingredients The furosine levels found in the ingredients are showed in Table V. Banana powder had a higher furosine content than orange powder. The latter has less glucose and fructose than the former as well as higher acidity. The ingredient with the lowest furosine content was soy flour sample (29 mg/100 g protein). The biscuit sample had a furosine content of 327 mg/100 g of protein. Products The furosine contents of the cereal samples with added ingredients are shown in Table II. Samples with fruits showed an increase in furosine content compared with the same sample without fruits. The furosine of cereal with fruit samples proceeds from the flours thermically treated as well as from the powder ingredients added. In conclusion, during the processing of infant cereal flours, furosine is mainly formed during the roller-drying step, since at this stage there is a combination of high temperature, intermediate aw and increased reducing sugar levels. Considerable increases in furosine contents were also observed during the amylolytic hydrolysis step. Furosine may therefore be a good indicator of the degree of Maillard reaction induced damage during the processing of infant cereal foods, and it may be possible to use it as a quality parameter for this type of infant food.
Furosine contents of ingredients used in producing infant cereal products
Ingredientsa
175
Protein contentd (%, w/w)
Furosine (mg/100 g of protein)
7·0b 5·6c 8·3c 47·1c
327 1902 371 29
n=2; bN×5·70; cN×6·25; dprotein expressed in dry matter.
E. Guerra-Hernandez et al.
176
Acknowledgements This work has been supported by the Comisio´n Interministerial de Ciencia y Tecnologı´a (Project ALI 970606-CO2-01). The authors would like to thank Abbott Laboratories for providing the samples.
11.
12.
REFERENCES 1. Kaanane, A. and Labuza, T.P. The Maillard reaction in foods. In ‘The Maillard Reaction in Aging, Diabetes and Nutrition’, ( J.W. Baynes and V.M. Monnier, eds), Alan R. Liss, New York (1989) pp 301–327. 2. Gil, A., Morales, D. and Valverde, E. Process for the preparation of ground cereal based foods and food products obtained thereby. European patent EP 453390, 23 October 1991 and 12 January 1994. 3. Friedman, M. Food browning and its prevention: an overview. Journal of Agricultural and Food Chemistry 44 (1996) 631–653. 4. O’Brien, J.O. and Morrisey, P.A. Nutritional and toxicological aspects of the Maillard browning reaction in foods. Critical Review of Food Science and Nutrition 8 (1989) 211–248. 5. Erbersdobler, H.F. and Hupe, A. Determination of lysine damage and calculation of lysine bio-availability in several processed foods. Zeitschrift fu¨r Erna¨hrungswissenschaft 30 (1991) 46–49. 6. Lopez-Fandin˜o, R., Corzo, N., Villamiel, M., Delgado, T., Olano, A. and Ramos, M. Assessment of quality of commercial UHT milks by chromatographic and electrophoretic methods. Journal of Food Protection 56 (1993) 263–265. 7. Henle, T., Zehetner, G. and Klostemeyer, H. Fast and sensitive determination of furosine. Zeitschrift fu¨r Lebensmittel-Untersuchung und-Forschung 200 (1995) 235–237. 8. Cefalu, W.T., Bell-Farrow, A., Wang, Z.Q. and Ralapati, S. Determination of furosine in biomedical samples employing and improved hydrolysis and HPLC technique. Carbohydrate Research 215 (1991) 117–125. 9. Harris, C.H. and Johnson, J.M. Monitoring non enzymatic browning in cakes prepared with high fructose corn syrup by high performance liquid chromatography. Journal of Food Quality 10 (1987) 417–424. 10. Resmini, P. and Pellegrino, L. Analysis of food heat
13.
14.
15.
16. 17. 18.
19.
20.
21.
damage by direct HPLC of furosine. International Chromatography Laboratory 6 (1991) 7–11. Carratu´, B., Boniglia, C., Filesi, C. and Bellmonte, G. Influenza del trattamento tecnologico sugli alimenti: Determinazione per cromatografia ionic di piridosina e furosina in prodotti dietetici ed alimentari. La Rivista di Scienza dell’Alimentazione 22 (1993) 455–457. Guerra-Herna´ndez, E. and Corzo, N. Furosine determination in baby cereals by ion-pair reversed-phase liquid chromatography. Cereal Chemistry 73 (1996) 729– 731. Finot, P.A., Bricout, J., Viani, R. and Mauron, J. Identification of a new lysine derivative obtained upon acid hydrolysis of heated milk. Experientia 24 (1968) 1097– 1099. Resmini, P., Pellegrino, L. and Batelli, G. Accurate quantification of furosine in milk and dairy products by a direct HPLC method. Italian Journal of Food Science 3 (1990) 173–183. Delgado, T., Corzo, N., Santa-Maria, G., Jimeno, M.L. and Olano, A. Determination of furosine in milk samples by ion pair reversed phase liquid chromatography. Chromatographia 33 (1992) 374–376. Association of Official Analytical Chemists. Official Methods of Analysis, AOAC, Washington, D.C. (1990) 15th edn. Method 920.87. Association of Official Analytical Chemists. Official Methods of Analysis, AOAC, Washington, D.C. (1990) 15th edn. Method 939.03. Acquistucci, R., Panfilini, G. and Marconi, E. Applications of the microwave hydrolysis to furosine determination in cereal and dairy foods. Journal of Agricultural and Food Chemistry 44 (1996) 3855–3857. Erbersdobler, H.F., Greulich, H.G. and Groß, A. Modellversuche u¨ber Einfu¨sse der Haushaltstechnik auf die Nahrungsmittelqualita¨t. Schriftenreihe der Agraewiss Fakulta¨t der Univ. Kiel Paul Parey, Hamburg und Berlin 63 (1982) 201–207. Hurrell, R.F., Finot, P.A. and Ford, J.E. Storage of milk powders under adverse conditions. 1. Losses of lysine and of other essential amino acids as determined by chemical and microbiological methods. British Journal of Nutrition 49 (1983) 343–354. Labuza, T.P. The effect of water activity on reaction kinetics of food deterioration. Food Technology 34 (1980) 36–41.
Maillard Reaction Evaluation by Furosine Determination During Infant Cereal Processing E. Guerra-Hernandez∗, N. Corzo† and B. Garcia-Villanova∗ ∗Universidad de Granada, Departamento de Nutricio´n y Bromatologı´a, 18012 Granada, Spain; †Instituto de Fermentaciones Industriales (C.S.I.C.), Juan de la Cierva 3, 28006 Madrid, Spain Received 15 September 1997
ABSTRACT The furosine content of infant cereals has been used to evaluate the thermal damage caused to those foods during their manufacture (toasting, amylolysis and drying steps). The mean furosine contents of raw and toasted flours were 13·7 and 13·8 mg/100 g of protein, respectively. The furosine content increased to 47·3 mg/100 g of protein after hydrolysis using alpha-amylase. The amylolysis process raised the reducing sugar content from 0·5% for raw samples to 12·7%. The furosine content of infant cereals rose to average levels of 500 mg/100 g of protein after roller-drying. The infant cereals with the highest levels of reducing sugars, with or without soy flour, exhibited the highest furosine contents. Furosine is a useful indicator to evaluate the progress of the Maillard reaction during the amylolysis and roller-drying stages of infant cereals manufacture. 1999 Academic Press
Keywords: Maillard reaction, furosine, infant cereal processing.
INTRODUCTION The manufacture of baby cereals represents a relatively simple food system where non-enzymic browning involving the Maillard reaction (MR) can occur easily. This is due to the type and concentration of the main reagents involved, the water activity (aw) of the system during drying, and the time-temperature conditions of the thermal process1. Infant cereals must be processed to improve their dispersibility in liquids and their digestibility since 3–4-month old babies have a reduced capacity to digest starch. Baby cereal processing involves sequential toasting, hydrolysis with alpha-amylase (amylolysis) and drying stages2. During the amylolysis process, the : HPLC=high performance liquid chromatography; PTFE=polytetrafluoroethylene.
∗ Corresponding author: E. Guerra-Hernandez. Tel: (958) 243867; Fax: (958) 243869; E-mail: [email protected] 0733–5210/99/020171+06 $30.00/0
reducing sugar level increases, thus facilitating the Maillard reaction during the toasting and drying processes. These treatments improve the sensory characteristics and starch digestibility, but adverse effects can occur. The loss in nutritional quality and potentially in safety is attributed to destruction of essential amino acids, decrease in digestibility, inhibition of proteolytic and glycolytic enzymes, interaction with metal ions, and formation of antinutritional and toxic compounds3,4. The evaluation of the early stages of Maillard reaction, can be achieved by the determination of furosine (e-N-(furoylmethyl L-lysine), an amino acid formed during acid hydrolysis of the Amadori compound fructosyl-lysine, lactulosyl-lysine and maltulosyl-lysine produced by reaction of e-amino groups of lysine with glucose, lactose and maltose5, respectively. This amino acid has been used to measure the early stages of Maillard reaction, not only in foods6,7, but also in biological samples8. In cereal products, furosine determination has been used to monitor the extent of non-enzymic browning in cakes9 and to monitor pasta processing10. Few studies have been carried out regarding 1999 Academic Press
E. Guerra-Hernandez et al.
172
Table I
Composition of different samples used for furosine determination
Samples
Type of cereal
Single flours
Wheat Oat Rice
Mixture of flours (A)
Ingredients
Seven cereals: wheat, rice, barley, rye, oat, corn and millet
Soy
Eight cereals: wheat, rice, barley, rye, oat, corn, millet and sorghum
Soy and honey
Rice, corn
Mixture of flours (B)
Rice, corn
Soy
Seven cereals: wheat, rice, barley, rye, oat, corn and millet
Soy, biscuit, orange and banana powders
Seven cereals: wheat, rice, barley, rye, oat, corn and millet
Soy, orange and banana powders
Rice, corn
Soy, orange and banana powders
the Maillard reaction in infant cereal products, although Carratu et al.11 and Guerra-Herna´ndez and Corzo12 determined furosine in two and 10 commercial baby cereal products, respectively. Since infant cereals are products for infant feeding, it is of interest to know the extent of Maillard reaction during their processing. The objective of this work was to study the early stages of Maillard reaction by furosine determination at the different steps of infant cereals processing as well as the usefulness of furosine as a parameter to control the process.
dried and then stored at −50 °C until they were analysed. Reagents preparation All chemicals were of analytical grade. A furosine standard was prepared by acid hydrolysis of eN-(1-deoxy-D-fructosyl)-L-lysine according to the procedure of Finot et al.13 and its authenticity confirmed by UV and NMR spectroscopy15. A standard stock solution, containing 1·2 mg/ 100 mL of furosine, was used to prepare the working standard solution.
EXPERIMENTAL Samples
Methods
Infant cereal samples were obtained from a Spanish dietetic products company2. The compositions of samples used in this study are shown in Table I. Single and mixed cereal flours (A) were analysed raw (before being processed), after toasting at 150 °C, after hydrolysis of the toasted material with alpha-amylase at 45–50 °C, and after rollerdrying of the amylase-treated material. Mixed cereal flours (B) with fruits were analysed only after roller-drying. Another study was carried out with wheat, rice and oat flour samples at two toasting temperatures (140 °C and 150 °C). Solid samples were stored at −50 °C until they were analysed. Hydrolysed samples were freeze-
Furosine determination Furosine was determined after hydrolysing the sample (0·15 g) with 10·6 HCl (4·5 mL) at 110 °C for 24 h in a Pyrex screw-cap vial with PTFEfaced septa. High-purity helium gas was bubbled through the solution for 2 min12. Sample preparation for HPLC analysis was carried out following the method of Resmini et al.14. The hydrolysate was filtered with a medium-grade paper filter. A portion (0·5 mL) of the filtrate was applied to a Sep-pak C18 cartridge (Millipore), prewetted with methanol (5 mL) and water (10 mL), then eluted with 3 HCl (3 mL) and evaporated under vacuum. The dried sample was dissolved in a mixture (3 mL) of water, acetonitrile and formic
Furosine formation infant cereal processing
acid (95:5:0·2) before HPLC analysis. Samples were analysed in duplicate. Furosine was quantified by ion-pair reversedphase HPLC (Beckman System Gold chromatograph) according to the method of Delgado et al.15, using a Spherisorb ODS2 5-l column (250×4·6 mm, i.d.; Phenomenex) operating at room temperature. The mobile phase consisted of a solution of 5m sodium heptane sulphonate with 20% (v/v) acetonitrile and 0·2% (v/v) formic acid. The elution was isocratic and the flow rate was 1·2 mL/min. The UV detector (Beckman) was set a 280 nm and the injection volume was 50 lL. Calibration of the chromatographic system for furosine determination was made by the external standard method.
Protein and reducing sugars determination Protein determination was carried out by the Kjeldahl method16. Reducing sugars were determined by a titrimetric method17. RESULTS AND DISCUSSION The precision of the entire assay procedure including acid hydrolysis, sample preparation, and RP-HPLC analysis was evaluated for toasted wheat samples (n=7). The relative standard deviation (RSD) was 6·21% obtained on a sample with an average furosine value of 14·3 mg/100 g of protein. The relative standard deviation (RSD) for samples with high levels of furosine was 3·73%12. Single and mixture of cereal (A) samples processing Figure 1 shows the HPLC chromatograms of furosine at different stages of infant oat flour processing. The greatest increase in furosine content was observed during the final roller-drying step. The furosine contents after the different processing steps of the samples studied are shown in Table II.
Raw flours The furosine contents of raw cereal flours ranged from only trace levels (rice flour) through 10·6 mg/ 100 g protein (wheat flour) to 23 mg/100 g of protein (rice-corn flour) (Table II). Furosine in raw flour samples may originate during grain growth, milling or storage. Similar levels of furosine were
173
found in durum wheat semolina in previous studies10,18, but lower values were found in wheat grain. These differences may be due to the milling and storage conditions used for the flour.
Toasted flours Toasting produced changes in furosine content in the cereal flours studied. Toasted wheat, rice and oat flours had increased furosine contents compared with the raw flours, especially rice (7·3 mg/ 100 g of protein). The furosine values for rice–corn and rice–corn–soy flours decreased after toasting, however (Table II). Since the usual temperatures used in commercial toasting processes are close to 140–150 °C, a study of the influence of temperature on Maillard reaction was carried out with other single infant cereal flour samples (Table III). Toasting at 150 °C gave a lower furosine content for wheat flour than toasting at 140 °C, whereas, for rice and oat flours, the higher temperature gave rise to greater levels of furosine. This observation may be explained by the degradation of the Amadori compound under the more severe heat treatment and a low content reducing sugar in flours. Previous studies19,20 of non-enzymic browning have shown that the increase in furosine is linear during the initial phase of heating but that in the later stages of the Maillard reaction furosine content decreases. Alpha-amylase hydrolysed flours As would be expected, the reducing sugar content increased during amylolysis of the toasted cereal flours (Table IV), favouring the Maillard reaction, and thus leading to increased furosine contents in all the samples (Table II), but especially in samples where rice flour was present in high proportion. This was also observed for the product containing eight cereals and honey. A large increase in furosine content was observed for the rice flour sample, which was six times greater than that observed for the toasted sample, but smaller increases were observed for wheat and oats. The reducing sugar content of the amylase-hydrolysed toasted rice flour sample was 22·8% (w/w) and for wheat and oat approximately 8% (w/w). Roller-dried flours Furosine levels in the roller-dried samples indicated a higher level of damage than for the toasted or the toasted and amylase-hydrolysed samples. The furosine content of the roller-dried rice sample (816 mg/100 g of protein) was much
E. Guerra-Hernandez et al.
174
Figure 1 High performance liquid chromatography (HPLC) chromatograms of furosine after different types of processing of infant oat flour products. (a) Raw, (b) toasted, (c) amylase-hydrolysed, (d) roller-dried.
Table II
Furosine contents of infant cereal foods after different types of processing Furosine content (mg/100 g of protein)
Samplesa Wheat 7 cereals–soy 7 cereals–soy–fruits 7 cereals–soy–biscuit–fruits 8 cereals–soy–honey Rice Rice–corn Rice–corn–soy Rice–corn–soy–fruits Oat
Proteine content (%, w/w) b
9·9 11·8b 7·9b 8·9b 13·5b 6·0c 7·7d 10·8d 8·1d 11·3d
Raw
Toasted
Hydrolised with alpha-amylase
10·6 nd
14·3 16·5
30·5 22·1
nd tr 23·1 22·8
13·7 7·3 17·5 14·4
75·4 44·3 68·1 66·0
11·9
13·3
24·7
Roller-dried 176·3 273·9 400·0 358·0 733·2 816·3 540·3 820·2 1026·0 143·0
a
n=2; bN×5·70; cN×5·95; dN×6·25; eprotein expressed in dry matter. tr=traces. nd=not determined.
greater than those for the roller-dried oat (143 mg/ 100 g of protein) and wheat (176 mg/100 g of protein) samples (Table II). Samples containing seven or eight cereals, which contained a high
proportion of wheat, had higher furosine contents after roller-drying than the corresponding single wheat sample; this could be due to the presence of soy in their formulation. The presence of honey
Furosine formation infant cereal processing
Mixture of cereal flours (B) and fruits
Table III Furosine contents of single cereal flour samples produced at different toasting temperatures
Floursa Wheat Rice Oat
Furosine (mg/100 g of protein)
Proteine content (%, w/w) b
10·6 6·4c 12·0d
140 °C
150 °C
15·4 7·1 10·4
11·1 9·0 12·4
a
n=2; bN×5·70; cN×5·95; dN×6·25; eprotein expressed in dry matter.
Table IV
Reducing sugar contents of raw and hydrolysed flour samples Reducing sugarsb,c (%, w/w)
Samplesa
Raw
Hydrolysed
Wheat 7 cereals–soy 8 cereals–soy–honey Rice Rice–corn Rice–corn–soy Oat
0·3 0·6 0·6 0·3 0·4 0·5 0·7
7·9 8·6 15·5 22·8 13·5 12·2 8·5
a
n=2; bsugars expressed in dry matter; creducing sugars expressed calculated as maltose.
in the product containing eight cereals may explain the high level of furosine in that sample. Infant cereal samples with a large proportion of rice had high furosine values; this cereal seems the most susceptible to Maillard browning reaction. The Maillard reaction is maximal at water activity values in the range 0·5–0·721. Thus, during the roller-drying process, intermediate moisture systems are produced and lysine losses increased dramatically. Table V
Biscuits Banana powder Orange powder Soy flour a
In order to study the influence of other ingredients on furosine content, analysis was carried out on three dried samples (containing seven cereals and fruits, with and without biscuit, and a product containing rice, corn, soy and fruits) to which fruit powders and biscuits had been added.
Ingredients The furosine levels found in the ingredients are showed in Table V. Banana powder had a higher furosine content than orange powder. The latter has less glucose and fructose than the former as well as higher acidity. The ingredient with the lowest furosine content was soy flour sample (29 mg/100 g protein). The biscuit sample had a furosine content of 327 mg/100 g of protein. Products The furosine contents of the cereal samples with added ingredients are shown in Table II. Samples with fruits showed an increase in furosine content compared with the same sample without fruits. The furosine of cereal with fruit samples proceeds from the flours thermically treated as well as from the powder ingredients added. In conclusion, during the processing of infant cereal flours, furosine is mainly formed during the roller-drying step, since at this stage there is a combination of high temperature, intermediate aw and increased reducing sugar levels. Considerable increases in furosine contents were also observed during the amylolytic hydrolysis step. Furosine may therefore be a good indicator of the degree of Maillard reaction induced damage during the processing of infant cereal foods, and it may be possible to use it as a quality parameter for this type of infant food.
Furosine contents of ingredients used in producing infant cereal products
Ingredientsa
175
Protein contentd (%, w/w)
Furosine (mg/100 g of protein)
7·0b 5·6c 8·3c 47·1c
327 1902 371 29
n=2; bN×5·70; cN×6·25; dprotein expressed in dry matter.
E. Guerra-Hernandez et al.
176
Acknowledgements This work has been supported by the Comisio´n Interministerial de Ciencia y Tecnologı´a (Project ALI 970606-CO2-01). The authors would like to thank Abbott Laboratories for providing the samples.
11.
12.
REFERENCES 1. Kaanane, A. and Labuza, T.P. The Maillard reaction in foods. In ‘The Maillard Reaction in Aging, Diabetes and Nutrition’, ( J.W. Baynes and V.M. Monnier, eds), Alan R. Liss, New York (1989) pp 301–327. 2. Gil, A., Morales, D. and Valverde, E. Process for the preparation of ground cereal based foods and food products obtained thereby. European patent EP 453390, 23 October 1991 and 12 January 1994. 3. Friedman, M. Food browning and its prevention: an overview. Journal of Agricultural and Food Chemistry 44 (1996) 631–653. 4. O’Brien, J.O. and Morrisey, P.A. Nutritional and toxicological aspects of the Maillard browning reaction in foods. Critical Review of Food Science and Nutrition 8 (1989) 211–248. 5. Erbersdobler, H.F. and Hupe, A. Determination of lysine damage and calculation of lysine bio-availability in several processed foods. Zeitschrift fu¨r Erna¨hrungswissenschaft 30 (1991) 46–49. 6. Lopez-Fandin˜o, R., Corzo, N., Villamiel, M., Delgado, T., Olano, A. and Ramos, M. Assessment of quality of commercial UHT milks by chromatographic and electrophoretic methods. Journal of Food Protection 56 (1993) 263–265. 7. Henle, T., Zehetner, G. and Klostemeyer, H. Fast and sensitive determination of furosine. Zeitschrift fu¨r Lebensmittel-Untersuchung und-Forschung 200 (1995) 235–237. 8. Cefalu, W.T., Bell-Farrow, A., Wang, Z.Q. and Ralapati, S. Determination of furosine in biomedical samples employing and improved hydrolysis and HPLC technique. Carbohydrate Research 215 (1991) 117–125. 9. Harris, C.H. and Johnson, J.M. Monitoring non enzymatic browning in cakes prepared with high fructose corn syrup by high performance liquid chromatography. Journal of Food Quality 10 (1987) 417–424. 10. Resmini, P. and Pellegrino, L. Analysis of food heat
13.
14.
15.
16. 17. 18.
19.
20.
21.
damage by direct HPLC of furosine. International Chromatography Laboratory 6 (1991) 7–11. Carratu´, B., Boniglia, C., Filesi, C. and Bellmonte, G. Influenza del trattamento tecnologico sugli alimenti: Determinazione per cromatografia ionic di piridosina e furosina in prodotti dietetici ed alimentari. La Rivista di Scienza dell’Alimentazione 22 (1993) 455–457. Guerra-Herna´ndez, E. and Corzo, N. Furosine determination in baby cereals by ion-pair reversed-phase liquid chromatography. Cereal Chemistry 73 (1996) 729– 731. Finot, P.A., Bricout, J., Viani, R. and Mauron, J. Identification of a new lysine derivative obtained upon acid hydrolysis of heated milk. Experientia 24 (1968) 1097– 1099. Resmini, P., Pellegrino, L. and Batelli, G. Accurate quantification of furosine in milk and dairy products by a direct HPLC method. Italian Journal of Food Science 3 (1990) 173–183. Delgado, T., Corzo, N., Santa-Maria, G., Jimeno, M.L. and Olano, A. Determination of furosine in milk samples by ion pair reversed phase liquid chromatography. Chromatographia 33 (1992) 374–376. Association of Official Analytical Chemists. Official Methods of Analysis, AOAC, Washington, D.C. (1990) 15th edn. Method 920.87. Association of Official Analytical Chemists. Official Methods of Analysis, AOAC, Washington, D.C. (1990) 15th edn. Method 939.03. Acquistucci, R., Panfilini, G. and Marconi, E. Applications of the microwave hydrolysis to furosine determination in cereal and dairy foods. Journal of Agricultural and Food Chemistry 44 (1996) 3855–3857. Erbersdobler, H.F., Greulich, H.G. and Groß, A. Modellversuche u¨ber Einfu¨sse der Haushaltstechnik auf die Nahrungsmittelqualita¨t. Schriftenreihe der Agraewiss Fakulta¨t der Univ. Kiel Paul Parey, Hamburg und Berlin 63 (1982) 201–207. Hurrell, R.F., Finot, P.A. and Ford, J.E. Storage of milk powders under adverse conditions. 1. Losses of lysine and of other essential amino acids as determined by chemical and microbiological methods. British Journal of Nutrition 49 (1983) 343–354. Labuza, T.P. The effect of water activity on reaction kinetics of food deterioration. Food Technology 34 (1980) 36–41.