- Email: [email protected]
Separation and identification of sugars and maltodextrines by thin layer chromatography: Application to biological fluids and human milk
Talanfa, Vol. 39, No. 11, pp. 1493-1498, 1992 Printed in GreatBritain.All rightsreserved
0039-9140/92 $5.00 + 0.00
PergamonPressLtd
SEPARATION AND IDENTIFICATION OF SUGARS AND MALTODEXTRINES BY THIN LAYER CHROMATOGRAPHY: APPLICATION TO BIOLOGICAL FLUIDS AND HUMAN MILK F. BOSCH-REIG, M. J. Macon,
M. D. MINANA
Departamento de Quimica Analitica, Fact&ad de Ciencias Quimicas, Valencia, Spain M. L. CABELLO*
Unidad de Metabolopatias CongWas, Hospital Universitario “La Fe”, Avda. Campanar, 46009 Vale&a, Spain (Received 27 March 1992. Accepted 1 April 1992) Summary-A monodimensional thin layer chromatography method to separate several sugars of clinical interest is described. The separation and identification of 14 sugars (L-fucose, D-galactose, D-glucose, lactose, N-acetylglucosamine, D-maltose, D-manose, L-sorbase, fructose, D-xylosc, glucuronic acid, N-acetyllactosamine, 3’ and 6’ sialyllactose) and maltodextrines (GrGs) is possible by using two different eluents mixtures, as well as two different detection reagents. The method has been applied to separate sugars, maltodextrines and oligosaccharides in several biological fluids (blood, urine and faeces), in an infant milk and in human milk. It is a very simple technique (with a high sensitivity) that can be used in any lab.
The presence of anomalous amounts of carbohydrate8 can be detected in blood, urine and faeces of patients with inborn errors in the metabolism of carbohydrates, such as essential fructosuria and others.‘-’ In the present method, urine does not need to be desalted as in most cases, either by coltmnP or by precipitation.5.g Over the last few years, the increasing popularity of breastfeeding has stimulated interest in the sugar content of human milk. It is known that the carbohydrate concentration of human milk is about 7 g/100 ml; 80% lactose and the rest oligosaccharides (l-l.2 g/100 ml in nature milk, 2-2.5 g/100 ml in colostrum).‘4 It has not yet been possible to reproduce the structure of these oligosaccharides. They are substituted by maltodextrines in “adapted” milk formula, keeping the total sugar concentration at 7-8 g/100 rn.15 The biological role of human milk oligosaccharides is not completely understood, but it
*Author for correspondence.
seems clear that they have a very important anti-infective action for breast-fed infants.16 Kobata et al. isolated and characterized most of the oligosaccharides in human milk using very sophisticated technology.“8 At first they used pooled milk from several donors as starting material, but then observed that the type of oligosaccharides present in individual samples of milk can vary with the ABO or Lewis blood type of the donor. The enzymes involved in their synthesis are also responsible for the formation of the structural determinants of these blood types. All oligosaccharides in human milk have some general structural principles in common, they all derive from lactose and contain, besides galactose and glucose, N-acetylglucosamine and/or more of L-fucose and N-acetyl-neuraminic acid molecules. Another advantage of the method is that sugars may be identified without using internal standards. Thin layer chromatography has been applied specifically to the study of the role placed by the oligosaccharides obtained from human blood, urine and milk’“~“~‘3in metabolic pathways.
1493
F.
1494
BGSCH-REIG
Several sugars obtained from animal urines have been studied by thin layer chromatography and high pressure liquid chromatography.‘2 It can be seen in the references in the literature that certain sugars have been studied, however a complete study of separation, identification and application to biological fhtids of carbohydrates has not been done.
EXPERIMENTAL Samples and solvent systems
Aqueous solutions of the following sugars were prepared: L-fucose, D-galactose, D-glucose, lactose, Nacetylglucasamine, D-maltose, D -manose, L sorbose, D-xylose, fructose (0.7 g/100 ml), glucuronic acid (0.21 g/100 ml), N-acetyllactosamine (5.00 g/100 ml) 3’ and 6’ sialyllactose (1.70 g/100 ml) and maltodextrines (G&J (8.33 g/100 ml). All sugars were obtained from commercial sources (Merck, Sigma). Less concentrated solutions were obtained by diluting the original ones. Blood samples, as well as fresh and twentyfour hour urine samples were taken from the ones available in the laboratory. Faeces samples belonged to preterm infants. The milk samples used were from either one donor, 15 days and 45 days after the delivery, or several donors, 15 days after the delivery. An infant milk formula containing lactose and maltodextrines (56.9 g/100 g milk), 15% in water. Blood was delipidated using a chloroform: metanol 2 : 1 mixture in a ratio 1: 10, mixed for 5 min and centrifuged for 25 min at 5000 rpm. Then it was ready to be applied to the plate. Urine required no previous treatment and could be directly applied to the plate. Faeces were analysed after being diluted in distilled water 1:2 and centrifuged 25 min at 5000 rpm. The supernatant was applied to the plate. Milk samples were delipidated by centrifugation at 1500 rpm for 30 min and diluted 1: 2 with distilled water. One-millilitre aliquots were kept at -20” until they were analysed. Two elements mixtures were used: in-butanol/ ethanol/water (3 : 2 : 1) and pyridin/ethyl acetate/ acetic acid/water (5 : 5 : 3 : 1).
et al.
Chromatography
Cellulose plates (20 x 20 cm, O.l-mm thick) were used (Merck Art. 5716). Cellulose plates (20 x 20 cm, 0.5~mm thick) were used for preparative chromatography (Merck Art.). Aliquots (3 pl), were spotted on the 0.1~mm plates with a micro syringe, at 1.5~cm intervals and 2 cm from the lower edge of the plates. Then they were developed by the ascending technique. A three-fold development, with the same eluent mixture, is used in order to improve separation. A good separation was obtained when the eluent was allowed to run 14.5 cm from the edge of the plate. The time required for a single run is approximately two and a half hours. In the preparative chromatography milk samples aliquots (850 ~1) were applied as a streak on the 0.5-mm plate. A three-fold development was also used. Guide strips were developed with silver nitrate reagent to locate the bands. The required bands on the unsprayed area were scrapped off onto test tubes, 1 ml of distilled water was added to each and tubes were shaken thoroughly. Sugar was separated from cellulose powder by centrifugation at 4.500 rpm for 5 min. Detection reagents
The following reagents were used: silver nitrate solution prepared by dissolving 3 g of silver nitrate in 12 ml of distilled water adding 500 ml of acetone. Ethanolic sodium hydroxide solution prepared by diluting 50 ml of 10N sodium hydroxide in 450 ml of ethanol. Sodium thiosulphate (5%) was also used. The developed plates were allowed to dry at room temperature and dipped into AgNO solution for one minute. When dry they were sprayed with the ethanolic sodium hydroxide solution until dark brown spots appeared on a light brown background. If the plates must be kept for a long time, they must be sprayed with a 5% NaZS203 solution once they are dry. Elson -Morgan reagent
1% Acetillacetone in butanol is treated with a l/20 volume of 50% potassium hydroxideethanol (1: 4 v/v). p-Dimethylaminobenzaldehide solution: 1 g of p-dimethylbenaldehide is dissolved in 30 ml
Separation and identification of sugars and maltodextrines
1495
Table 2. R, values, using two different eluents mixtures, of maltodextrines in aqueous solutions
of ethanol of concentrated HCl and 30 ml of N-butanol. When the plate was dry, it was sprayed with the acetylacetone solution, heated at 100” for 5 min and then sprayed with pdimenthylbenzaldehide solution. N-Acetylaminated sugars appear immediately as violet spots.
Sugar GZ G3
G4 G, G, G, G*
R, (eluent 1)
R, (eluent 2)
0.73 0.65 0.56 0.46 0.38 0.31 0.24
0.48 0.35 0.26 0.19 0.14 0.10 -
RESULTS AND DISCUSSION
Aqueous solutions of the following sugars were chromatographed on 0. I-mm cellulose plates: N-acetylglucosamine, fucose, xylose, manose, sorbose, glucose, galactose, N-acetyllactosamine, glucuronic acid, maltose, lactose, 3’-sialyllactose and 6’-sialyllactose, using Nbutanol/ethanol/water (3 : 2 : 2) as eluent and with a 3-fold development. All sugars were detected with both silver nitrate and ElsonMorgan reagents. The RIvalues and sensitivities are shown in Table 1. Similarly, an aqueous solution of maltodextrines was also chromatographed using two different eluents: (1) n-butanol/ethanol/water (3 : 2 : 2) and (2) pyridine/ethyl acetate/acetic acid/water (5 : 5 : 1: 3). Then maltodextrines were located with silver nitrate reagent. The method sensitivity for maltodextrines is 6.25 pg and the Rf values are shown in Table 2. As can be seen in Table 1 fucose and Nacetylglucosamine have the same RI but they could be identified by the detection reagents. Both of them can be visualized with silver nitrate reagent, but only N-acetylglucosamine can be detected with Elson-Morgan reagent. Table 2 shows that only G2 to G, components of maltodextrines can be separated with eluent (2). G2 to G8 can be separated with eluent (1).
However G9 and G10 components cannot be separated either with eluent (1) or with eluent (2). It can be said that the described aqueous solutions of sugars and maltodextrines have been identified either by their Rfor by selecting the suitable detection reagent. When sugars were added to blood, fresh urine, twenty-four hour urine and faeces, these biological fluids were chromatographed and it was seen that there is very good agreement between the Rf of the standard solutions and those of the biological fluids, as shown in Table 3. Samples of infant milk were also chromatographed. Its sugar content according to the manufacturer was lactose and maltodextrines. N-Butanol/ethanol/water (3 : 2 : 2) was used as eluent and the detection reagent was silver nitrate. On the chromatogram shown in Fig. 1 can be seen that lactose and maltodextrines were identified, as well as minor sugars like galactose and glucose. Therefore, this procedure allows the identification of sugars with a sensitivity that may be considered high. Seven spots were detected when samples of milk from a donor who had delivered a full term infant 15 days before, were chromatographed.
Table 1. RFvalues and sensitivity in pg of sugars in aqueous solutions
Table 3. R, values of sugars in several biological fluids (blood, faeces and urine) Sugar
Sugar 6’-Sialillactose 3’-Sialillactose Lactose Maltose Glucuronic Acid N-Acetillactosamine Galactose Glucose Sorbose Manose Xilose Function Fucose N-Acetilglucosamine
TM 3Q:ll-6
R,
0.35 0.42 0.48 0.48 0.55 0.58 0.65 0.69 0.70 0.73 0.78 0.74 0.81 0.8 1
(AgI’f6,
3.19 3.19 0.56 0.21 0.09 3.0 0.14 0.11 0.21 0.21 0.11 0.11 0.11 0.42
R.) @son-%rganR.) 3.0 2.1
R, values
Xilose Fructose Glucose Galactose Glucuronic Acid Lactose
Aqueous solution 0.78 0.14 0.69 0.65 0.55 0.48
Sugar Xilose Fructose Glucose Galactose Glucuronic Acid Lactose
Fresh urine 0.78 0.74 0.69 0.66 0.55 0.48
Blood Faeces 0.78 0.78 0.73 0.74 0.69 0.69 0.65 0.65 0.53 0.54 0.47 0.48 Twenty-four hour urine 0.78 0.74 0.69 0.66 0.55 0.48
F.
1496
~SCH-&lG
et al.
2
I
3
4
5
Fig. 1. Chromatogram of sugars of an infant milk; 1: infant milk :water (1: 1); 2: glucose: 14 mg/lOO ml; 3: galactose: 14 mg/lOO ml; 4: lactose: 70 mg/lOO ml; 5: maltodextrines: 830 mg/lOO ml.
As shown in Fig. 2, spots 1, 2, 3 and 4 were confirmed to be glucose, lactose, 3’-sialyllactose and 6’-sialyllactose respectively. Spots 5, 6 and 7 were confirmed to be fractions of oligosaccharides. They were isolated by preparative chromatography and they all released glucose, galactose and fucose when mild hydrolysis
I
2
3
was carried out. Spot 5 also contains N-acetyl glucosamine as it reacts with Elson-Morgan reagent. Samples of milk of the same donor 45 days after she gave birth were also chromatographed (Fig. 3). The chromatogram was quite similar to the former. However, the fractions of oligo-
4
5
6
Fig. 2. Chromatogram of human milk; 1: human milk: water (1: 1); 2: lactose: 70 mg/lOO ml; 3: 3’ and 6’ sialyllactose: 210 mg/lOO ml; 4: human milk: 3’ and 6’ sialyllactose (1: 1); 5: glucose: 14 mg/lOO ml; 6: human milk: glucose (1: 1)
Separation and identification of sugars and maltodextrines
2
I
3
4
5
6
1497
7
Fig. 3. Chromatogram of human milk from the same donor 1, 3, 5 and 7: sample of milk after 45 days of delivery. 2, 4 and 6: sample of milk after 15 days of delivery.
saccharides were in less concentration and subsequently the glucose concentration was increased. When samples of milk from different donors, who had delivered babies 15 days before, were chromatographed, strong differences were observed in the chromatogram as a function of the donors (Fig. 4).
Therefore, it can be said that the chromatogram of sugars from milk depends fundamentally on the donor, the maturity of the milk and its origin (human and infant milk formula). All of that can be detected by thin layer chromatography which is a very simple technique that can be used in any laboratory.
i I
Fig. 4. Chromatogram
2
3
4
5
6
7
6
9
IO
of sugars of milks from several donors I-10 spots belong to milks from different donors.
F. BOSCH-REIGet al.
1498 REFERENCES
1. I. Lytt, Gardner. Salvat., Enferrmpdades congknitus y endocrinas de la infmcia.
2. Lynch, Raphael, Mellor, Spare, Inwood, Laboratory Methods, 2nd ed. 3. M. Ghebregzabher, S. Rufmi, B. Monaldi and M. Lato, J. Chromatogr., 1976, 127, 133. 4. Vladimir Vitek and Kveta Vitek, ibid., 1971, 60, 381. 5. Idem, ibid., 1977, 143, 65. 6. Gerard Streaker and Anick Lemaire-Poitau, ibid., 1977, 143, 65. 7. M. Lato, B. Brunelli and G. CiulTini and T. Mezzetti, ibid., 1968, 36, 191. 8. Wolfgang Prinz, William Meldrum and Lynne Wilkinson, Clin. Chim. Acta, 1978, 82, 229. 9. Zeijko Zilic and Nenad Blau and Margarethe Knob, J. Chomalogr., 1979, 164, 91. 10. R. Ramphal, C. Carnoy, S. Fieure, J. C. Michalski, N. Hondret, G. Lamblin, G. Strecker and P. Roussel, Inlect. Iunmnn., 1991, 59, 700. 11. P. Scudder, A. M. Lawson, E. F. Hounsell, R. A. Carruthem, R. A. Childs and T. Feiz, Eur. J. Biochem., 1978, 163, 585.
12. C. D. Warren, S. Sadeh, P. F. Daniel, B. Bugge, L. F. James and R. W. Jeanloz, Febs Lert., 1983, 163, 99. 13. A. Cahour, P. Debeire, L. Hartmann and J. Montrenil, Biochem. J., 1983, 211, 55. 14. Comite sobre nutricibn ESPAGN (1982). Acta Paediatr. Stand. Suplemento 302. 15. Comite sobre nutrici6n ESPAGN (1977). Acta Paediatr. Stand. Suplemento 262. 16. J. Cruz and C. Arevalo, Pedriutric Infections Disease 5, S-148. 17. A. Kobata, E. Grollman and V. Ginsburg, Biochem. Biophys. Res. Commun., 1968, 32, 272. 18. A. Kobata and V. Ginsburg, Arch. Biochem. Biophys., 1969, 130, 509. 19. I&m, J. Biol. Chem., 1969, 244, 5496. 20. Idem, ibid., 1970, 245, 1984. 21. I&m, ibid., 1972, 247, 1525. 22. Idem, Arch. Biochem. Biophys., 1972, 150, 273. 23. K. Yamashita and A. Kobata, ibid., 1974, 161, 164. 24. K. Yamashita, Y. Tachibana and A. Kobata, ibid., 174, 582. 25. A. Kobata, Merho& in Enzymology, 1972, 28, 262. 26. A. Kobata, K. Yamashita and Y. Tachibana, Methods in Enzymology, 1978, 50, 216.
0039-9140/92 $5.00 + 0.00
PergamonPressLtd
SEPARATION AND IDENTIFICATION OF SUGARS AND MALTODEXTRINES BY THIN LAYER CHROMATOGRAPHY: APPLICATION TO BIOLOGICAL FLUIDS AND HUMAN MILK F. BOSCH-REIG, M. J. Macon,
M. D. MINANA
Departamento de Quimica Analitica, Fact&ad de Ciencias Quimicas, Valencia, Spain M. L. CABELLO*
Unidad de Metabolopatias CongWas, Hospital Universitario “La Fe”, Avda. Campanar, 46009 Vale&a, Spain (Received 27 March 1992. Accepted 1 April 1992) Summary-A monodimensional thin layer chromatography method to separate several sugars of clinical interest is described. The separation and identification of 14 sugars (L-fucose, D-galactose, D-glucose, lactose, N-acetylglucosamine, D-maltose, D-manose, L-sorbase, fructose, D-xylosc, glucuronic acid, N-acetyllactosamine, 3’ and 6’ sialyllactose) and maltodextrines (GrGs) is possible by using two different eluents mixtures, as well as two different detection reagents. The method has been applied to separate sugars, maltodextrines and oligosaccharides in several biological fluids (blood, urine and faeces), in an infant milk and in human milk. It is a very simple technique (with a high sensitivity) that can be used in any lab.
The presence of anomalous amounts of carbohydrate8 can be detected in blood, urine and faeces of patients with inborn errors in the metabolism of carbohydrates, such as essential fructosuria and others.‘-’ In the present method, urine does not need to be desalted as in most cases, either by coltmnP or by precipitation.5.g Over the last few years, the increasing popularity of breastfeeding has stimulated interest in the sugar content of human milk. It is known that the carbohydrate concentration of human milk is about 7 g/100 ml; 80% lactose and the rest oligosaccharides (l-l.2 g/100 ml in nature milk, 2-2.5 g/100 ml in colostrum).‘4 It has not yet been possible to reproduce the structure of these oligosaccharides. They are substituted by maltodextrines in “adapted” milk formula, keeping the total sugar concentration at 7-8 g/100 rn.15 The biological role of human milk oligosaccharides is not completely understood, but it
*Author for correspondence.
seems clear that they have a very important anti-infective action for breast-fed infants.16 Kobata et al. isolated and characterized most of the oligosaccharides in human milk using very sophisticated technology.“8 At first they used pooled milk from several donors as starting material, but then observed that the type of oligosaccharides present in individual samples of milk can vary with the ABO or Lewis blood type of the donor. The enzymes involved in their synthesis are also responsible for the formation of the structural determinants of these blood types. All oligosaccharides in human milk have some general structural principles in common, they all derive from lactose and contain, besides galactose and glucose, N-acetylglucosamine and/or more of L-fucose and N-acetyl-neuraminic acid molecules. Another advantage of the method is that sugars may be identified without using internal standards. Thin layer chromatography has been applied specifically to the study of the role placed by the oligosaccharides obtained from human blood, urine and milk’“~“~‘3in metabolic pathways.
1493
F.
1494
BGSCH-REIG
Several sugars obtained from animal urines have been studied by thin layer chromatography and high pressure liquid chromatography.‘2 It can be seen in the references in the literature that certain sugars have been studied, however a complete study of separation, identification and application to biological fhtids of carbohydrates has not been done.
EXPERIMENTAL Samples and solvent systems
Aqueous solutions of the following sugars were prepared: L-fucose, D-galactose, D-glucose, lactose, Nacetylglucasamine, D-maltose, D -manose, L sorbose, D-xylose, fructose (0.7 g/100 ml), glucuronic acid (0.21 g/100 ml), N-acetyllactosamine (5.00 g/100 ml) 3’ and 6’ sialyllactose (1.70 g/100 ml) and maltodextrines (G&J (8.33 g/100 ml). All sugars were obtained from commercial sources (Merck, Sigma). Less concentrated solutions were obtained by diluting the original ones. Blood samples, as well as fresh and twentyfour hour urine samples were taken from the ones available in the laboratory. Faeces samples belonged to preterm infants. The milk samples used were from either one donor, 15 days and 45 days after the delivery, or several donors, 15 days after the delivery. An infant milk formula containing lactose and maltodextrines (56.9 g/100 g milk), 15% in water. Blood was delipidated using a chloroform: metanol 2 : 1 mixture in a ratio 1: 10, mixed for 5 min and centrifuged for 25 min at 5000 rpm. Then it was ready to be applied to the plate. Urine required no previous treatment and could be directly applied to the plate. Faeces were analysed after being diluted in distilled water 1:2 and centrifuged 25 min at 5000 rpm. The supernatant was applied to the plate. Milk samples were delipidated by centrifugation at 1500 rpm for 30 min and diluted 1: 2 with distilled water. One-millilitre aliquots were kept at -20” until they were analysed. Two elements mixtures were used: in-butanol/ ethanol/water (3 : 2 : 1) and pyridin/ethyl acetate/ acetic acid/water (5 : 5 : 3 : 1).
et al.
Chromatography
Cellulose plates (20 x 20 cm, O.l-mm thick) were used (Merck Art. 5716). Cellulose plates (20 x 20 cm, 0.5~mm thick) were used for preparative chromatography (Merck Art.). Aliquots (3 pl), were spotted on the 0.1~mm plates with a micro syringe, at 1.5~cm intervals and 2 cm from the lower edge of the plates. Then they were developed by the ascending technique. A three-fold development, with the same eluent mixture, is used in order to improve separation. A good separation was obtained when the eluent was allowed to run 14.5 cm from the edge of the plate. The time required for a single run is approximately two and a half hours. In the preparative chromatography milk samples aliquots (850 ~1) were applied as a streak on the 0.5-mm plate. A three-fold development was also used. Guide strips were developed with silver nitrate reagent to locate the bands. The required bands on the unsprayed area were scrapped off onto test tubes, 1 ml of distilled water was added to each and tubes were shaken thoroughly. Sugar was separated from cellulose powder by centrifugation at 4.500 rpm for 5 min. Detection reagents
The following reagents were used: silver nitrate solution prepared by dissolving 3 g of silver nitrate in 12 ml of distilled water adding 500 ml of acetone. Ethanolic sodium hydroxide solution prepared by diluting 50 ml of 10N sodium hydroxide in 450 ml of ethanol. Sodium thiosulphate (5%) was also used. The developed plates were allowed to dry at room temperature and dipped into AgNO solution for one minute. When dry they were sprayed with the ethanolic sodium hydroxide solution until dark brown spots appeared on a light brown background. If the plates must be kept for a long time, they must be sprayed with a 5% NaZS203 solution once they are dry. Elson -Morgan reagent
1% Acetillacetone in butanol is treated with a l/20 volume of 50% potassium hydroxideethanol (1: 4 v/v). p-Dimethylaminobenzaldehide solution: 1 g of p-dimethylbenaldehide is dissolved in 30 ml
Separation and identification of sugars and maltodextrines
1495
Table 2. R, values, using two different eluents mixtures, of maltodextrines in aqueous solutions
of ethanol of concentrated HCl and 30 ml of N-butanol. When the plate was dry, it was sprayed with the acetylacetone solution, heated at 100” for 5 min and then sprayed with pdimenthylbenzaldehide solution. N-Acetylaminated sugars appear immediately as violet spots.
Sugar GZ G3
G4 G, G, G, G*
R, (eluent 1)
R, (eluent 2)
0.73 0.65 0.56 0.46 0.38 0.31 0.24
0.48 0.35 0.26 0.19 0.14 0.10 -
RESULTS AND DISCUSSION
Aqueous solutions of the following sugars were chromatographed on 0. I-mm cellulose plates: N-acetylglucosamine, fucose, xylose, manose, sorbose, glucose, galactose, N-acetyllactosamine, glucuronic acid, maltose, lactose, 3’-sialyllactose and 6’-sialyllactose, using Nbutanol/ethanol/water (3 : 2 : 2) as eluent and with a 3-fold development. All sugars were detected with both silver nitrate and ElsonMorgan reagents. The RIvalues and sensitivities are shown in Table 1. Similarly, an aqueous solution of maltodextrines was also chromatographed using two different eluents: (1) n-butanol/ethanol/water (3 : 2 : 2) and (2) pyridine/ethyl acetate/acetic acid/water (5 : 5 : 1: 3). Then maltodextrines were located with silver nitrate reagent. The method sensitivity for maltodextrines is 6.25 pg and the Rf values are shown in Table 2. As can be seen in Table 1 fucose and Nacetylglucosamine have the same RI but they could be identified by the detection reagents. Both of them can be visualized with silver nitrate reagent, but only N-acetylglucosamine can be detected with Elson-Morgan reagent. Table 2 shows that only G2 to G, components of maltodextrines can be separated with eluent (2). G2 to G8 can be separated with eluent (1).
However G9 and G10 components cannot be separated either with eluent (1) or with eluent (2). It can be said that the described aqueous solutions of sugars and maltodextrines have been identified either by their Rfor by selecting the suitable detection reagent. When sugars were added to blood, fresh urine, twenty-four hour urine and faeces, these biological fluids were chromatographed and it was seen that there is very good agreement between the Rf of the standard solutions and those of the biological fluids, as shown in Table 3. Samples of infant milk were also chromatographed. Its sugar content according to the manufacturer was lactose and maltodextrines. N-Butanol/ethanol/water (3 : 2 : 2) was used as eluent and the detection reagent was silver nitrate. On the chromatogram shown in Fig. 1 can be seen that lactose and maltodextrines were identified, as well as minor sugars like galactose and glucose. Therefore, this procedure allows the identification of sugars with a sensitivity that may be considered high. Seven spots were detected when samples of milk from a donor who had delivered a full term infant 15 days before, were chromatographed.
Table 1. RFvalues and sensitivity in pg of sugars in aqueous solutions
Table 3. R, values of sugars in several biological fluids (blood, faeces and urine) Sugar
Sugar 6’-Sialillactose 3’-Sialillactose Lactose Maltose Glucuronic Acid N-Acetillactosamine Galactose Glucose Sorbose Manose Xilose Function Fucose N-Acetilglucosamine
TM 3Q:ll-6
R,
0.35 0.42 0.48 0.48 0.55 0.58 0.65 0.69 0.70 0.73 0.78 0.74 0.81 0.8 1
(AgI’f6,
3.19 3.19 0.56 0.21 0.09 3.0 0.14 0.11 0.21 0.21 0.11 0.11 0.11 0.42
R.) @son-%rganR.) 3.0 2.1
R, values
Xilose Fructose Glucose Galactose Glucuronic Acid Lactose
Aqueous solution 0.78 0.14 0.69 0.65 0.55 0.48
Sugar Xilose Fructose Glucose Galactose Glucuronic Acid Lactose
Fresh urine 0.78 0.74 0.69 0.66 0.55 0.48
Blood Faeces 0.78 0.78 0.73 0.74 0.69 0.69 0.65 0.65 0.53 0.54 0.47 0.48 Twenty-four hour urine 0.78 0.74 0.69 0.66 0.55 0.48
F.
1496
~SCH-&lG
et al.
2
I
3
4
5
Fig. 1. Chromatogram of sugars of an infant milk; 1: infant milk :water (1: 1); 2: glucose: 14 mg/lOO ml; 3: galactose: 14 mg/lOO ml; 4: lactose: 70 mg/lOO ml; 5: maltodextrines: 830 mg/lOO ml.
As shown in Fig. 2, spots 1, 2, 3 and 4 were confirmed to be glucose, lactose, 3’-sialyllactose and 6’-sialyllactose respectively. Spots 5, 6 and 7 were confirmed to be fractions of oligosaccharides. They were isolated by preparative chromatography and they all released glucose, galactose and fucose when mild hydrolysis
I
2
3
was carried out. Spot 5 also contains N-acetyl glucosamine as it reacts with Elson-Morgan reagent. Samples of milk of the same donor 45 days after she gave birth were also chromatographed (Fig. 3). The chromatogram was quite similar to the former. However, the fractions of oligo-
4
5
6
Fig. 2. Chromatogram of human milk; 1: human milk: water (1: 1); 2: lactose: 70 mg/lOO ml; 3: 3’ and 6’ sialyllactose: 210 mg/lOO ml; 4: human milk: 3’ and 6’ sialyllactose (1: 1); 5: glucose: 14 mg/lOO ml; 6: human milk: glucose (1: 1)
Separation and identification of sugars and maltodextrines
2
I
3
4
5
6
1497
7
Fig. 3. Chromatogram of human milk from the same donor 1, 3, 5 and 7: sample of milk after 45 days of delivery. 2, 4 and 6: sample of milk after 15 days of delivery.
saccharides were in less concentration and subsequently the glucose concentration was increased. When samples of milk from different donors, who had delivered babies 15 days before, were chromatographed, strong differences were observed in the chromatogram as a function of the donors (Fig. 4).
Therefore, it can be said that the chromatogram of sugars from milk depends fundamentally on the donor, the maturity of the milk and its origin (human and infant milk formula). All of that can be detected by thin layer chromatography which is a very simple technique that can be used in any laboratory.
i I
Fig. 4. Chromatogram
2
3
4
5
6
7
6
9
IO
of sugars of milks from several donors I-10 spots belong to milks from different donors.
F. BOSCH-REIGet al.
1498 REFERENCES
1. I. Lytt, Gardner. Salvat., Enferrmpdades congknitus y endocrinas de la infmcia.
2. Lynch, Raphael, Mellor, Spare, Inwood, Laboratory Methods, 2nd ed. 3. M. Ghebregzabher, S. Rufmi, B. Monaldi and M. Lato, J. Chromatogr., 1976, 127, 133. 4. Vladimir Vitek and Kveta Vitek, ibid., 1971, 60, 381. 5. Idem, ibid., 1977, 143, 65. 6. Gerard Streaker and Anick Lemaire-Poitau, ibid., 1977, 143, 65. 7. M. Lato, B. Brunelli and G. CiulTini and T. Mezzetti, ibid., 1968, 36, 191. 8. Wolfgang Prinz, William Meldrum and Lynne Wilkinson, Clin. Chim. Acta, 1978, 82, 229. 9. Zeijko Zilic and Nenad Blau and Margarethe Knob, J. Chomalogr., 1979, 164, 91. 10. R. Ramphal, C. Carnoy, S. Fieure, J. C. Michalski, N. Hondret, G. Lamblin, G. Strecker and P. Roussel, Inlect. Iunmnn., 1991, 59, 700. 11. P. Scudder, A. M. Lawson, E. F. Hounsell, R. A. Carruthem, R. A. Childs and T. Feiz, Eur. J. Biochem., 1978, 163, 585.
12. C. D. Warren, S. Sadeh, P. F. Daniel, B. Bugge, L. F. James and R. W. Jeanloz, Febs Lert., 1983, 163, 99. 13. A. Cahour, P. Debeire, L. Hartmann and J. Montrenil, Biochem. J., 1983, 211, 55. 14. Comite sobre nutricibn ESPAGN (1982). Acta Paediatr. Stand. Suplemento 302. 15. Comite sobre nutrici6n ESPAGN (1977). Acta Paediatr. Stand. Suplemento 262. 16. J. Cruz and C. Arevalo, Pedriutric Infections Disease 5, S-148. 17. A. Kobata, E. Grollman and V. Ginsburg, Biochem. Biophys. Res. Commun., 1968, 32, 272. 18. A. Kobata and V. Ginsburg, Arch. Biochem. Biophys., 1969, 130, 509. 19. I&m, J. Biol. Chem., 1969, 244, 5496. 20. Idem, ibid., 1970, 245, 1984. 21. I&m, ibid., 1972, 247, 1525. 22. Idem, Arch. Biochem. Biophys., 1972, 150, 273. 23. K. Yamashita and A. Kobata, ibid., 1974, 161, 164. 24. K. Yamashita, Y. Tachibana and A. Kobata, ibid., 174, 582. 25. A. Kobata, Merho& in Enzymology, 1972, 28, 262. 26. A. Kobata, K. Yamashita and Y. Tachibana, Methods in Enzymology, 1978, 50, 216.