- Email: [email protected]
Use of activated carbon for removing phenylalanine from reconstituted skim milk powder hydrolysates
ARTICLE IN PRESS
LWT 38 (2005) 447–453 www.elsevier.com/locate/lwt
Use of activated carbon for removing phenylalanine from reconstituted skim milk powder hydrolysates Daniella Cristine Fialho Lopes, Fernanda Meneghello Delvivo, Marialice Pinto Coelho Silvestre Departamento de Alimentos, Faculdade de Farma´cia, UFMG, Av. Antoˆnio Carlos 6627 sala 3070-B3, Cep 31270-901 Belo Horizonte, MG, Brazil Received 1 April 2004; received in revised form 5 July 2004; accepted 8 July 2004
Abstract With the aim of preparing dietary supplements for phenylketonurics, the activated carbon was used in this work to remove phenylalanine (Phe) from skim milk powder enzymatic hydrolysates. Six hydrolysates were prepared, using a protease from Aspergillus oryzae (AO), isolated or in association with papain (PA). Different conditions were tested for removing Phe, and the best one showed to be the use of a activated carbon:casein ratio of 118 (g) and a temperature of 25 1C, which produced 96–99% of Phe removal. Among the hydrolytic conditions employed, the association of AO with PA (1 h, 1 g of enzyme/100 g of substrate and 4 h, 2 g of enzyme/100 g of substrate, respectively) led to the lowest absolute value for the final Phe concentration (0.060 104 mg/ 100 mg of protein). r 2004 Swiss Society of Food Science and Technology. Published by Elsevier Ltd. All rights reserved. Keywords: Dietary supplement; Phenylketonuria; Phenylalanine; Skim milk powder; Activated carbon
1. Introduction Phenylketonuria (PKU) is a metabolic disease associated with the metabolism disorder of phenylalanine (Phe), in which the oxidation of this amino acid is impaired due to the deficiency of the hydroxylase Phe enzyme (Moszczynski & Idziak, 1993). Untreated patients show serious mental retardation and their expectation of life is drastically reduced (Lopez-Bajonero, Lara-Calderon, Galvez-Mariscal, Velasquez-Arellano, & Lopez-Munguia, 1991; Shimamura et al., 1999; Mira & Marquez, 2000). The nutritional therapy for PKU is based on limitation of protein ingestion, reducing Phe supply to the minimum and promoting the normal growth of
Corresponding author. Tel.: +55-3134996919; fax: +553134996928. E-mail address: [email protected] (M.P.C. Silvestre).
patients with other nutrients (Mahan & Stump, 1998; Dutra-De-Oliveira, 1998, Chap. 3). Considering that the amount of Phe in vegetable or animal proteins is from 3–6g/100 g of protein, the diet for phenylketonurics is generally deficient in proteins in order to attempt the low Phe level required by these patients (Outinen, Tossavainen, Harju, & Linko, 1996; Shimamura et al., 1999). The PKU diet is normally supplemented with a medicinal food containing a mixture of free amino acids or oligopeptides, which provides 50–90% of protein equivalents, 90–100% of vitamins and trace elements and 50–70% of energy (Mira & Marquez, 2000). Among several protein sources that may be used for preparing dietary supplements for phenylketonurics, isolated casein, the main milk protein, is the choice in most cases (Lopez-Bajonero et al., 1991; Outinen et al., 1996; Shimamura et al., 1999). However, in underdeveloped countries, this protein needs to be imported which represents an important increase in production
0023-6438/$30.00 r 2004 Swiss Society of Food Science and Technology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.lwt.2004.07.021
ARTICLE IN PRESS 448
D.C.F. Lopes et al. / LWT 38 (2005) 447–453
costs. Thus, the use of alternative sources must be investigated. Among them, the skim milk powder may be tested since its cost is lower than casein. Most of the methods used for Phe removal from protein hydrolysates are based on the principle that a sufficient amount of Phe is liberated by enzymatic hydrolysis, and the free Phe is, then, removed by gel filtration, adsorption by activated carbon or resins (Lopez-Bajonero et al., 1991; Outinen et al., 1996; Shimamura et al., 1999). The second derivative spectrophotometry (SDS) is an analytical technique of great utility for obtaining qualitative and quantitative data about several compounds (O’haver, 1979; Grant & Bhattacharyya, 1985; Rojas, Ojeda, & Pavon, 1998), such as aromatic amino acids, which absorb specifically between 250 and 300 nm (Ichikawa & Terada, 1977, 1981; Cahill & Padera, 1980; Miclo, Perrin, Driou, Mellet, & Linden, 1995). Concerning the complex spectra of proteins, SDS showed to be a useful tool for separating, identifying and quantifying these amino acids, as well as to reveal important differences among native and denaturated proteins (Ragone, Colonna, Balestrieri, Servillo, & Irace, 1984). Several authors reported the great reliability of using SDS, between 250 and 270 nm, for quantifying Phe in proteins, since parameters such as pH and the addition of other elements are controlled (Brandts & Kaplan, 1973; O’haver, 1979; Matsushima, Inoue, & Shibata, 1975; Ichikawa & Terada, 1979; Cahill & Padera, 1980; Grant & Bhattacharyya, 1985; Rojas et al., 1998). Concerning the study of protein hydrolysates, SDS was used by our group for determining the hydrolytic degree of casein hydrolysates, as well as for evaluating the modifications in the protein chain, around aromatic residues, which normally take place when the native structure is broken up. Moreover, we found that the peak intensity, in the second derivative spectra of proteins or peptides is related to the degree of exposure of aromatic amino acids, which becomes higher as the group nears the C- or N-terminal position (Silvestre, Dauphin, & Hamon 1993). In our group’s more recent work, a study of second derivative spectra of aromatic amino acids was carried out in two pH values (7.0 and 13.0), and SDS was used for estimating the degree of Phe exposure in casein hydrolysates prepared by using papain (Barbosa et al., 2002). The aim of the present work was to test several conditions using activated carbon for removing Phe from protein hydrolysates, in order to prepare dietary supplements for phenylketonurics. The protein source used here was skimed milk powder. SDS was used as a screening method for choosing the best condition for removing Phe by activated carbon, and also for estimating the efficiency of this treatment.
2. Material and methods 2.1. Material L-pheylalanine, L-tyrosine, L-tryptophan and a protease from Aspergillus oryzae (XXIII type) were purchased from Sigma Chemical Co. (St. Louis, MO, EUA). Papain was kindly furnished by BIOBRA´S (Montes Claros, MG, Brazil). The skim milk powder (SMP) was purchased in a supermarket of Belo Horizonte, MG, Brazil.
2.2. Methods 2.2.1. Preparation of skim milk powder (SMP) hydrolysates Six hydrolysates were prepared from solutions of SMP (0.35 g/100 ml) in 0.01 mol/l phosphate buffer (pH 6.0). Initially, they were pre-heated in a water-bath, at 80 1C for 10 min. Then, the temperature was adjusted to 50 1C, and the enzymes, a protease of Aspergillus Oryzae (AO) isolated or in association with papain (PA), were added in such a concentration to attain the desired enzime:substrate ratio (Table 1). The hydrolytic reactions were stopped by lowering the temperature to 10 1C in an ice bath and the hydrolysates were, then, freezedried. The other parameters of hydrolysis are listed in Table 1. 2.2.2. Characterization of phenylalanine by SDS Stock solutions of Phe (6.05 104 mol/l), Tyr (5.52 104 mol/l) and Trp (4.90 104 mol/l) were prepared in a 0.01 mol/l phosphate buffer, pH 6.0. Then, 10 ml of each solution were mixed and successive dilutions of this mixture were made to have Phe concentrations in a range from 0.13 to 1.01 104 mol/l. Spectra of these diluted solutions were recorded from 250 to 280 nm (CECIL spectrophotometer, CE2041 model, Buck Scientific, England). A software GRAMS-UV (Galactic Industries Corporation,
Table 1 Hydrolytic conditions employed for preparing skim milk powder hydrolysates Hydrolysates
H1 H2 H3 H4 H5 H6
Hydrolysis time (h)
AO AO AO AO AO AO
(5 h) (1 h)+PA (4 h) (1 h)+PA (4 h) (2 h)+PA (8 h) (5 h)+PA (20 h) (15 h)+PA (10 h)
E:S (g/100 g) AO
PA
1 1 10 1 1 1
— 2 20 2 2 2
E:S=enzyme:subsgtrate ratio; AO=protease from Aspergillus oryzae; PA=papain.
ARTICLE IN PRESS D.C.F. Lopes et al. / LWT 38 (2005) 447–453
Salem, NH, EUA) was used to draw the second derivative spectra. Then, eight standard curves were drawn, two for each negative peak (‘‘a’’, ‘‘b’’, ‘‘c’’ and ‘‘d’’) (Fig. 2), using their area or height, in function of Phe concentration.
2.2.3. Use of activated carbon to remove Phe from SMP hydrolysates 2.2.3.1. Determining the ideal conditions. Eight solutions of hydrolysate H5 (same hydrolytic conditions stated by Lopez-Bajonero et al., 1991) in purified water (Aries, Vaponics, EUA), were prepared with a concentration of 0.8 g/100 ml. Then, different amounts of activated carbon were added to each solution (Table 2), and the mixtures were stirred just to keep the carbon in suspension (Frisatom stirrer, Sa˜o Paulo, SP, Brazil). Also, the effects of stirring time (10 and 30 min) and of temperature (20 and 30 1C) were investigated (Table 2). The mixtures were centrifuged (Jouan centrifuge, BR4i model, France) at 10,000 rpm for 10 min at 25 1C, and filtered (qualitative filter paper, Whatman no. 1, Maidstone, England). The absorbance of the filtrates were measured from 250 to 280 nm, the second derivative spectra were drawn and the areas or heights of negative peaks were used to calculate the exposure rate (ER) of Phe, employing the standard curve of Phe. Considering that no acid hydrolysis was performed, these results correspond to Phe that was exposed by the enzymatic hydrolysis (Barbosa et al., 2002). The best condition for Phe removal was the one that produced the highest percentage of ER, calculated according to ERð%Þ ¼ 100 ½ðB=AÞ 100;
(1)
where A=ER of Phe in the protein hydrolysate, before treatment with activated carbon, and B=ER of Phe in the protein hydrolysate, after treatment with activated carbon.
Table 2 Tested conditions for removing Phe from skim milk powder hydrolysates Filtrates
Added amount of activated carbon (g)
Stirring time (min)
Temperature (1C)
F1 F2 F3 F4 F5 F6 F7 F8
0.20 0.20 0.20 0.40 0.80 0.80 0.80 1
10 20 30 30 30 30 30 30
20 20 20 20 20 25 30 25
F=filtrates from hydrolysate H5.
449
2.2.3.2. Evaluating the efficiency of Phe removal. Initially, the best condition to remove Phe (Section 2.2.3.1) was applied to all hydrolysates. Then, a 10 ml aliquot of filtrates (Section 2.2.3.1) was evaporated in Centrivap (Labconco, USA) and hydrolysed (5.7 mol/l HCl, 110 1C, 24 h). The residue was dissolved in 10 ml of purified water, the pH adjusted to 6.0 with 1 mol/l dibasic sodium phosphate solution and the absorbance was measured from 250 to 280 nm. Second derivative spectra were drawn and the areas or heights of negative peaks were used to calculate the amount of Phe in SMP hydrolysates, employing the standard curve. The same process was applied to SMP, and the efficiency of Phe removal was calculated according to Phe Removalð%Þ initial amount of Phe-final amount of Phe ¼ 100; ð2Þ initial amount of Phe where initial amount of Phe=amount of Phe in SMP; final amount of Phe=amount of Phe in hydrolysates treated by activated carbon. 2.2.4. Statistical analysis All experiments were carried out in triplicate. Differences between means of areas or heights were evaluated by analysis of variance (ANOVA) and Duncan test (Gomes, 1990). Differences were considered to be significant at Po0:05 throughout this study. The least-squares method was used to fit the standard curve and the adequacy of the linear model (Y ¼ aX þ b) was tested at Po0:05:
3. Results and discussion 3.1. SDS spectrum of Phe The absorbance and SDS spectra of Phe, in a mixture of aromatic amino acids, and of hydrolysate H2, in pH 6.0, are shown in Fig. 1. In case of Phe (Fig. 1b), we can see four negative peaks, indicated by letters ‘‘a’’, ‘‘b’’, ‘‘c’’ and ‘‘d’’, situated within the range from 280 nm with maxima at 253, 258, 263, 268 and 273 nm and minima at 257, 262, 267 and 272 nm. The SDS spectrum of H2 is close to that of Phe, with negative peaks situated in almost the same wavelengths. The similarity among the spectra of standard amino acids and proteins had previously been described by Ichikawa and Terada (1977), working with several native and denaturated proteins. The same result was previously achieved in our laboratory using papain for hydrolysing casein (Barbosa et al., 2002). On the other hand, in some reports in the literature, the number of negative peaks for Phe was different from that found in the present work. Thus, five peaks for this amino acid in pH 7.0 were shown by Ichikawa and
ARTICLE IN PRESS D.C.F. Lopes et al. / LWT 38 (2005) 447–453
450
50E-07
0.14 Absorbance
Absorbance
0.16
0.12 0.10 0.08 0.06
a 0
b c d
-50E-07
0.04 255
(a)
260
265
270
255
275
260
(b)
Wavelength (nm)
265
270
275
Wavelength (nm)
0.7 40E-06 Absorbance
Absorbance
a 0.6 0.5
20E-06 b 0
c d
0.4 40E-06 255
(c)
260
265
270
275
280
Wavelength (nm)
255
(d)
260
265
270
275
Wavelength (nm)
Fig. 1. Absorbance and second derivative spectra of Phe in a solution containing Tyr and Trp (a and b, respectively) and of hydrolysate H2 (c and d, respectively), in pH 6.0.
Terada (1977), while Miclo et al. (1995) described the presence of six, in pH 1.9. These discrepancies could be associated to several factors such as forms of Phe (free or eather), type of equipment (spectrophotometer and software used for measuring absorption and derivative spectra), properties of the solvent and pH used (Ichikawa & Terada, 1977; Levillain & Fompeydie, 1986).
F5 treatment produced the same ER of F8, it is somehow disadvantageous because a further stage was needed to cool the sample to 20 1C using an ice bath. It is worth stating that the measure of ER by SDS showed to be a useful screening method for choosing the best condition for removing Phe, avoiding acid hydrolysis of samples, which is a time-consuming stage.
3.2. Linearity of standard curve of Phe
3.3.2. Efficiency of Phe removal As shown in Table 4, the activated carbon, in the conditions determined above (Section 2.2.3.1) was efficient to remove Phe from all SMP hydrolysates, and the reduction changed from 96% to 99%. Also, the data in Table 4 indicate that the isolated action of a protease from Aspergillus oryzae (H1) gave similar results to its association with papain (H2, H3, H4, H5, H6), as well as the fact that the total reaction time had no influence on the Phe removal, since no significant difference was observed among hydrolysates prepared with 5 h (H1 and H2), 10 h (H4) and 25 h (H5 and H6). Only the increase of enzyme:substrate ratio of the two enzymes reduced the Phe removal (H3). A probable explanation for this could be associated with the higher amount of Phe inside the large peptide chains in the hydrolysate H3, which would hinder its removal. In fact, as previously reported by Shimamura et al. (1999) stated before, only a small amount of Phe can be removed by gel filtration or activated carbon, when this amino acid is insufficiently liberated by enzymatic hydrolysis.
The standard curve of Phe which presented the best correlation coefficient (the highest and most significant) was that using the area of negative peak ‘‘d’’ (Fig. 2). The correlation coefficient and the curve equation were y ¼ 2:1748x þ 0:3913 and R2 ¼ 0:9921: This result is in agreement with others in the literature (Ichikawa & Terada, 1977; Zhao, Sannier, Garreau, Lecoeur, & Piot, 1996) and also with a previous study carried out in our laboratory (Barbosa et al., 2002), since in all these works a linearity for the standard curve of Phe, in presence of Tyr and Trp in several concentrations, was shown. 3.3. Phe removal from SMP hydrolysates 3.3.1. Ideal condition for removing Phe Among all tested conditions for removing Phe from SMP hydrolysates, the best one used 118 g of activated carbon per g of casein, stirring for 30 min at 25 1C (F8), since it led to the highest ER of Phe (Table 3). Although
ARTICLE IN PRESS D.C.F. Lopes et al. / LWT 38 (2005) 447–453
451
3
Peak area "d" (x10-5)
2.5 2 1.5 y = 2.1748x + 0.3913
1
R2 = 0.9921 0.5 0 0
0.2
0.4
0.6
0.8
1
1.2
-4
Phe content (x10 )
Fig. 2. Standard curve of Phe. Peak ‘‘d’’=negative peak in the second derivative spectrum of Phe.
Table 3 Exposition rate of Phe in supernatants, after treatment with activated carbon
Table 4 Efficiency of Phe removal from skim milk powder hydrolysates by activated carbon
Filtrates
ER
Hydrolysates
Removal percentage
F1 F2 F3 F4 F5 F6 F7 F8
55d73.61 45e71.91 50e79.25 82c72.48 98a71.58 91b75.13 93b73.20 98a70.36
Final Phe contenta (mg 104/100 mg protein)
H1 H2 H3 H4 H5 H6
0.098b,c70.022 0.060c70.032 0.231a70.078 0.072b,c70.005 0.104b70.001 0.133a,b70.070
98b70 99a,b71 96c71 99a70 98b70 98b71
F=filtrates of hydrolysates; ER=exposition rate of Phe (%), calculated according to Eq. (1). Different letters are significantly different (Po0:05).
Concerning the hydrolysate H5, it is worth stating that it was prepared using the same hydrolytic conditions described by Lopez-Bajonero et al. (1991). However, instead of removing only 92% of Phe reported by these authors, in the present work, we were able to remove 98% of this amino acid. This discrepancy may be related to the higher amount of activated carbon used here (38.5 times). Nevertheless, no mention was made by these authors regarding the conditions of activated carbon use, such as the length and the velocity of stirring, as well as the temperature of the protein solution, which could also interfere in the Phe removal. Moreover, although these authors stated that they used skim milk powder and casein as substrate, the results presented in their paper involved only casein. Although we used a larger amount of activated carbon than Lopez-Bajonero et al. (1991), the advantage of this work is related to the lower hydrolysis length (5 times) and the use of only one enzyme, which are important factors to reduce costs and increase purity of these formulations. It is well known that using a reaction time longer than 5 h to hydrolyse proteins,
Different letters are significantly different (Po0:05). a Final Phe content=Phe content after treatment with activated carbon, calculated according to Eq. (2).
increases the probability of microbial contamination (Chataud, Desreumeux, & Cartwright, 1988; Loosen, Bresspollier, Juleen, Pejoan, & Verneuil, 1991; Moszczynski & Idziak, 1993; Shimamura et al., 1999). Other authors also used activated carbon to remove Phe from protein hydrolysates. Thus, Kitagawa et al. (1987), after hydrolysing whey proteins with actinase, in pH 6.5 at 37 1C, treated these preparations with activated carbon and removed 97% of Phe. However, the conditions for the treatment with activated carbon were not mentioned. Moszczynski & Idziak (1993) removed 95% of Phe from casein hydrolysates. However, these authors employed more severe conditions than those used here, i.e. a very long time for hydrolysis (72 h) and for the treatment with activated carbon (5.5 h). Other methods were employed by some authors, in order to remove Phe from protein hydrolysates, such as gel filtration (Shimamura et al., 1999). However, the removal efficiency was not mentioned by these authors. The use of polystyrene resins (XAD-4 e XAD-16) was reported by Outinen, et al. (1996), leading to an almost
ARTICLE IN PRESS 452
D.C.F. Lopes et al. / LWT 38 (2005) 447–453
complete removal of Phe (99.9%). However, this method has the disadvantage related to the much higher cost of these resins compared to activated carbon (Moszczynski & Idziak, 1993; Lopez-Bajonero et al., 1991; Shimamura et al., 1999). It is important to mention that, concerning the recommendation of Phe for phenyketonurics, although a consensus was not found in the literature, the tendency is to consider a blood concentration between 2 and 6 mg/dL (Wappner, Cho, Kronmal, Schuett, & Seashore, 1999), and the amount of Phe in the diet must change according to this blood serum level (Acosta, 1997). In this way, the final concentration of Phe found in protein hydrolysates used for phenyketonurics treatment must be informed in order to adjust the level of this amino acid to the requirements of each patient. Thus, although no significant difference was observed among almost all Phe contents of SMP hydrolysates in the present work, if the clinical aspect of phenylketonuria is considered, one can conclude that hydrolysate H2 would be the best choice for preparing dietary supplements for the treatment of this disease, considering its lower absolute value found for Phe content (0.060 104 mg/100 mg of protein) (Table 4).
4. Conclusion The use of activated carbon showed to be an efficient method for removing Phe from hydrolysates giving rise to skim milk hydrolysate powder useful as an ingredient for phenyketonurics diets. This adsorption material was able to remove up to 99% of Phe from skim milk hydrolysates, using AO associated with PA. The second derivative spectrophotometry (SDS) showed to be a simple and fast method, either for choosing the best condition for Phe removal or for the quantification of this amino acid in SMP and in its hydrolysates.
Acknowledgements The authors thank CNPq and FAPEMIG for their financial support. References Acosta, P. B., & Yannicelli, S. (1997). Phenylketonuria., In The ross metabolic formula system—nutricion support protocol. Ross Products Division, Abbott Laboratories, Keziaz Strvat, Columbia. Barbosa, C. M. S., Morais, H. A., Silva, V. D. M., Oliveira, M. C., & Silvestre, M. P. C. (2002). Padronizac- a˜o de me´todo analı´ tico para avaliac- a˜o do grau de exposic- a˜o da fenilalanina em hidrolisados de caseı´ na, por espectrofotometria derivada Segunda. Revista Brasileira de Cieˆncias Farmaceˆuticas, 38, 113–119. Brandts, J. F., & Kaplan, L. J. (1973). Derivate spectroscopy applied to tyrosiyl chromophores. Studies on ribonuclease, lima bean
inhibitors, insulin, and pancreatic trypsin inhibitor. Biochemical Journal, 12, 121–127. Cahill, J. E., & Padera, F.G. (1980). Derivate analysis of uv/visible spectra. American Laboratory (pp. 101–112). Chataud, J., Desreumeux, S., & Cartwright, T. (1988). Proce´de´ de fabrication d’un hydrolysat enzymatique de prote´ines riche en di- et tri-peptides, utilisable notamment en nutrition artificielle et en diete´tique. Laborato´rio Roger Bellon, Neuilly sur-Seine-FR. A23J3/00, FR87402837.6, 0.274946A1. Dutra-De-Oliveira, J. E. (1998). Cieˆncias nutricionais, Sa˜o Paulo: Sarvier. Grant, A., & Bhattacharyya, P. K. (1985). Application of derivate spectroscopy to the determination of chromatografhic peak purity. Journal of Chromatography, 347, 219–235. Gomes, F. P. (1990). Curso de estatı´stica experimental, Piracicaba: Nobel. Ichikawa, T., & Terada, H. (1977). Second derivative spectrophotometry as an effective tool for examining phenylalanine residues in proteins. Biochimica and Biophysica Acta, 494, 267–270. Ichikawa, T., & Terada, H. (1979). Estimation of state and amount of phenylalanine residues in proteins by second derivative spectrophotometry. Biochimica and Biophysica Acta, 580, 120–128. Ichikawa, T., & Terada, H. (1981). Determination of phenylalanine, tryptophan and tyrosine in a mixture of amino acids by second derivative spectrophotometry. Chemical Pharmacy, 29, 438–444. Kitagawa, T., Owada, M., Aoki, K., Arai, S., Oura, T., Matsuda, I., Igarashi, Y., Tada, K., Katayama, S., & Hashida, W. (1987). Treatment of phenylketonuria with a formula consisting of lowphenylalanine peptide. Enzyme, 38, 321–327. Levillain, P., & Fompeydie, D. (1986). Spectrophotome´trie de´rive´e: inte´reˆt, limites et apllications. Analysis, 14, 1–20. Loosen, P. C., Bresspollier, P. R., Julieen, A. R., Pejoan, C. H. & Verneuil, B. (1991). Procede pour preparer um hydrolysat enzymatique Tessenderlo Cheemie n. v. [BE/BE]; Stationsstraat, B-3980 Tessenderlo (BE). A23J3/34, C12P21/06 C12S3/14, C07K15/00//A61K37/18, A23J3/04, 3/14. FR-PCT/BE91/00001, W091/10369. 11/01/1991; 25/07/1991. Lopez-Bajonero, L. J., Lara-Calderon, P., Galvez-Mariscal, A., Velasquez-Arellano, A., & Lopez-Munguia, A. (1991). Enzimatic production of a low-phenylalanine product from skim milk powder and caseinate. Journal of Food Science, 56, 938–942. Mahan, L. K., & Stump, S. E. (1998). Krause—Alimentos, nutric- a˜o e dietoterapia, Sa˜o Paulo: Roca. Matsushima, A., Inoue, Y., & Shibata, K. (1975). Derivate absortion spectrophotometry of native proteins. Analytical Biochemistry, 65, 362–368. Miclo, L., Perrin, E., Driou, A., Mellet, M., & Linden, G. (1995). Determination of the ratios of the aromatic amino acids residues by first- or second-derivative UV spectrometry for a simple characterization of peptides. International Journal of Peptides and Protein Research, 46, 186–192. Mira, N. V. M., & Marquez, U. M. L. (2000). Importaˆncia do diagno´stico e tratamento da fenilcetonu´ria. Revista Sau´de Pu´blica, 34, 86–96. Moszczynski, P., & Idziac, J. (1993). Preparation of enzymatic hydrolysates of casein depleted in phenyllalanine. Biochemistry and Microbiology, 29, 302–306. O’haver, T. C. (1979). Potential clinical applications of derivative and wavelength-modulation spectrometry. Clinical Chemistry, 25, 1548–1559. Outinen, M. T., Tossavainen, O., Harju, M., & Linko, P. (1996). Method for removing phenylalanine from proteinaceous compositions, a product obtained and use thereof. Patents US 5547687.
ARTICLE IN PRESS D.C.F. Lopes et al. / LWT 38 (2005) 447–453 Ragone, R., Colonna, G., Balestrieri, C., Servillo, L., & Irace, G. (1984). Determination of tyrosine exposure in proteins by second derivative spectroscopy. Biochemical Journal, 23, 1871–1875. Rojas, F. S. ., Ojeda, C. B., & Pavon, J. M. C. (1998). Derivate ultraviolet-visible region absorption spectrophotometry and its analytical applications. Talanta, 35, 753–761. Silvestre, M. P. C., Dauphin, C., & Hamon, M. (1993). Application of UV absorption and second-derivate spectrophotometry for analysing casein hydrolysates. Analytical Chemistry Acta, 282, 603–612. Shimamura, S., Tamura, Y., Miyakawa, H., Saito, H., Kawaguchi, Y., Isomura, N., Akazome, Y., Ochi, H., & Kawamoto, M.
453
(1999). Peptide mixture and products thereof. Patents US 5952193. Wappner, R., Cho, S., Kronmal, R. A., Schuett, V., & Seashore, M. R. (1999). Management of phenylketonuria for optimal outcome: a review of guidelines for phenylketonuria management and a report of surveys of parents, patients, and clinic directors. Pediatric, 104, 4–9. Zhao, Q., Sannier, F., Garreau, I., Lecoeur, C., & Piot, J. M. (1996). Reversed-phase high-performance liquid chromatography coupled with second-order derivative spectroscopy for the quantitation of aromatic amino acids in peptides: application to hemorphins. Journal of Chromatography, 723, 35–41.
LWT 38 (2005) 447–453 www.elsevier.com/locate/lwt
Use of activated carbon for removing phenylalanine from reconstituted skim milk powder hydrolysates Daniella Cristine Fialho Lopes, Fernanda Meneghello Delvivo, Marialice Pinto Coelho Silvestre Departamento de Alimentos, Faculdade de Farma´cia, UFMG, Av. Antoˆnio Carlos 6627 sala 3070-B3, Cep 31270-901 Belo Horizonte, MG, Brazil Received 1 April 2004; received in revised form 5 July 2004; accepted 8 July 2004
Abstract With the aim of preparing dietary supplements for phenylketonurics, the activated carbon was used in this work to remove phenylalanine (Phe) from skim milk powder enzymatic hydrolysates. Six hydrolysates were prepared, using a protease from Aspergillus oryzae (AO), isolated or in association with papain (PA). Different conditions were tested for removing Phe, and the best one showed to be the use of a activated carbon:casein ratio of 118 (g) and a temperature of 25 1C, which produced 96–99% of Phe removal. Among the hydrolytic conditions employed, the association of AO with PA (1 h, 1 g of enzyme/100 g of substrate and 4 h, 2 g of enzyme/100 g of substrate, respectively) led to the lowest absolute value for the final Phe concentration (0.060 104 mg/ 100 mg of protein). r 2004 Swiss Society of Food Science and Technology. Published by Elsevier Ltd. All rights reserved. Keywords: Dietary supplement; Phenylketonuria; Phenylalanine; Skim milk powder; Activated carbon
1. Introduction Phenylketonuria (PKU) is a metabolic disease associated with the metabolism disorder of phenylalanine (Phe), in which the oxidation of this amino acid is impaired due to the deficiency of the hydroxylase Phe enzyme (Moszczynski & Idziak, 1993). Untreated patients show serious mental retardation and their expectation of life is drastically reduced (Lopez-Bajonero, Lara-Calderon, Galvez-Mariscal, Velasquez-Arellano, & Lopez-Munguia, 1991; Shimamura et al., 1999; Mira & Marquez, 2000). The nutritional therapy for PKU is based on limitation of protein ingestion, reducing Phe supply to the minimum and promoting the normal growth of
Corresponding author. Tel.: +55-3134996919; fax: +553134996928. E-mail address: [email protected] (M.P.C. Silvestre).
patients with other nutrients (Mahan & Stump, 1998; Dutra-De-Oliveira, 1998, Chap. 3). Considering that the amount of Phe in vegetable or animal proteins is from 3–6g/100 g of protein, the diet for phenylketonurics is generally deficient in proteins in order to attempt the low Phe level required by these patients (Outinen, Tossavainen, Harju, & Linko, 1996; Shimamura et al., 1999). The PKU diet is normally supplemented with a medicinal food containing a mixture of free amino acids or oligopeptides, which provides 50–90% of protein equivalents, 90–100% of vitamins and trace elements and 50–70% of energy (Mira & Marquez, 2000). Among several protein sources that may be used for preparing dietary supplements for phenylketonurics, isolated casein, the main milk protein, is the choice in most cases (Lopez-Bajonero et al., 1991; Outinen et al., 1996; Shimamura et al., 1999). However, in underdeveloped countries, this protein needs to be imported which represents an important increase in production
0023-6438/$30.00 r 2004 Swiss Society of Food Science and Technology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.lwt.2004.07.021
ARTICLE IN PRESS 448
D.C.F. Lopes et al. / LWT 38 (2005) 447–453
costs. Thus, the use of alternative sources must be investigated. Among them, the skim milk powder may be tested since its cost is lower than casein. Most of the methods used for Phe removal from protein hydrolysates are based on the principle that a sufficient amount of Phe is liberated by enzymatic hydrolysis, and the free Phe is, then, removed by gel filtration, adsorption by activated carbon or resins (Lopez-Bajonero et al., 1991; Outinen et al., 1996; Shimamura et al., 1999). The second derivative spectrophotometry (SDS) is an analytical technique of great utility for obtaining qualitative and quantitative data about several compounds (O’haver, 1979; Grant & Bhattacharyya, 1985; Rojas, Ojeda, & Pavon, 1998), such as aromatic amino acids, which absorb specifically between 250 and 300 nm (Ichikawa & Terada, 1977, 1981; Cahill & Padera, 1980; Miclo, Perrin, Driou, Mellet, & Linden, 1995). Concerning the complex spectra of proteins, SDS showed to be a useful tool for separating, identifying and quantifying these amino acids, as well as to reveal important differences among native and denaturated proteins (Ragone, Colonna, Balestrieri, Servillo, & Irace, 1984). Several authors reported the great reliability of using SDS, between 250 and 270 nm, for quantifying Phe in proteins, since parameters such as pH and the addition of other elements are controlled (Brandts & Kaplan, 1973; O’haver, 1979; Matsushima, Inoue, & Shibata, 1975; Ichikawa & Terada, 1979; Cahill & Padera, 1980; Grant & Bhattacharyya, 1985; Rojas et al., 1998). Concerning the study of protein hydrolysates, SDS was used by our group for determining the hydrolytic degree of casein hydrolysates, as well as for evaluating the modifications in the protein chain, around aromatic residues, which normally take place when the native structure is broken up. Moreover, we found that the peak intensity, in the second derivative spectra of proteins or peptides is related to the degree of exposure of aromatic amino acids, which becomes higher as the group nears the C- or N-terminal position (Silvestre, Dauphin, & Hamon 1993). In our group’s more recent work, a study of second derivative spectra of aromatic amino acids was carried out in two pH values (7.0 and 13.0), and SDS was used for estimating the degree of Phe exposure in casein hydrolysates prepared by using papain (Barbosa et al., 2002). The aim of the present work was to test several conditions using activated carbon for removing Phe from protein hydrolysates, in order to prepare dietary supplements for phenylketonurics. The protein source used here was skimed milk powder. SDS was used as a screening method for choosing the best condition for removing Phe by activated carbon, and also for estimating the efficiency of this treatment.
2. Material and methods 2.1. Material L-pheylalanine, L-tyrosine, L-tryptophan and a protease from Aspergillus oryzae (XXIII type) were purchased from Sigma Chemical Co. (St. Louis, MO, EUA). Papain was kindly furnished by BIOBRA´S (Montes Claros, MG, Brazil). The skim milk powder (SMP) was purchased in a supermarket of Belo Horizonte, MG, Brazil.
2.2. Methods 2.2.1. Preparation of skim milk powder (SMP) hydrolysates Six hydrolysates were prepared from solutions of SMP (0.35 g/100 ml) in 0.01 mol/l phosphate buffer (pH 6.0). Initially, they were pre-heated in a water-bath, at 80 1C for 10 min. Then, the temperature was adjusted to 50 1C, and the enzymes, a protease of Aspergillus Oryzae (AO) isolated or in association with papain (PA), were added in such a concentration to attain the desired enzime:substrate ratio (Table 1). The hydrolytic reactions were stopped by lowering the temperature to 10 1C in an ice bath and the hydrolysates were, then, freezedried. The other parameters of hydrolysis are listed in Table 1. 2.2.2. Characterization of phenylalanine by SDS Stock solutions of Phe (6.05 104 mol/l), Tyr (5.52 104 mol/l) and Trp (4.90 104 mol/l) were prepared in a 0.01 mol/l phosphate buffer, pH 6.0. Then, 10 ml of each solution were mixed and successive dilutions of this mixture were made to have Phe concentrations in a range from 0.13 to 1.01 104 mol/l. Spectra of these diluted solutions were recorded from 250 to 280 nm (CECIL spectrophotometer, CE2041 model, Buck Scientific, England). A software GRAMS-UV (Galactic Industries Corporation,
Table 1 Hydrolytic conditions employed for preparing skim milk powder hydrolysates Hydrolysates
H1 H2 H3 H4 H5 H6
Hydrolysis time (h)
AO AO AO AO AO AO
(5 h) (1 h)+PA (4 h) (1 h)+PA (4 h) (2 h)+PA (8 h) (5 h)+PA (20 h) (15 h)+PA (10 h)
E:S (g/100 g) AO
PA
1 1 10 1 1 1
— 2 20 2 2 2
E:S=enzyme:subsgtrate ratio; AO=protease from Aspergillus oryzae; PA=papain.
ARTICLE IN PRESS D.C.F. Lopes et al. / LWT 38 (2005) 447–453
Salem, NH, EUA) was used to draw the second derivative spectra. Then, eight standard curves were drawn, two for each negative peak (‘‘a’’, ‘‘b’’, ‘‘c’’ and ‘‘d’’) (Fig. 2), using their area or height, in function of Phe concentration.
2.2.3. Use of activated carbon to remove Phe from SMP hydrolysates 2.2.3.1. Determining the ideal conditions. Eight solutions of hydrolysate H5 (same hydrolytic conditions stated by Lopez-Bajonero et al., 1991) in purified water (Aries, Vaponics, EUA), were prepared with a concentration of 0.8 g/100 ml. Then, different amounts of activated carbon were added to each solution (Table 2), and the mixtures were stirred just to keep the carbon in suspension (Frisatom stirrer, Sa˜o Paulo, SP, Brazil). Also, the effects of stirring time (10 and 30 min) and of temperature (20 and 30 1C) were investigated (Table 2). The mixtures were centrifuged (Jouan centrifuge, BR4i model, France) at 10,000 rpm for 10 min at 25 1C, and filtered (qualitative filter paper, Whatman no. 1, Maidstone, England). The absorbance of the filtrates were measured from 250 to 280 nm, the second derivative spectra were drawn and the areas or heights of negative peaks were used to calculate the exposure rate (ER) of Phe, employing the standard curve of Phe. Considering that no acid hydrolysis was performed, these results correspond to Phe that was exposed by the enzymatic hydrolysis (Barbosa et al., 2002). The best condition for Phe removal was the one that produced the highest percentage of ER, calculated according to ERð%Þ ¼ 100 ½ðB=AÞ 100;
(1)
where A=ER of Phe in the protein hydrolysate, before treatment with activated carbon, and B=ER of Phe in the protein hydrolysate, after treatment with activated carbon.
Table 2 Tested conditions for removing Phe from skim milk powder hydrolysates Filtrates
Added amount of activated carbon (g)
Stirring time (min)
Temperature (1C)
F1 F2 F3 F4 F5 F6 F7 F8
0.20 0.20 0.20 0.40 0.80 0.80 0.80 1
10 20 30 30 30 30 30 30
20 20 20 20 20 25 30 25
F=filtrates from hydrolysate H5.
449
2.2.3.2. Evaluating the efficiency of Phe removal. Initially, the best condition to remove Phe (Section 2.2.3.1) was applied to all hydrolysates. Then, a 10 ml aliquot of filtrates (Section 2.2.3.1) was evaporated in Centrivap (Labconco, USA) and hydrolysed (5.7 mol/l HCl, 110 1C, 24 h). The residue was dissolved in 10 ml of purified water, the pH adjusted to 6.0 with 1 mol/l dibasic sodium phosphate solution and the absorbance was measured from 250 to 280 nm. Second derivative spectra were drawn and the areas or heights of negative peaks were used to calculate the amount of Phe in SMP hydrolysates, employing the standard curve. The same process was applied to SMP, and the efficiency of Phe removal was calculated according to Phe Removalð%Þ initial amount of Phe-final amount of Phe ¼ 100; ð2Þ initial amount of Phe where initial amount of Phe=amount of Phe in SMP; final amount of Phe=amount of Phe in hydrolysates treated by activated carbon. 2.2.4. Statistical analysis All experiments were carried out in triplicate. Differences between means of areas or heights were evaluated by analysis of variance (ANOVA) and Duncan test (Gomes, 1990). Differences were considered to be significant at Po0:05 throughout this study. The least-squares method was used to fit the standard curve and the adequacy of the linear model (Y ¼ aX þ b) was tested at Po0:05:
3. Results and discussion 3.1. SDS spectrum of Phe The absorbance and SDS spectra of Phe, in a mixture of aromatic amino acids, and of hydrolysate H2, in pH 6.0, are shown in Fig. 1. In case of Phe (Fig. 1b), we can see four negative peaks, indicated by letters ‘‘a’’, ‘‘b’’, ‘‘c’’ and ‘‘d’’, situated within the range from 280 nm with maxima at 253, 258, 263, 268 and 273 nm and minima at 257, 262, 267 and 272 nm. The SDS spectrum of H2 is close to that of Phe, with negative peaks situated in almost the same wavelengths. The similarity among the spectra of standard amino acids and proteins had previously been described by Ichikawa and Terada (1977), working with several native and denaturated proteins. The same result was previously achieved in our laboratory using papain for hydrolysing casein (Barbosa et al., 2002). On the other hand, in some reports in the literature, the number of negative peaks for Phe was different from that found in the present work. Thus, five peaks for this amino acid in pH 7.0 were shown by Ichikawa and
ARTICLE IN PRESS D.C.F. Lopes et al. / LWT 38 (2005) 447–453
450
50E-07
0.14 Absorbance
Absorbance
0.16
0.12 0.10 0.08 0.06
a 0
b c d
-50E-07
0.04 255
(a)
260
265
270
255
275
260
(b)
Wavelength (nm)
265
270
275
Wavelength (nm)
0.7 40E-06 Absorbance
Absorbance
a 0.6 0.5
20E-06 b 0
c d
0.4 40E-06 255
(c)
260
265
270
275
280
Wavelength (nm)
255
(d)
260
265
270
275
Wavelength (nm)
Fig. 1. Absorbance and second derivative spectra of Phe in a solution containing Tyr and Trp (a and b, respectively) and of hydrolysate H2 (c and d, respectively), in pH 6.0.
Terada (1977), while Miclo et al. (1995) described the presence of six, in pH 1.9. These discrepancies could be associated to several factors such as forms of Phe (free or eather), type of equipment (spectrophotometer and software used for measuring absorption and derivative spectra), properties of the solvent and pH used (Ichikawa & Terada, 1977; Levillain & Fompeydie, 1986).
F5 treatment produced the same ER of F8, it is somehow disadvantageous because a further stage was needed to cool the sample to 20 1C using an ice bath. It is worth stating that the measure of ER by SDS showed to be a useful screening method for choosing the best condition for removing Phe, avoiding acid hydrolysis of samples, which is a time-consuming stage.
3.2. Linearity of standard curve of Phe
3.3.2. Efficiency of Phe removal As shown in Table 4, the activated carbon, in the conditions determined above (Section 2.2.3.1) was efficient to remove Phe from all SMP hydrolysates, and the reduction changed from 96% to 99%. Also, the data in Table 4 indicate that the isolated action of a protease from Aspergillus oryzae (H1) gave similar results to its association with papain (H2, H3, H4, H5, H6), as well as the fact that the total reaction time had no influence on the Phe removal, since no significant difference was observed among hydrolysates prepared with 5 h (H1 and H2), 10 h (H4) and 25 h (H5 and H6). Only the increase of enzyme:substrate ratio of the two enzymes reduced the Phe removal (H3). A probable explanation for this could be associated with the higher amount of Phe inside the large peptide chains in the hydrolysate H3, which would hinder its removal. In fact, as previously reported by Shimamura et al. (1999) stated before, only a small amount of Phe can be removed by gel filtration or activated carbon, when this amino acid is insufficiently liberated by enzymatic hydrolysis.
The standard curve of Phe which presented the best correlation coefficient (the highest and most significant) was that using the area of negative peak ‘‘d’’ (Fig. 2). The correlation coefficient and the curve equation were y ¼ 2:1748x þ 0:3913 and R2 ¼ 0:9921: This result is in agreement with others in the literature (Ichikawa & Terada, 1977; Zhao, Sannier, Garreau, Lecoeur, & Piot, 1996) and also with a previous study carried out in our laboratory (Barbosa et al., 2002), since in all these works a linearity for the standard curve of Phe, in presence of Tyr and Trp in several concentrations, was shown. 3.3. Phe removal from SMP hydrolysates 3.3.1. Ideal condition for removing Phe Among all tested conditions for removing Phe from SMP hydrolysates, the best one used 118 g of activated carbon per g of casein, stirring for 30 min at 25 1C (F8), since it led to the highest ER of Phe (Table 3). Although
ARTICLE IN PRESS D.C.F. Lopes et al. / LWT 38 (2005) 447–453
451
3
Peak area "d" (x10-5)
2.5 2 1.5 y = 2.1748x + 0.3913
1
R2 = 0.9921 0.5 0 0
0.2
0.4
0.6
0.8
1
1.2
-4
Phe content (x10 )
Fig. 2. Standard curve of Phe. Peak ‘‘d’’=negative peak in the second derivative spectrum of Phe.
Table 3 Exposition rate of Phe in supernatants, after treatment with activated carbon
Table 4 Efficiency of Phe removal from skim milk powder hydrolysates by activated carbon
Filtrates
ER
Hydrolysates
Removal percentage
F1 F2 F3 F4 F5 F6 F7 F8
55d73.61 45e71.91 50e79.25 82c72.48 98a71.58 91b75.13 93b73.20 98a70.36
Final Phe contenta (mg 104/100 mg protein)
H1 H2 H3 H4 H5 H6
0.098b,c70.022 0.060c70.032 0.231a70.078 0.072b,c70.005 0.104b70.001 0.133a,b70.070
98b70 99a,b71 96c71 99a70 98b70 98b71
F=filtrates of hydrolysates; ER=exposition rate of Phe (%), calculated according to Eq. (1). Different letters are significantly different (Po0:05).
Concerning the hydrolysate H5, it is worth stating that it was prepared using the same hydrolytic conditions described by Lopez-Bajonero et al. (1991). However, instead of removing only 92% of Phe reported by these authors, in the present work, we were able to remove 98% of this amino acid. This discrepancy may be related to the higher amount of activated carbon used here (38.5 times). Nevertheless, no mention was made by these authors regarding the conditions of activated carbon use, such as the length and the velocity of stirring, as well as the temperature of the protein solution, which could also interfere in the Phe removal. Moreover, although these authors stated that they used skim milk powder and casein as substrate, the results presented in their paper involved only casein. Although we used a larger amount of activated carbon than Lopez-Bajonero et al. (1991), the advantage of this work is related to the lower hydrolysis length (5 times) and the use of only one enzyme, which are important factors to reduce costs and increase purity of these formulations. It is well known that using a reaction time longer than 5 h to hydrolyse proteins,
Different letters are significantly different (Po0:05). a Final Phe content=Phe content after treatment with activated carbon, calculated according to Eq. (2).
increases the probability of microbial contamination (Chataud, Desreumeux, & Cartwright, 1988; Loosen, Bresspollier, Juleen, Pejoan, & Verneuil, 1991; Moszczynski & Idziak, 1993; Shimamura et al., 1999). Other authors also used activated carbon to remove Phe from protein hydrolysates. Thus, Kitagawa et al. (1987), after hydrolysing whey proteins with actinase, in pH 6.5 at 37 1C, treated these preparations with activated carbon and removed 97% of Phe. However, the conditions for the treatment with activated carbon were not mentioned. Moszczynski & Idziak (1993) removed 95% of Phe from casein hydrolysates. However, these authors employed more severe conditions than those used here, i.e. a very long time for hydrolysis (72 h) and for the treatment with activated carbon (5.5 h). Other methods were employed by some authors, in order to remove Phe from protein hydrolysates, such as gel filtration (Shimamura et al., 1999). However, the removal efficiency was not mentioned by these authors. The use of polystyrene resins (XAD-4 e XAD-16) was reported by Outinen, et al. (1996), leading to an almost
ARTICLE IN PRESS 452
D.C.F. Lopes et al. / LWT 38 (2005) 447–453
complete removal of Phe (99.9%). However, this method has the disadvantage related to the much higher cost of these resins compared to activated carbon (Moszczynski & Idziak, 1993; Lopez-Bajonero et al., 1991; Shimamura et al., 1999). It is important to mention that, concerning the recommendation of Phe for phenyketonurics, although a consensus was not found in the literature, the tendency is to consider a blood concentration between 2 and 6 mg/dL (Wappner, Cho, Kronmal, Schuett, & Seashore, 1999), and the amount of Phe in the diet must change according to this blood serum level (Acosta, 1997). In this way, the final concentration of Phe found in protein hydrolysates used for phenyketonurics treatment must be informed in order to adjust the level of this amino acid to the requirements of each patient. Thus, although no significant difference was observed among almost all Phe contents of SMP hydrolysates in the present work, if the clinical aspect of phenylketonuria is considered, one can conclude that hydrolysate H2 would be the best choice for preparing dietary supplements for the treatment of this disease, considering its lower absolute value found for Phe content (0.060 104 mg/100 mg of protein) (Table 4).
4. Conclusion The use of activated carbon showed to be an efficient method for removing Phe from hydrolysates giving rise to skim milk hydrolysate powder useful as an ingredient for phenyketonurics diets. This adsorption material was able to remove up to 99% of Phe from skim milk hydrolysates, using AO associated with PA. The second derivative spectrophotometry (SDS) showed to be a simple and fast method, either for choosing the best condition for Phe removal or for the quantification of this amino acid in SMP and in its hydrolysates.
Acknowledgements The authors thank CNPq and FAPEMIG for their financial support. References Acosta, P. B., & Yannicelli, S. (1997). Phenylketonuria., In The ross metabolic formula system—nutricion support protocol. Ross Products Division, Abbott Laboratories, Keziaz Strvat, Columbia. Barbosa, C. M. S., Morais, H. A., Silva, V. D. M., Oliveira, M. C., & Silvestre, M. P. C. (2002). Padronizac- a˜o de me´todo analı´ tico para avaliac- a˜o do grau de exposic- a˜o da fenilalanina em hidrolisados de caseı´ na, por espectrofotometria derivada Segunda. Revista Brasileira de Cieˆncias Farmaceˆuticas, 38, 113–119. Brandts, J. F., & Kaplan, L. J. (1973). Derivate spectroscopy applied to tyrosiyl chromophores. Studies on ribonuclease, lima bean
inhibitors, insulin, and pancreatic trypsin inhibitor. Biochemical Journal, 12, 121–127. Cahill, J. E., & Padera, F.G. (1980). Derivate analysis of uv/visible spectra. American Laboratory (pp. 101–112). Chataud, J., Desreumeux, S., & Cartwright, T. (1988). Proce´de´ de fabrication d’un hydrolysat enzymatique de prote´ines riche en di- et tri-peptides, utilisable notamment en nutrition artificielle et en diete´tique. Laborato´rio Roger Bellon, Neuilly sur-Seine-FR. A23J3/00, FR87402837.6, 0.274946A1. Dutra-De-Oliveira, J. E. (1998). Cieˆncias nutricionais, Sa˜o Paulo: Sarvier. Grant, A., & Bhattacharyya, P. K. (1985). Application of derivate spectroscopy to the determination of chromatografhic peak purity. Journal of Chromatography, 347, 219–235. Gomes, F. P. (1990). Curso de estatı´stica experimental, Piracicaba: Nobel. Ichikawa, T., & Terada, H. (1977). Second derivative spectrophotometry as an effective tool for examining phenylalanine residues in proteins. Biochimica and Biophysica Acta, 494, 267–270. Ichikawa, T., & Terada, H. (1979). Estimation of state and amount of phenylalanine residues in proteins by second derivative spectrophotometry. Biochimica and Biophysica Acta, 580, 120–128. Ichikawa, T., & Terada, H. (1981). Determination of phenylalanine, tryptophan and tyrosine in a mixture of amino acids by second derivative spectrophotometry. Chemical Pharmacy, 29, 438–444. Kitagawa, T., Owada, M., Aoki, K., Arai, S., Oura, T., Matsuda, I., Igarashi, Y., Tada, K., Katayama, S., & Hashida, W. (1987). Treatment of phenylketonuria with a formula consisting of lowphenylalanine peptide. Enzyme, 38, 321–327. Levillain, P., & Fompeydie, D. (1986). Spectrophotome´trie de´rive´e: inte´reˆt, limites et apllications. Analysis, 14, 1–20. Loosen, P. C., Bresspollier, P. R., Julieen, A. R., Pejoan, C. H. & Verneuil, B. (1991). Procede pour preparer um hydrolysat enzymatique Tessenderlo Cheemie n. v. [BE/BE]; Stationsstraat, B-3980 Tessenderlo (BE). A23J3/34, C12P21/06 C12S3/14, C07K15/00//A61K37/18, A23J3/04, 3/14. FR-PCT/BE91/00001, W091/10369. 11/01/1991; 25/07/1991. Lopez-Bajonero, L. J., Lara-Calderon, P., Galvez-Mariscal, A., Velasquez-Arellano, A., & Lopez-Munguia, A. (1991). Enzimatic production of a low-phenylalanine product from skim milk powder and caseinate. Journal of Food Science, 56, 938–942. Mahan, L. K., & Stump, S. E. (1998). Krause—Alimentos, nutric- a˜o e dietoterapia, Sa˜o Paulo: Roca. Matsushima, A., Inoue, Y., & Shibata, K. (1975). Derivate absortion spectrophotometry of native proteins. Analytical Biochemistry, 65, 362–368. Miclo, L., Perrin, E., Driou, A., Mellet, M., & Linden, G. (1995). Determination of the ratios of the aromatic amino acids residues by first- or second-derivative UV spectrometry for a simple characterization of peptides. International Journal of Peptides and Protein Research, 46, 186–192. Mira, N. V. M., & Marquez, U. M. L. (2000). Importaˆncia do diagno´stico e tratamento da fenilcetonu´ria. Revista Sau´de Pu´blica, 34, 86–96. Moszczynski, P., & Idziac, J. (1993). Preparation of enzymatic hydrolysates of casein depleted in phenyllalanine. Biochemistry and Microbiology, 29, 302–306. O’haver, T. C. (1979). Potential clinical applications of derivative and wavelength-modulation spectrometry. Clinical Chemistry, 25, 1548–1559. Outinen, M. T., Tossavainen, O., Harju, M., & Linko, P. (1996). Method for removing phenylalanine from proteinaceous compositions, a product obtained and use thereof. Patents US 5547687.
ARTICLE IN PRESS D.C.F. Lopes et al. / LWT 38 (2005) 447–453 Ragone, R., Colonna, G., Balestrieri, C., Servillo, L., & Irace, G. (1984). Determination of tyrosine exposure in proteins by second derivative spectroscopy. Biochemical Journal, 23, 1871–1875. Rojas, F. S. ., Ojeda, C. B., & Pavon, J. M. C. (1998). Derivate ultraviolet-visible region absorption spectrophotometry and its analytical applications. Talanta, 35, 753–761. Silvestre, M. P. C., Dauphin, C., & Hamon, M. (1993). Application of UV absorption and second-derivate spectrophotometry for analysing casein hydrolysates. Analytical Chemistry Acta, 282, 603–612. Shimamura, S., Tamura, Y., Miyakawa, H., Saito, H., Kawaguchi, Y., Isomura, N., Akazome, Y., Ochi, H., & Kawamoto, M.
453
(1999). Peptide mixture and products thereof. Patents US 5952193. Wappner, R., Cho, S., Kronmal, R. A., Schuett, V., & Seashore, M. R. (1999). Management of phenylketonuria for optimal outcome: a review of guidelines for phenylketonuria management and a report of surveys of parents, patients, and clinic directors. Pediatric, 104, 4–9. Zhao, Q., Sannier, F., Garreau, I., Lecoeur, C., & Piot, J. M. (1996). Reversed-phase high-performance liquid chromatography coupled with second-order derivative spectroscopy for the quantitation of aromatic amino acids in peptides: application to hemorphins. Journal of Chromatography, 723, 35–41.