J. Agrie. Food Chem. 1981, 29, 540-543
`540
`Effect of Some Metals on the Maillard Reaction of Ovalbumin
`
`Yasuko Kato, Kenji Watanabe,* and Yasushi Sato
`
`Studies were conducted to investigate the effect of metal ions such as Na+, Cu2+, and Fe3+ or Fe2+ on
`the Maillard reaction during storage (50 °C; 65% relative humidity) of freeze-dried ovalbumin-glucose
`mixtures and the changes of some ovalbumin properties ascribed to the reaction. Cupric ion and Fe3+
`accelerated the denaturation (the decrease in  -helix content) of ovalbumin by promoting the Maillard
`reaction which was ascertained from the browning color development, the increase and decrease in
`solubility, the variation in isoelectric point determined by isoelectric focusing, the destruction of lysine
`and arginine residues, and the remaining ratios of free amino groups. However, Na+ had no effect on
`the reaction. The browning reaction was promoted somewhat faster in the presence of Fe3+ than of
`Fe2+. The addition of increasing amounts of Cu2+ ion from 0 to 0.5 mg % resulted in the acceleration
`of the browning reaction and the denaturation of ovalbumin.
`
`The interaction of metallic ions with proteins has been
`studied extensively in its biochemical as well as physico-
`chemical aspects (Vallee and Wacker, 1970). As one of the
`subjects of the metal-protein interaction, the significance
`of metals for the Maillard reaction has also been recognized
`inhibition effect on the re-
`because of its acceleration or
`action between protein or amino acid and sugar in terms
`of the amount and type of metal (Ellis, 1959).
`In our earlier studies (Kato et al., 1978; Watanabe et al.,
`1980), the browning reaction in the dried egg white solid-
`glucose system was found to be promoted more remarkably
`than in the dried ovalbumin-glucose one. We also sug-
`gested that trace amounts of metal in egg white accelerated
`the rate of browning. The present study was designed to
`investigate the effect of metal ions such as Na+, Cu2+, and
`Fe3+ or Fe2+, which were generally present in egg white,
`on the rate of the Maillard reaction in the dried ov-
`albumin-glucose system and the change of some properties
`of ovalbumin ascribed to the reaction.
`MATERIALS AND METHODS
`Sample Preparation. The ovalbumin preparation was
`dissolved with water at the protein concentration of 1%
`and divided into two parts. One part was adjusted to pH
`10.0 and freeze-dried. This sample was termed OV, as
`described in a previous study by Watanabe et al. (1980).
`To the several divided solutions of the other part, some
`metal ions were added with D-glucose corresponding to
`30% of the dry weight of the protein. The rates of the
`metal ion additions were as follows: FeCl2, FeCl3, and their
`mixture in equal amounts at the rate of 5 mg % to com-
`pare the effect of oxidative or reductive forms of iron, and
`NaCl at the rate of 0.2% and CuS04 and Fe(N03)3 at the
`rate of 5 mg % to evaluate the effect of the type of metal.
`These solutions were adjusted to pH 10.0 and then
`freeze-dried. Samples containing FeCl2, FeCl3, FeCl2 +
`FeCl3, NaCl, CuS04, Fe(N03)3, and with no metal were
`designated OVG-Fe2+, OVG-Fe3+, OVG-(Fe2+ + Fe3+),
`OVG-Na, OVG-Cu, OVG-Fe, and OVG, respectively. The
`various samples containing Cu2+ at the rate of 0.001 mg
`% (OVG-Cu-I), 0.005 mg % (OVG-Cu-II), 0.05 mg %
`(OVG-Cu-  ), and 0.5 mg % (OVG-Cu-IV) for the OVG
`solutions were prepared by the same method as mentioned
`above. For the control, an OV-Cu system containing
`CuS04 at the rate of 0.5 mg % (OV-Cu-IV) was treated
`in the same manner.
`
`Faculty of Agriculture, Nagoya University, Chikusa-ku,
`Nagoya 464, Japan (K.W. and Y.S.), and Women’s College
`of Tokaigakuen, Tenpaku-ku, Nagoya 468, Japan (Y.K.).
`0021-8561/81/1429-0540S01.25/0
`
`These samples were stored at 50 °C and 65% relative
`humidity as described previously (Kato et al., 1978).
`Measurement of Brown Color, Solubility, Amino
`Acid Composition, and  -Helix Content. All methods
`used for the measurements of brown color, solubility, am-
`ino acid composition and  -helix content were the same
`as reported in a previous study by Watanabe et al. (1980).
`Determination of Free Amino Groups. The content
`of free amino groups was determined by the method of
`fluorometric assay (Bohlen et al., 1973), using a Hitachi
`fluorescence spectrophotometer, MPF-2A, with the exci-
`tation wavelength at 390 nm and emission at 475 nm. The
`remaining ratios of free amino groups were expressed as
`percentages in relation to the content of free amino groups
`in native ovalbumin.
`Isoelectric Focusing. The isoelectric points of the
`unstored and stored samples with or without metal ions
`were determined by isoelectric focusing in 5% polyacryl-
`amide gel, using the technique based on the method of
`Catsimpoolas (1968). Each sample was dissolved in pH
`7.0 phosphate buffer (/ = 0.01) at the rate of 5% and
`focused in polyacrylamide gel containing 2% ampholine
`(pH range of ampholine: unstored sample, pH 3.5-10.0;
`stored sample, pH 3.0-5.0) for 4 h at 200 V. Each point
`on the pH gradient represents the average pH value ob-
`tained when duplicate unstained gels were cut into 0.5-cm
`slices and eluted with distilled water for one night. Pro-
`fixed with 12.5% trichloroacetic acid and
`teins were
`stained with Coomassie brilliant blue G-250. A Shimadzu
`dual-wavelength scanner, Model CS-910, was used to ob-
`tained the absorbance migration tracings.
`RESULTS
`Effect of the Added Ferrous and Ferric Ions. The
`browning color development on the addition of Fe2+ and
`Fe3+ ions is shown in Figure 1. From the comparison of
`four lines in Figure 1, it was found that the browning was
`positively catalyzed by iron ions and that the Fe3+ ion
`accelerated the color development reaction somewhat
`faster than that observed in the presence of the Fe2+ ion.
`Effect of the Added Three Metallic Ions. The shape
`of the browning curve on the addition of three metallic ions
`is shown in Figure 2. The browning color in the OVG-Cu
`system developed from its initial storage time and showed
`~4 times higher optical density than the others after 2
`days of storage. That in the OVG-Fe system developed
`gradually after storage for 2 days. The lines on the OVG
`and OVG-Na systems were the same and almost linear
`with storage time.
`Figure 3 shows the changes in solubility which occurred
`in the stored samples. Solubilities of OVG and OVG-Na
`© 1981 American Chemical Society
`
`Downloaded via 47.150.137.144 on September 12, 2023 at 05:55:14 (UTC).
`
`See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
`
`Ex. 1184
`
`

`

`Maillard Reaction of Ovalbumin
`
`Figure 1. Effect of Fe2+ and Fe3+ additions on the browning color
`(·) OVG; (O)
`development in ovalbumin-glucose mixtures.
`OVG-Fe2+; (O) OVG-Fe3+; ( ) OVG-(Fe2+ + Fe3+).
`
`J. Agrie. Food Chem., Vol. 29, No. 3, 1981
`Table I. Effect of Na*, Cu2+, and Fe3> Additions on the
`Remaining Ratio of Basic Amino Acid Residues in
`Ovalbumin-Glucose Mixtures
`remaining ratio of
`basic amino acid residues
`OVG-
`OVG-
`OVG-
`Na
`Cu
`Fe
`100
`100
`100
`100
`100
`100
`59
`57
`58
`58
`56
`56
`60
`58
`59
`56
`59
`51
`
`541
`
`days OVG
`100
`3
`100
`6
`59
`3
`58
`6
`60
`3
`59
`6
`
`,
`
`.
`
`amino
`acid
`His
`
`Lys
`
`Arg
`
`Figure 2. Effect of Na+, Cu2*, and Fe3+ additions on the browning
`color development in ovalbumin-glucose mixtures. (·) OVG; (X)
`OVG-Na; ( ) OVG-Cu; ( ) OVG-Fe.
`
`Figure 4. Effect of Na+, Cu2+, and Fe3+ additions on the change
`of the free amino group content in ovalbumin-glucose mixtures.
`(·) OVG; (X) OVG-Na; ( ) OVG-Cu; ( ) OVG-Fe.
`
`Figure 3. Effect of Na+, Cu2+, and Fe3+ additions on the change
`(·) OVG; (X)
`of solubility in ovalbumin-glucose mixtures.
`OVG-Na; ( ) OVG-Cu; ( ) OVG-Fe.
`during storage for the 5 days were much better than those
`of unstored samples, as noted in the OVG system in a
`previous study by Watanabe et al. (1980). Moreover, these
`for both samples showed almost the same pattern
`curves
`throughout the storage period. On the other hand, solu-
`bilities of OVG-Fe and OVG-Cu began to decrease ab-
`ruptly after 2 days of storage, followed by a more moderate
`decrease. As seen in Figure 3, OVG-Cu had the lowest
`solubility, which decreased to 25% within the first 4 days.
`The remaining ratios of basic amino acid during the
`storage obtained from the amino acid composition are
`shown in Table I. The destruction of lysine and arginine
`remarkable, and their remaining ratios
`residures was
`reached 60% in the first 3 days of storage, followed by only
`slight decreases during the 3 days to follow. However,
`there was no destruction of histidine residue in all samples
`used.
`The changes in free amino group content occurring
`during storage shown in Figure 4 indicated that the losses
`of free amino groups were similar in extent among all kinds
`
`Figure 5. Effect of Na+, Cu2+, and Fe3+ additions on the change
`in the isoelectric point in ovalbumin-glucose mixtures. Storage
`time: 0, without storage; 3, 3 days of storage. pH range of
`ampholine used: unstored sample, pH 3.5-10.0; stored sample,
`pH 3.0-5.0.
`of samples, reaching ~70% in the storage for the first day,
`followed by a slow increase. This was similar to the
`phenomenon encountered in the destruction of lysine
`(Table I).
`Each isoelectric point of various peaks separated by
`isoelectric focusing in polyacrylamide gel columns on the
`unstored and stored samples is shown in Figure 5.
`Isoe-
`lectric points on the unstored OVG-Cu and OVG-Fe
`showed only little differences for OVG and OVG-Na
`samples in the presence of compounds which have a more
`acidic isoelectric point than native ovalbumin. The major
`observable change with the samples stored for 3 days was
`the disappearance of the native ovalbumin and the oc-
`currence of various compounds having more acidic isoe-
`lectric points, in the order of OVG-Cu, OVG-Fe, OVG-Na,
`and OVG, based on the changes in net charge from the
`blocking of basic amino groups.
`
`Ex. 1184
`
`

`

`542
`
`J. Agrie. Food Chem., Vol. 29, No. 3, 1981
`
`Kato, Watanabe, and Sato
`
`Table II. Effect of Na% Cu2+, and Fe3 + Additions on the
`Loss of a-Helix Content in Ovalbumin-Glucose Mixtures
` -helix content, %,
`at storage time, days, of
`12
`6
`3
`29
`31
`27
`29
`28
`25
`14
`17
`21
`22
`26
`18
`
`20
`19
`14
`6
`12
`
`OVG
`OVG-Na
`OVG-Cu
`OVG-Fe
`
`0
`33
`33
`33
`33
`
`Figure 6. Effect of the concentration of added Cu2* on the change
`(·) OVG; (X)
`of solubility in ovalbumin-glucose mixtures.
`OVG-Cu-I; (a) OVG-Cu-II; ( ) OVG-Cu-III; ( ) OVG-Cu-IV;
`(O) OV-Cu-IV.
`The variation in the  -helix content of the ovalbumin
`calculated from the CD curves of the soluble part (pH 7.0
`phosphate buffer, I = 0.1) of unstored and stored samples
`is shown in Table II. Watanabe et al. (1980) reported that
`native ovalbumin contained an a helix of 33%. The «-
`helix contents of the four unstored kinds of samples were
`~33% and did not show changes in secondary form during
`their preparation. However, the  -helix content gradually
`decreased with storage time, showing that Cu2+ and Fe3+
`ions with the exception of Na+ accelerated the destruction
`of the ovalbumin secondary structure.
`Effect of Cupric Ion Concentrations. Solubilities of
`the unstored and stored samples containing CuS04 in
`various concentrations are shown in Figure 6.
`For the
`first 4 days, there were only marginal differences in the
`solubilities of all samples of OVG system, but thereafter
`their solubilities seriously decreased with the addition of
`In
`increasing amounts of Cu2+ ion from 0 to 0.5 mg %.
`contrast, the solubilities of the OV-Cu-IV containing no
`glucose showed a slight decrease during the storage time
`studied. From the fact that 0.023 mg % of copper ion is
`present in edible egg white liquid (Cotterill et al., 1977),
`it was expected that the browning curve of OVG containing
`the amount of copper ion present in egg white would be
`situated between the curve of OVG-Cu-I and that of
`OVG-Cu-II.
`The  -helix content of ovalbumin also decreased with
`the increase of Cu2+ ion concentration and the period of
`storage time, as shown in Table HI. However, there was
`no change in the  -helix content in the samples of OV-
`Cu-IV. This is in contrast to the decrease in the secondary
`structure for the OVG system.
`DISCUSSION
`The results reported here, combined with information
`from the literature, led to the conclusion that Cu2+ and
`Fe3+ accelerated the denaturation of ovalbumin by pro-
`moting the browning reaction of ovalbumin with glucose,
`but Na+ had no effect on it.
`Solutions of copper, iron, zinc, or cadmium normally
`precipitate metal hydroxide under the alkaline pH value.
`However, no precipitate appears if a complexing agent is
`
`Table III. Effect of the Concentration of Added Cu2 * on
`the Loss of a Helix in Ovalbumin-Glucose Mixtures0
`a-helix content, %,
`at storage time, days, of
`( 
`~0
`3
`OVG
`33
`29
`31
`OVG-Cu-I
`33
`31
`27
`OVG-Cu-II
`33
`29
`26
`OVG-Cu-III
`33
`28
`25
`OVG-Cu-IV
`28
`33
`22
`OV-Cu-IV
`33
`33
`33
`I, +0.001 mg % CuSO„; II, +0.005 mg % CuS04; III,
`+ 0.05 mg % CuS04; IV, +0.5 mg % CuS04.
`
`°
`
`present which removes
`free metal ions from the solution.
`If amino acids or peptides such as L-lysine, L-glycyl-L-
`leucine (Nakasuka et al., 1975), and L-histidine + L-
`threonine (Freeman and Martin, 1969) exist together,
`copper is soluble even in the alkaline solutions. Tanford
`(1952) investigated the interaction between bovine serum
`albumin and a number of metals (i.e., Cu2+, Zn2+, Cd2+,
`and Pb2+) and reported that the principal site on the
`protein molecule responsible for metal binding was the
`imidazole groups above pH 7. This might be proof that
`at pH 10 copper and iron had combined with ovalbumin
`under the existing conditions in our experiments.
`Certain metal ions promote Schiff base formation by
`forming stable complexes and thereby providing a more
`favorable free-energy reaction (Leussing and Bai, 1968).
`formation of N-
`However, a study of the rate of
`salicylideneglycinate has shown that Cu2+ and Ni2+ are
`kinetically inactive with the highly stable Schiff base
`complexes being formed through proton-catalyzed paths
`(Bai and Leussing, 1967).
`The isoelectric points of unstored samples did not show
`a great change, compared with the 25% loss of free amino
`group content in the freeze-drying process during sample
`preparation. This is natural from the fact that nonionized
`e-amino groups reacted with carbonyl groups in the first
`In the initial stage of the
`step of the Maillard reaction.
`reaction, comparing the variation of free amino groups and
`the remaining ratios of basic amino acids with the decrease
`of  -helix content among OVG, OVG-Na, OVG-Cu, and
`OVG-Fe, it could be said that amino groups on the surface
`of ovalbumin were at first blocked by the added glucose
`before denaturation in spite of the presence or absence of
`ions in the reaction mixture. The net negative
`metal
`change on the protein might increase with the progress of
`the Maillard reaction, because the decrease of the positive
`in terms of the irreversible transformation
`charge occurs
`of NH3+ to NH2, which can be blocked by a carbonyl
`group. The ways in which arginine residues participate
`in the reaction have not been demonstrated clearly thus
`far. The changes in the charge balance of the ovalbumin
`molecules might bring about a state in which the unfolding
`It has been demonstrated that
`of proteins is encouraged.
`maleylation of e-amino groups of ovalbumin caused ap-
`preciable but incomplete unfolding of ovalbumin (Qasin
`and Salahuddin, 1979). This incomplete unfolding, derived
`from the charge imbalance, might progress further with
`advances of the browning reaction which was positively
`catalyzed by transition elements such as Fe3+ and Cu2+ and
`led to the increase of denaturation and insolubility of
`ovalbumin. These findings might be corroborated by ob-
`servations revealing that traces of cupric salts had no effect
`on the rate of loss of amino groups of protein (Mohammad
`et al., 1949), metal ions catalyzed oxidative reaction of
`ascorbic acid (Taqui Khan and Martell, 1967) or catechols
`
`Ex. 1184
`
`

`

`J. Agrie. Food Chem. 1981, 29, 543-547
`543
`the studies on the rate of the overall browning reaction in
`(Tyson and Martell, 1972), and Fe3+ promoted the de-
`struction of Amadori rearrangement products in the oxy-
`the complex system such as egg white.
`gen-dependent browning system of a glucose-diglycine
`LITERATURE CITED
`mixture (Hashiba, 1975).
`Azari, P. R.; Feeney, R. E. Arch. Biochem. Biophys. 1961,92,44.
`The metal-binding proteins, such as conalbumin from
`Bai, K. S.; Leussing, D. L. J. Am. Chem. Soc. 1967, 89, 6126.
`egg white (Azari and Feeny, 1961) and transferrin from
`Bohart, G. S.; Carson, J. E. Nature (London) 1955, 175, 470.
`serum (Koechlin, 1952), have a greater stability to pro-
`Bóhlen, P.; Stein, S.; Dairman, W.; Udenfriend, S. Arch. Biochem.
`teolytic hydrolysis and thermal denaturation. In the stored
`Biophys. 1973,155, 213.
`ovalbumin-Cu2+ system (OV-Cu-IV), few effects on the
`Catsimpoolas, N. Anal. Biochem. 1968, 26, 480.
`denaturation and solubility of ovalbumin could be found
`Cotterill, O. J.; Marion, W. W.; Naber, E. C. Poult. Sci. 1977,56,
`even if metal bound to the ovalbumin in any form. Al-
`1927.
`though copper catalyzes oxidation steps involving cys-
`Ellis, G. P. Ado. Carbohydr. Chem. 1959,14, 91.
`Feeney, R. E.; MacDonnell, L. R.; Ducay, E. D. Arch. Biochem.
`tines-cysteines (Feeney et al., 1956), which might be in-
`volved in the aggregation phenomenon of denatured pro-
`Biophys. 1956, 61, 72.
`Freeman, H. C.; Martin, R. P. J. Biol. Chem. 1969, 244, 4823.
`tein, this action might be negligible in the OV-Cu-IV
`Hashiba, H. J. Agrie. Food Chem. 1975, 23, 539.
`system used.
`Kato, Y.; Watanabe, K.; Sato, Y. Agrie. Biol. Chem. 1978,42,2233.
`The difference in the capacity to accelerate the Maillard
`Koechlin, B. A. J. Am. Chem. Soc. 1952, 74, 2649.
`reaction between Fe2+ and Fe3+ reflects the possibility that
`Leussing, D. L.; Bai, K. S. Anal. Chem. 1968, 40, 575.
`the first step of metal catalysis is “oxidation activation”
`Mohammad, A; Fraenkel-Conrat, H.; Olcott, H. S. Arch. Biochem.
`involving the reduction of metal.
`1949, 24, 157.
`It has been clearly demonstrated that there are many
`Nakasuka, N.; Martin, R. P.; Scharff, J. P. Bull. Soc. Chim. Fr.
`species of minerals like calcium, copper, iron, magnesium,
`1975, 1973.
`Qasin, M. A.; Salahuddin, A. J. Biochem. (Tokyo) 1979,85,1029.
`manganese, potassium, sodium, and zinc in egg white
`Tanford, C. J. Am. Chem. Soc. 1952, 74, 211.
`In the present study, the effects
`(Cotterill et al., 1977).
`Taqui Khan,  . M.; Martell, A. E. J. Am. Chem. Soc. 1967,89,
`of only three kinds of metals on the browning reaction in
`7104.
`the ovalbumin-glucose mixture were studied. However,
`Tyson, C. A.; Martell, A. E. J. Am. Chem. Soc. 1972, 94, 939.
`it is considered that Cu2+ and Fe3+ might accelerate the
`Vallee, B. L; Wacker, W. E. C. Proteins (2nd Ed.) 1970, 5, 1.
`browning reaction in the egg white solid-glucose system
`Watanabe, K.; Kato, Y.; Sato, Y. J. Food Process. Presero. 1980,
`reported in the previous study (Kato et al., 1978). The
`3, 263.
`addition of a mere 0.003 ppm of manganese was found to
`inhibit the rate of browning reaction of the glucose-glycine
`system in air or oxygen (Bohart and Carson, 1955). At-
`tention must be paid to the existence of trace metals in
`
`Received for review January 22,1980. Revised September 29,
`1980. Accepted February 6,1981.
`
`Binding of Methylmercury to Ovalbumin as Methylmercuric Cysteine
`Welsonia J. Magat and Jerry L. Sell*
`
`After the administration of methylmercury in the form of CH3203HgCl to laying hens, the egg ovalbumin
`was isolated, and the binding of ^Hg was determined. Enzymic hydrolysis of ovalbumin, followed by
`covalent chromatography of the hydrolysate on 2-pyridyl-S-S-propyl-Sepharose, enabled the separation
`of a ^Hg-labeled digestion product. Amino acid analysis of the latter, after acid hydrolysis and performic
`acid oxidation, produced only cysteic acid. The hydrolyzed, but unoxidized, sample had an elution time
`different from that of cysteine or cystine. The ^Hg-labeled, Sepharose-bound fraction showed a higher
`Rf value than either cysteine or cystine but a value less than that of methionine. It had a mobility similar
`It was suggested that the fraction separated from
`to that of methylmercuric cysteine prepared in vitro.
`egg ovalbumin contained methylmercuric cysteine and that the binding of the ^Hg to the cysteine
`probably involved the SH group.
`
`The interactions of mercurials with proteins usually
`involve the SH and S-S groups, and such interactions have
`been linked to the toxic effects of mercury (Hg) (Vallee
`and Ulmer, 1972; Webb, 1966). Despite much research
`during the past 20 years on sulfur-mercury interactions,
`the relationship between this chemical reaction and tox-
`icological observations is yet to be explained.
`Many studies on Hg-binding sites of proteins have
`utilized albumins as model systems. Binding of the
`mercuric ion (Hg2+) to albumins has been demonstrated
`in vitro, and the reactive groups of the protein were the
`
`Department of Animal Science, Iowa State University,
`Ames, Iowa 50011.
`
`0021-8561/81/1429-0543301.25/0
`
`SH groups, as in mercaptalbumin (Hughes and Dintzis,
`1964), or the COOH group, as in human serum albumin
`(Perkins, 1961).
`The formation of methylmercuric cysteine in biological
`systems has been reported. A study of the accumulation
`of methylmercury has shown that ~95% of the muscle
`in the form of the methylmercury-
`methylmercury was
`cysteinyl complex (Westoo, 1966). After the injection of
`^Hg-labeled methylmercuric chloride (CH3203HgCl) in
`rats, methylmercury cysteine was found in the bile (Nor-
`seth and Clarkson, 1971).
`It has been reported that, when methylmercuric chloride
`was administered to hens, the ovalbumin contained 97%
`of the egg white ^Hg (Magat and Sell, 1979). The data
`presented in this paper deal with the chemical nature of
`© 1981 American Chemical Society
`
`Ex. 1184
`
`