pH on Structure and Function of Amaranth Protein Isolates

September 27/ September 2010 - October 2010Cereal Chemistry - pg. 448 Vol. 87 No. 5 ISSN: 0009-0352 -- Structural and functional properties of two amaranth protein isolates as a function of pH were studied. Isolates. A9 and All, were obtained by alkaline extraction at pH 9 and 11, respectively. Gel filtration chromatograms of A9 and A11 showed similar profiles. The A11 isolate contained mainly albumins and globulins, and a small proportion of globulin-P aggregates, suggesting the presence of species with a higher degree of denaturation compared to A9. Differential scanning calorimetry (DSC) showed that A9 was characterized by two thermal transitions (65.8 and 98°C); A11 exhibited only a small endotherm (66.6°C) and a second, less defined one.

DSC analysis of A9 at pH 2-4 did not show endotherms. but at pH 5, some protein structures were observed. A11 showed a greater degree of denaturation. FPLC results showed that the proteins in A9 are more folded and their conformation is closer to the native state than those in A11, which are more unfolded due to pH-mediated denaturation, mainly in acid media. The surface hydrophobicity of the isolates in acid media was lower than in alkaline media. The fluorescence emission spectra of the isolates showed differences in acidic pH conditions. As expected, the highest solubility was at alkaline pH. The water-holding capacity was similar for both isolates. The water-imbibing capacity and speed of foaming was higher for A11 than for A9. In summary, intense pH treatment of amaranth isolates generated partial or total protein denaturation and differences in the functional properties.

Use of vegetable seed protein preparations as ingredients in the food industry (Bernardino-Nicanor et al 2005; Adebowale et al 2007) requires knowledge of the protein physicochemical and functional properties. Amaranth is an excellent potential source of proteins because its seeds have high protein content (14-18%, versus ±10% of commercial cereals) with a balanced proportion of essential amino acids (Teutonico and Knor 1985; Bresani 1989). The nutritional quality of amaranth proteins resides not only in amino acid composition but also in digestibility, which is higher than that of cereals and close to that of casein (Bejosano and Corke 1998; Guzmán-Maldonado and Paredes-López 1999). The main protein fractions in amaranth grain are albumins, globulins, and glutelins, which differ in solubility. Amaranth globulins are composed of 11S-globulin, globulin-P, and a small amount of 7S globulin (Konishi et al 1991; Segura-Nieto et al 1994; Martínez et al 1997; Castellani et al 1998; Marcone 1999). Besides foods obtained from processing amaranth seeds, isolated amaranth proteins constitute a nutritional ingrethent for preparing traditional foods as well as for developing new ones. The quality of a protein isolate depends not only on its nutritional properties but also on the versatility of its functional properties, which are directly related to the structure of its constituent proteins and the conditions of their milieu. Different methods have been proposed for preparing amaranth isolates (Paredes-López et al 1998; Fidantsi and Doxastakis 2001; Cordero-de-los-Santos et al 2005) and some studies have been made regarding the functionality of the proteins. Acceptable emulsifying and foaming properties have been described for amaranth isolates and protein fractions (Konishi and Yoshimoto 1998; Marcone and Kakuda 1999; Fidantsi and Doxastakis 2001). Other studies have established the ability of protein isolates to coagulate upon heating and to form gels as well as determining their solubility in different media (Marcone and Kakuda 1999; Scilingo et al 2002; Avanza et al 2005). In view of the relationship between protein structure and functionality (Sorgentini et al 1991), structural modification through different treatments allows improvement of the functional properties and development of new ones. In particular, little is known about the capacity of these proteins for incorporating and retaining water (water-imbibition capacity and water-holding capacity). Also, there is no detailed information on the influence of ionic strength and pH on the structural and functional properties of amaranth protein isolates. The goal of the present work was to determine the physicochemical and functional properties of amaranth protein isolates as a function of pH.

Raw material: Seeds of Amaranthus hypochondiracus, Mercado, were obtained from the Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP), Chapingo, Mexico.

Flour preparation: Amaranth seed flour was obtained as in Abugoch et al (2003). Protein content was determined by the Kjeldahl technique (AOAC 1996) and calculated as N Ö 5.85 (Becker et al 1981) at 17.0 ± 0.1% w/w.

Preparation of protein isolates: The isolates were prepared by extraction of proteins at pH 9 and 11 and precipitation at pH 5. To this end, for each pH level, the flour was suspended in water (10% w/v), and the pH adjusted to the required value by adding 2M NaOH. Suspensions were stirred for 30 min at room temperature and then centrifuged at 9,000 Ö g for 20 min. Supernatants were adjusted to pH 5 with IM HCl, followed by centrifugation at 9,000 Ö g for 20 min at 4°C. The precipitates were resuspended in water, neutralized with 0.1M NaOH, and freeze-dried. Isolates obtained were termed A9 and A11, with the numbers referring to the original pH of extraction.

Solutions with a different pH: Gel-filtration chromatography (fast protein liquid chromatography, FPLC), differential scanning calorimetry (DSC), fluorescence spectrophotometry, hydrophobicity and solubility assays were conducted at different pH levels, including 0.17M C^sub 6^H^sub 8^O^sub 7^/0.03M C^sub 6^H^sub 7^O^sub 7^^sup -^ (pH 2.3); 0.08M C^sub 6^H^sub 8^O^sub 7^/0.12M C^sub 6^H^sub 7^O^sub 7^^sup -^ (pH 3.1); 0.015M C^sub 6^H^sub 8^O^sub 7^/0.15M C^sub 6^H^sub 7^O^sub 7^^sup -^/0.035M C^sub 6^H^sub 6^O^sub 7^^sup 2-^ (pH 4.1); 0.11M C^sub 6^H^sub 7^O^sub 7^^sup -^/0.09M C^sub 6^H^sub 6^O^sub 7^^sup 2-^ (pH 4.7); 0.06M C^sub 6^H^sub 7^O^sub 7^^sup -^/0.14M C^sub 6^H^sub 6^O^sub 7^^sup 2-^ (pH 5.1); 0.1 8M H^sub 2^PO^sub 4^^sup -^/0.02M HPO^sub 4^^sup 2-^ (pH 6.3); 0.1 2M H^sub 2^PO^sub 4^^sup -^/0.08M HPO^sub 4^^sup 2-^ (pH 7.5); 0.1 32M H^sub 3^BO^sub 3^/0.068/W H^sub 2^BO^sub 3^ (pH 8.8); 0.046M H^sub 3^BO^sub 3^/0.154M H^sub 2^BO^sub 3^^sup -^ (pH 9.7); 0.1 28M HCO^sub 3^^sup -^/0.072M CO^sub 3^^sup 2-^ (pH 10.1); 0.03M HCO^sub 3^^sup -^ /0.17/M CO^sub 3^^sup 2-^ (pH 11.0); 0.1 63M H^sub 2^BO^sub 3^^sup -^/0.037M HBO^sub 3^^sup 2-^ (pH 12.2). When necessary, NaCI was added as required to each solution to reach an ionic strength of 0.5.

Differential scanning calorimetry: DSC runs were performed in a calorimeter (Polymer Laboratories, Rheometric Scientific, UK) fitted with Plus V5.41 software. Calibration was at a heating rate of 10°C/min by using indium, lauric acid, and stearic acid (p.a.) as standards. Freeze-dried samples (12-16mg) were suspended in the corresponding buffer (already detailed) at 200-250mg/L, and hermetically sealed in aluminum pans using a double pan as reference. Samples were ran at a rate of 10°C/min from 27-120°C. After each run, the dry matter content was determined by puncturing the pans and keeping them at 107°C overnight. Denaturation temperature (T^sub d^) was taken as the maximum deflection temperature, and the transition enthalpy, calculated from the area under the transition peaks, was determined. At least three analyses were performed for each sample.

Gel-filtration chromatography: Isolates were suspended at 25mg/mL in the different buffers (already indicated). Samples were analyzed at room temperature on a Superóse 6B HR 10/30 column using a Pharmacia LKB FPLC system. Samples contained 1-3mg of protein in 200µL of buffer. Elution was performed with the same buffer as that used for the sample at a flow rate of 0.2mL/min. Fractions (0.3mL) were collected, and the elution profile (absorbance at 280nm) was recorded. The column was calibrated with Dextran Blue (Vo) and with proteins of known molecular weight: thyroglobulin (669,000), apoferritin (443,000), ?-amylase (200,000), and alcohol dehydrogenase (150,000). The calibration curve obtained from duplicate runs was determined as logMW = 6.7 - 0.29 Ve; where Ve is the elution volume in mL and MW is the molecular weight. Three replicates of each sample were analyzed by chromatography.

SDS-PAGE: All gels were run in minislabs (Bio-Rad Mini Protean II model). SDS-PAGE was performed using the method in Laemmli (1970) as modified by Petruccelli and Añón (1994). Runs were made in continuous buffer systems: 0.0125M Tris-HCl pH 6.8/0.1% (w/v) SDS for the stacking gel; 0.0375M Tris-HCl, pH 8.8/0.1% (w/v) SDS for the separating gel; and 0.025/W Tris-HCl/0A92M glycine/0.1% (w/v) SDS, pH 8.3 for the running buffer. A linear grathent separating gel of 5-15% w/v Polyacrylamide was used. The protein samples (10 mg/mL) were dissolved in 0.125M Tris-HCl, pH 6.8/20% (v/v) glycerol/0.1% (w/v)/0.05% (w/v) bromophenol blue, and centrifuged at 15,800 ? g for 5 minutes at room temperature. The supernatants were loaded onto the gel at 30-40 µg of protein per lane. Some samples were run under reducing conditions; in this case, the samples were boiled for 1 min in sample buffer containing 5% (v/v) 2-mercaptoethanol (2-ME) before centrifugation. Electrophoresis runs were conducted for 1 hr at a constant voltage of 200V. Molecular weight standards (Pharmacia) used to estimate the molecular masses of polypeptides were Phosphorylase b (94,000 kDa); bovine serum albumin (67,000 kDa); ovalbumin (45,000 kDa); carbonic anhydrase (30,000 kDa); trypsin inhibitor (20,100 kDa); a-lactalbumin (14,400).

Fluorescence spectrophotometry: Isolates were suspended in the different buffers, dispersions were gently stirred for 1 hr at room temperature and centrifuged at 8,500 ? g for 30 minutes at 150C. The concentration of the soluble fraction of protein was normalized at 0.2mg/mL (Lowry et al 1951). Fluorescence spectra of the samples were determined twice on a Perkin-Elmer LS-50B spectrofluorometer using 1cm light path glass cells at 25°C. A spectral scan was performed at 310-560nm using an excitation wavelength of 290nm and a scanning rate of 300nm/min. Maximum emission wavelength of the samples and maximum fluorescence intensity were determined. At least two analyses were performed for each sample.

Hydrophobicity: The surface hydrophobicity of isolate suspensions at pH 4, 7, and 9, were gently stirred for 1 hour at room temperature and centrifuged at 8,500 Ö g for 30 minutes at 15°C, and the concentration of the soluble fraction of protein was measured according to Lowry et al (1951). Ho was evaluated by the method of Kato and Nakai (1980) using ANS (8-anilino-1-naphthalene-sulfonic acid) as a fluorescent probe. The index of protein Ho was calculated as the initial slope of fluorescence intensity versus protein concentration. The fluorescence intensities of the blank (FIb) and of the ANS-protein conjugate (FIe) were recorded at ?e?: 363 nm and ?em: 475 nm, using 5-nm emission and excitation slit widths, on a Perkin-Elmer 2000 fluorescence spectrometer. At least three analyses were performed for each sample.

Solubility: Samples of freeze-dried protein isolates were dispersed in buffers with at pH 2.3-11.0 (the same buffers used in the calorimetric analysis), with a protein concentration of 1.0% w/v, and were kept at room temperature for 2 hours with periodic stirring every 15 minutes in a Thermolyne Maxi-Mix II mixer set at maximum speed. They were then centrifuged at 18,000 Ö g for 30 min to determine protein in the supernatants by the method of Lowry et al (1951). Solubility was expressed as 100 Ö protein content in the supernatant/total protein content (g% w/w). Bovine serum albumin was used as the standard. These determinations were performed in triplicate.

Water holding capacity: Samples were dispersed in distilled water at 1% w/v using a vortex mixer and then stirred for 30 min at 25°C. After the mixture was thoroughly wetted, the samples were centrifuged (8,500 Ö g for 30 min). After centrifugation, the amount of added distilled water remaining in the supernatant liquid (soluble fraction) in the test tube was recorded and soluble proteins in the supernatant were determined according to Lowry et al (1951). WHC (g of water/g of sample) was calculated as WHC = (m2 2[mL - m3])/ml d: where mL is the weight of dry sample (g); m2 is the weight of sediment (insoluble fraction, g); m3 is the weight of soluble protein from the supernatant (g), and d is the density (mL/g). Determinations were performed in triplicate.

Water-imbibing capacity: WIC was determined using Baumann's apparatus (Torgensen and Toledo 1977). It was expressed as mL of water imbibed/g of sample. Determinations were performed in triplicate.

Determination of foaming properties: Foaming properties were evaluated according to Wagner and Gueguen (1999). Isolates were suspended in different buffers (pH 4, 7, 9) where the dispersions werer gently stirred for 1 hr at room temperature and centrifuged at 8,500 Ö g for 30 min at 15°C. The solution concentration used was 1.0 mg/mL. These solutions were placed in a graduated glass column with a fritted glass plate at the base. Formation of foam was started by bubbling N^sub 2^ (1.7 mL/sec). Bubbling was continued until a fixed volume of foam (100 mL) was reached. The maximum volume of liquid incorporated into the foam (Vmax, mL) and the rate of liquid incorporation to the foam (V^sub 0^, mL/min) were determined, and foam stability was calculated as the specific rate constant of liquid drainage: K = 1/(V^sub max^ t^sub 1/2^) (mL^sup -1^ min^sup -1^). The time was determined for decreasing half the foam conductivity at the end of bubbling (t^sub 1/2^, min). Determinations were performed in triplicate.

Statistical Analyses
Isolates were prepared at least in triplicate and determinations were performed at least in duplicate. The data obtained were statistically evaluated by variance analysis (ANOVA). The comparison of means was done by the least significant difference (LSD) test at a significance level of 0.05. Both used the Statgraphic plus v.5.1 statistical analysis package.

Composition and Structural Characteristic of Protein Isolates
The protein content of the A9 and A11 isolates was 84.4 ± 4.0% (w/w db) and 82.2 ± 5.0% (w/w db), respectively; values similar to those of soy and amaranth protein isolates (Martinez and Añon 1996; Tomotake et al 2002), while the protein content of A9 and A1 was higher than that of sesame isolate (Khalid et al 2003). A9 and A11 isolates had a yield of 52.9-58.8%. These results agree with those obtained by Paredes López et al (1988), Soriano-Santos and Córdoba-Salgado (1995), and Martinez and Añón (1996).

Amaranth protein isolates were analyzed by SDS-PAGE (Fig. 1) and showed similar polypeptide bands. The SDS-PAGE profiles of isolates without 2-ME (Fig. 1B) show peptides at &55,000; polypeptides of molecular mass 80,000 and 40,000 have also been found. Polypeptidic composition is similar to that of the 11S type globulin, glutelins, and albumins SDS-PAGE (Fig. 1B) (Martinez and Añón 1996; Castellani et al 1998; Abugoch et al 2003). Also, SDS-PAGE with 2-ME contained acid (Ac) and B polypeptides linked by a disulfide bond corresponding to MW 33,000 (A) and 21,000 (B) (Fig. 1A). The SDS-PAGE profile in the presence of 2-ME also showed the presence of a peptide monomer of [approximate]57,000 and polypeptides with MW 20,000 corresponding to albumins, both already described by Martínez and Añón (1996).

According to gel filtration (FPLC) results, the A9 isolate contained 23.5% of globulin-P and 1 lS-globulin, and 55.9% of albumins and globulins. The All isolate contained 65% of albumins and globulins, and a small proportion of globulin aggregates, which suggests the presence of species with a higher degree of denaturation or a higher degree of molecular dissociation compared to A9. The presence of glutelins extracted during flour treatment at pH 11 cannot be ruled out.

The DSC analysis of the A9 isolate revealed the presence of two thermal transitions with denaturation temperatures (Td) of 65.8 ± 0.2 and 98 ± 3°C, respectively. In contrast, the All isolate exhibited a small endotherm at T¿ 66.6 ± 0.2°C and a second less defined endotherm at 95-110°C. The denaturation enthalpy of proteins present in A9 was 4.4 ± 0.9 J/g and that of A11 was 2.3 ± 0.7 J/g, both presenting significant differences (P 0.05) between A9 and A11. These results indicate that proteins present in A9 contained a less denatured protein population compared to the A11 isolate proteins.

Structural Changes Produced by pH Treatment
Gel-filtration chromatograms (FPLC) were obtained (Fig. 2) for the A9 isolate at pH 3, 5, 7, 9, and 10. The profile at pH 7 indicates that aggregates (identified by A in the figure) were present in very low proportion, while hexameric molecules of P-globulin and 11S globulin contributed 23% of the protein content. There are species of low molecular weight too, corresponding to part of the protein faction of globulins and albumins. When the pH became more acidic, low molecular weight species predominated. At pH 5, there were virtually no molecules of MW 280,000 (zone I), probably because this pH is close to the pi of globulins, while species of lower molecular weight predominate. Molecules of low molecular size were observed at pH 3, which probably constituted dissociated polypeptides (zones II and III) because proteins were denatured and dissociated at this pH. The profile obtained at pH 9 presented a more pronounced peak than that at pH 7, which was probably formed by high molecular weight soluble globulin-P aggregates (A) (Castelanni et al 1998). Within zone I, together with the unit molecules, there was a peak that would correspond to subunits derived from the dissociation of some unit molecules. At pH 11, there was an increase of high molecular weight aggregates (A) and dissociated subunits. On the basis of these results, it can be concluded that acidification renders the protein molecules present in A9 insoluble, while alkalinization leads to the solubilization of insoluble aggregates and a partial dissociation of globulin molecules.

The results of the size-exclusion chromatography of the A11 isolate are shown in Fig. 2. According to these results, the isolate presented a behavior similar to that described for A9. Low molecular weight species, corresponding to albumins, predominated at pH 7, while molecules of higher molecular weight, corresponding to P-globulin and HS globulin, increased at alkaline pH, and disappeared at acidic pH values.

The surface hydrophobicity (Ho) of the isolates at pH 4, 7, and 9 is shown in Fig. 3. At all these pH values, there were significant differences (P 0.05) between both isolates. At pH 4, both isolates showed low surface hydrophobicity, which increased with increasing pH. This behavior may have been due to partial unfolding and also dissociation of proteins, as observed by size-exclusion chromatography. The A11 isolate presented a higher Ho at pH 9, which can be attributed to a higher degree of unfolding of its molecular species. The Ho values obtained suggest that, although both isolates are constituted by protein species of similar size according to FPLC results, such species would differ in their composition or conformation.

Fluorescence emission spectra of the isolates as a function of pH is shown in Fig. 4. For both isolates, it can be seen that the spectra obtained at acidic pH (pH 3 and 5) differed from those at pH 7 more than those obtained at alkaline pH. These variations can be attributed to changes induced by pH in the tryptophan environment of a given protein or to the presence of different proteins soluble at these pH values, which may differ in tryptophan content. At acidic pH, the fluorescence intensity of the isolates (FI) decreased and the maximum wavelength (?^sub max^) shifted to higher values, while at basic pH, the latter parameter remained unchanged and FI increased. The changes observed at acidic pH would indicate the presence of tryptophan residues more exposed to the polar medium under these conditions. Since Ho values were lower under these conditions, these results suggest that species soluble at acidic pH present a lower content of tryptophan residues. The behavior of the isolates differed slightly at alkaline pH, which may be due to the solubilization of new species or conformational changes. The greater increase of FI in A11 correlates with the higher Ho of this isolate at alkaline pH, suggesting that the conformation of proteins present in A11 and A9 is different.

The results of the calorimetric analysis of the A9 isolate at different pH levels are shown in Fig. 5. The thermograms obtained at pH 2, 3, and 4 did not show endotherms, indicating that the molecules present under these conditions were totally denatured. At pH 5, the proteins had some structure and yielded a T^sub d^ of 97.2 ± 0.1°C. The conditions of highest stability were observed at pH 6, 7, and 8, with T^sub d^ values from 100.7 ± 0.1 to 104.3 ± 0.8°C. The T^sub d^ decreased at pH 9 (99.8°C), and two small endotherms were observed at pH 10 and 11. The denaturation enthalpy was maximum at pH 6-9 (values of 7-7.8 J/g), indicating molecules in the highest structured state. Enthalpy values decreased significantly at pH 5, 10, and 11. The Î"T^sub 1/2^ values, which reflect the cooperativity of the denaturation process, were lowest between pH 6 and 8, indicating a higher cooperativity under these conditions than at pH 5, 9, and 10. The results obtained with A11 were similar to those for A9 (Fig. 6), except that proteins had a lower degree of structure, exhibiting the highest enthalpy values between pH 7 and 9 (5.8-6.6 J/g) while being completely denatured or highly denatured at pH 6.

Functional Properties
The results obtained for isolate solubility as a function of pH are shown in Fig. 7. The isolates exhibited similar solubilities along the whole pH range analyzed. Solubility at pH 2 was higher than at pH 3 and 5, and tended to increase significantly at alkaline pH, reaching 80-90% (w/w) at pH 9-11. Solubility changes with the acidity of the medium are related to changes in the net charge of the protein. These variations in solubility agree with those reported for amaranth and soy globulins (Marcone and Kakuda 1999) and for buckwheat proteins (Tomotake et al 2002).

The WHC of the insoluble fraction in aqueous medium was similar for both isolates (1.9 ± 0.5 mL water/g of protein). If these values are compared with those obtained for other protein isolates such as soy, sesame, and cashew nut, the WHC of amaranth proteins is equal or higher (Khalid et al 2003; Ragab et al 2004).

The analysis of the WIC indicated that the All isolate had the higher initial water absorption (1.1 ± 0.1mL/ming), while the corresponding value for A9 was 0.5 ± 0.1mL/min g. Equilibrium was reached at 0.6 minutes in All and at 0.9 minutes in A9. WIC values were 2.5 ± 0.1mL/g and 1.7 ± 0.3mL/g, respectively. These results reveal a greater WIC for A11, which has a higher content of denatured proteins.

The influence of pH (pH 4, 7, and 9, with ? = 0.5) on the foaming capacity of amaranth proteins from A9 and A11 isolates was also analyzed. Calculated values for the initial rate of foam formation (V^sub o^) ranged from 9-19mL/min (Fig. 8A). These values agree with those obtained for soy proteins by Sorgentini and Wagner (2002). The A11 isolate exhibited a tendency to form foam at a higher rate than the A9 isolate, especially at acidic pH. The stability of the foam was evaluated with the draining rate constant (K) values shown in Fig. 8. The stability of isolates to drainage increased with increasing pH. At acidic pH, the stability of foams formed by A11 was higher than those of A9. At acidic or neutral pH, the amaranth protein foams obtained under the present assay conditions were less stable than bovine serum albumin foam. In contrast, at alkaline pH, the stability of amaranth protein foams was the same as that of bovine serum albumin foam, with a drainage rate constant of 0.12mL/min.

The results of the present study show that amaranth proteins soluble at acidic (3-4), neutral, or alkaline pH (9-10) and constant ionic strength (0.5) differ in structure. At neutral pH (6-8), A9 are more folded and the conformation is closer to the native state than those in A11. A rapid loss in stability was observed at extreme pH, either acidic or alkaline, with a partial unfolding of the molecular structure that was reflected by a reduction of the temperature and enthalpy of thermal treatment denaturation.

In addition, the soluble protein exhibited changes in surface hydrophobicity, UV fluorescence spectra, and size-exclusion chromatography patterns, probably due to changes in the net charge of proteins and modification of electrostatic repulsions. The structural changes produced by alkalinization were less important than those induced by acidification. At pH 7, a greater solubilization of P-globulin and 11S globulin aggregates was also detected, together with a partial dissociation of P-globulin, in agreement with data obtained by other authors for amaranth and soy proteins (Peng et al 1984; Marcone and Yada 1991).

These structural changes were reflected in the hydration and foaming properties of proteins present in A9 and A11 isolates. For both isolates, the maximum solubility was obtained at alkaline pH, in which proteins are partially denatured, an effect similar to that described for other plant proteins (Mahajan and Dua 2001; Adebowale et al 2007; Abugoch et al 2008). As to water retention capacity, no correlation was found with the degree of protein denaturation. The WIC was higher in the more denaturated isolate (A11). The structural modifications produced by pH changes did not affect the tensoactive properties of amaranth proteins, as reflected by a similar foaming capacity upon bubbling. However, the pH of the medium influenced markedly the stability of the foam, which was less stable at acidic and neutral pH. This behavior may be due to a lower solubility and the presence of dissociated and denatured protein species under acidic conditions. In contrast, foam obtained at alkaline pH had high stability, comparable to that of bovine serum albumin foam.

In conclusion, extreme pH treatment of amaranth isolates generates partial or total protein denaturation and enhancement of some functional properties. Protein isolates A9 and A11 differed in structural and thermal properties, denaturation temperature, and enthalpy of denaturation. Proteins present in A9 are more folded and the conformation is closer to the native state, while those of A11 are more unfolded. A9 protein isolate differed significantly from A11 with respect to WIC and foaming properties. Protein solubility profiles showed decreasing solubility with increasing pH until it reached a minimum at the isoelectric point (pH 4.0-4.5).

Lilian E. Abugoch,1,2 Nora E. Martinez, 3 and María Cristina Añón2,3
1 Departamento Ciencia de los Alimentos y Tecnología Química. Facultad de Ciencias Químicas y Farmacéuticas. Universidad de Chile. Vicuña Mackenna 20. Santiago, Chile.
2 Corresponding author. Phone (56)2-9781635. Fax (56)2-2227900. E-mail address;
3 Centro de Investigación y Desarrollo en Criotecnología de Alimentos (CIDCA). Facultad de Ciencias Exactas, Universidad Nacional de La Plata y Consejo Nacional de Investigaciones Científicas y Te'cnicas. calle 47 y 116, 1900 La Plata, Argentina.

Abugoch, L., Martínez, N., and Añon, M. C. 2003. Influence of the extracting solvent upon the structural properties of amaranth (Amaranthus hypochondriacus) glutelin. J. Agric. Food Chem. 51:4060-4065.
Abugoch, L., Romero, N., Tapia, C, Silva, J., and Rivera, M. 2008. Study of some physicochemical and functional properties of quinoa (Chenopodium quinoa Willd.) protein isolates. J. Agric. Food Chem. 56:4745-4750.
Adebowale, Y., Adeyemi. L1 Oshodi, A., and Niranjan. K. 2007. Isolation, fractionation and characterization of proteins from Mucuna bean. Food Chem. 104:287-299.
AOAC. 1996. Official Methods of Analysis of AOAC International. 16th Ed. Method 945.18. AOAC Int.: Gaithersburg, MD.
Avanza, M. V., Puppo, M. C, and Añón. M. C. 2005. Structural characterization of amaranth protein gels. J. Food Sci. 70:E223-229
Becker, R., Wheeler, E. L., Lorenz, K., Stafford, A. E., Grosjean, O. K., Betschart, A. A., and Saunders, R. M. 1981. A compositional study of amaranth grain. J. Food Sci. 46: 1 175-1 1 80.
Bejosano, F, and Corke, H. 1998. Effect of amaranthus and buckwheat proteins on wheat dough properties and noodle quality. Cereal Chem. 75:171-176.
Bernardino-Nicanor, A.. Añón, M. C, Scilingo, A., and Dávila-Ortiz, G. 2005. Functional properties of guava seed glutelins. J. Agric. Food Chem. 53:3613-3617.
Bressani, R. 1989. The proteins of grain amaranth. Food Rev. Int. 5:13-38.
Castellani, O., Martínez, N., and Anón. M. C. 1998. Structural modifications of an amaranth globulin induced by pH and NaCl. J. Agric. Food Chem. 46:4846-4853.
Cordero-de-los-Santos. J., Osuna-Castro, A.. Borodanenko, A., and Paredes-López, O. 2005. PhysicochemicaJ and functional characterisation of amaranth (Amaranthus hypochondriacus) protein isolates obtained by isoelectric precipitation and micellisation. Food Sci. Technol. Int. 11:269-280.
Fidantsi, A., and Doxastakis. G. 2001. Emulsifying and foaming properties of amaranth seed protein isolates. Colloids Surfaces B: Biointerfaces 21:119-124.
Guzmán-Maldonado. S., and Paredes-López, O. 1999. Functional products of indigenous plants to Latin America: Amaranth, quinoa, common beans and botanicals. Pages 293-328 in: Functional Foods: Biochemicals and Processing Aspects. G. Mazza, ed. Technomic Publishing: Lancaster, PA.
Kato. A., and Nakai, S. 1980. Hydrophobicity determined by a fluorescence probe method and its correlation with surface properties of proteins. Bioch. Biophys. Acta 624:13-20.
Khalid, E.. Babiker, E.. and Tinay, A. 2003. Solubility and functional properties of sesame seed proteins as influenced by pH and/or salt concentration. Food Chem. 82:361-366.
Konishi, Y. and Yoshimoto, N. 1989. Amaranth globulin as a heat-stable emulsifying agent. Agric. Biol. Chem. 53:3327-3328
Laemmli. U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.
Lowry, O. H., Rosebrough, N. J., Fair. A. L.. and Randall. R. J. 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193:265-275.
Mahajan, A., and Dua, S. 2001. Salts and pH Induced changes in functional properties of amaranth (Amaranthus tricolor L.) seed meal. Cereal Chem. 79:834837.
Marcone, M., and Kakuda. Y. 1999. Functional properties of amaranth and soybean globulins. A comparative study of the functional properties of amaranth and soybean globulin isolates. Nahrung Food 43:368-373.
Marcone, M. F.. and Yada. R. Y 1991. Isolation, purification, and characterization of the oligomeric seed globulin from Amaranthus hypochondriacus. Agrie. Biol. Chem. 55:2281-2289.
Martínez, N. E.. and Anón. M. C. 1996. Composition and structural characterization of amaranth protein isolates: An electrophoretic and calorimetrie study. J. Agric. Food Chem. 44:2523-2530.
Martínez, N. E., Castellani, O. F., and Añón, M. C. 1997. Common molecular features among amaranth storage proteins. J. Agric. Food Chem. 45:3832-3839.
National Academy of Sciences. l975.Underexploited tropical plants with promising economic value. National Acad. Sci. Press: Washington. DC.
Paredes-López, O.. Mora-Escobedo, R.. and Odorica-Falomir. C. 1988. Isolation of amaranth proteins. LWT Food Sci. Technol. 21:59-61.
Peng. I. C, Quass, D. W., Dayton. W. R.. and Allen, C. E. 1984. Review: Physicochemical and functional properties of soybean 11s globulin. Cereal Chem. 61:480-490.
Petruccelli, S., and Añón. M. C. 1994. Relationship between the method of obtention and the structural and functional properties of soy protein isolates. 1. Structural and hydration properties. J. Agric. Food Chem. 42:2161-2169.
Ragab, D., Babiker, E., and Eltinay, A. 2004. Fractionation, solubility and functional properties of cowpea (Vigna unguiculata) proteins as affected by pH and/or salt concentration. Food Chem. 84:207-212.
Saunders, R., and Becker, R. 1984. Amaranthus: A potential food and feed resource. Pages 357-397 in: Advances in Cereal Science and Technology. Y. Pomeranz, ed. AACC International: St. Paul, MN.
Scilingo, A., Molina, S., Martínez, N., and Añón. M. C. A. 2002. Amaranth protein isolates modified by hydrolytic and thermal treatments. Relationship between structure and solubility. Food Res. Int. 35:855-862.
Segura-Nieto. M., Barba de la Rosa, A. P.. and Paredes-López, O. 1994. Biochemistry of amaranth proteins. Pages 75-106 in: Amaranth. Biology. Chemistry, and Technology. O. Paredes-López. ed. CRC: Boca Raton, FL.
Sorgentini, D. A., and Wagner. J. R. 2002. Comparative study of foaming properties of whey and isolate soybean proteins. Food Res. Int. 35: 721-729.
Sorgentini. D. A., Wagner. J. R.. Arrese. E. L., and Añón. M. C. 1991. Water imbibing capacity of soy protein isolates: Influence of protein denaturation. J. Agric. Food Chem. 39: 1386-1391.
Teutonico, R. A., and Knorr, D. 1985. Amaranth: Composition, properties and applications of a rediscovered food crop. Food Technol. Chicago 39:49-61.
Tomotake, H., Shimaoka, L. Kayashita. J., Nakajoh, M., and Kato, N. 2002. Physicochemical and functional properties of buckwheat protein product. J. Agric. Food Chem. 50:2125-2129.
Torgensen, H., and Toledo. R. T. 1977. Physical properties of proteins preparations related to their functional characteristics in comminuted meat systems. J. Food Sci. 42:1615-1620.
Wagner, J., and Gueguen, J. 1999. Surface functional properties of native, acid-treated, and reduced soy glycinin. 1. Foaming properties. J. Agric. Food Chem. 47:2173-2180.

From the October 4, 2010, Prepared Foods E-dition