- Email: [email protected]
Characterization of iron-binding phosphopeptide released by gastrointestinal digestion of egg white
Food Research International 67 (2015) 308–314
Contents lists available at ScienceDirect
Food Research International journal homepage: www.elsevier.com/locate/foodres
Characterization of iron-binding phosphopeptide released by gastrointestinal digestion of egg white Ravindranadh Palika a, Purna Chandra Mashurabad a, Madhavan K. Nair a, G. Bhanuprakash Reddy b, Raghu Pullakhandam a,⁎ a b
Biophysics Division, National Institute of Nutrition, Indian Council of Medical Research, Jamai Osmania, Hyderabad 500 007, India Biochemistry Division, National Institute of Nutrition, Indian Council of Medical Research, Jamai Osmania, Hyderabad 500 007, India
a r t i c l e
i n f o
Article history: Received 8 October 2014 Accepted 25 November 2014 Available online 3 December 2014 Chemical compounds studied in this article:: Ferric chloride (PubChem CID: 23673676) Ferrozine (PubChem CID: 34127) α-Cyano-4-hydroxycinnamic acid (PubChem CID: 5328791) 2-(N-Morpholino) ethanesulfonic acid sodium salt (PubChem CID: 78165) Acetonitrile (PubChem CID: 6342) Ascorbic acid (PubChem CID: 54670067)
a b s t r a c t Binding and solubilization of ferric iron by food peptides, released during digestion, facilitate intestinal iron absorption. In the present study, we investigated the release of iron-binding peptides during in vitro gastrointestinal digestion of chicken (Gallus gallus) egg white. The iron-binding activity of the egg white protein increased upon gastrointestinal digestion. The iron-binding fraction of egg white digesta was purified by gel filtration chromatography followed by reverse phase HPLC. Subsequently, this fraction was identified as an internal fragment of ovalbumin (DKLPGFGDS(PO)4IEAQ, 61–73 residues, GenBank AAB59956.1) by MALDI-MS/MS followed by de novo sequencing. The synthetic peptide corresponding to the identified iron-binding peptide sequence bound and increased the 59Fe-iron uptake. Further, the synthetic peptide also stimulated the iron-induced ferritin synthesis in intestinal Caco-2 cells. While, dephosphorylation of synthetic peptide completely inhibited the iron-binding activity, methyl-esterification of its carboxyl groups partially inhibited the activity. These results suggest that food derived peptides modulate intestinal iron absorption and that the isolated iron-binding egg peptide could be a potential nutraceutical for improving iron absorption. © 2014 Elsevier Ltd. All rights reserved.
Keywords: Bioavailability Caco-2 cells De novo sequencing Iron-binding peptides In vitro digestion
1. Introduction Iron exists in ferric (Fe3+) and ferrous (Fe2+) forms in nature. Dietary iron is available mainly in two forms, the heme- and nonheme iron. Heme-iron is better absorbed than non-heme iron (Baker, Anderson, & Baker, 2003). Dietary non-heme iron present predominantly in ferric form is reduced by duodenal cytochrome-B (DcytB) prior to its intestinal absorption via divalent metal ion transporter 1 (DMT1) (Sharp & Srai, 2007). The intestinal absorption of ferric-iron is limited by its poor solubility at near neutral pH, and chelation improves its solubility and intestinal absorption (Hurrell et al., 2004; Palika et al., 2013). Phytic acid (inositol-hexaphosphate, IP6) and polyphenols, the major secondary metabolites of plant foods, are potent inhibitors of non-heme iron absorption (Hurrell, 2002; Nair & Iyengar, 2009). Therefore, the observed high prevalence of iron deficiency anemia in ⁎ Corresponding author at: Biophysics Division, National Institute of Nutrition, Jamai Osmania, Hyderabad 500 604, India. Tel.: +91 40 27197323; fax: +91 40 27019074. E-mail address: [email protected] (R. Pullakhandam).
http://dx.doi.org/10.1016/j.foodres.2014.11.049 0963-9969/© 2014 Elsevier Ltd. All rights reserved.
vegetarians could be due to poor density and bioavailability of iron (Nair & Iyengar, 2009). Therapeutic iron supplementation and food fortification are potential strategies to prevent iron deficiency anemia (Hurrell, 2002; Hurrell et al., 2004; Nair & Iyengar, 2009). However, oral iron therapy is associated with free radical damage, and the addition of iron to food results in unacceptable organoleptic properties due to the pro-oxidant nature of iron (Casanueva & Viteri, 2003; Douglas, Rainey, Wong, Edmonson, & LaCroix, 1981). Although more inert (encapsulated) or chelated forms of iron increase the shelf life of products, the iron bioavailability from such sources remains a serious concern (Hurrell et al., 2004). Therefore, alternate strategies such as selection of iron fortificants with higher iron solubility/bioavailability or addition of components that increase iron solubility and hence absorption are needed. Studies over the past two decades have demonstrated that iron bioavailability from protein rich foods, particularly from animal sources, is high (Cook & Monsen, 1976; Hurrell, 2002; Hurrell et al., 2004; Nair & Iyengar, 2009). Histidine or cysteine-rich peptides released during the
R. Palika et al. / Food Research International 67 (2015) 308–314
digestion of dietary protein are reported to stimulate intestinal iron absorption (Hurrell, Reddy, Juillerat, & Cook, 2006; Storcksdieck & Hurrell, 2007; Swain, Tabatabai, & Reddy, 2002; Taylor, Martinez-Torres, Romano, & Layrisse, 1986). Indeed, casein phosphopeptides (CPP) isolated from in vitro digests of milk have been demonstrated to improve mineral bioavailability including that of iron (Bouhallab & Bougle, 2004). Ironbinding peptides have also been isolated from soybeans, whey protein and yeast (de la Hoz et al., 2014; Lv et al., 2009). These observations suggest wide-spread occurrence of iron-binding peptides in foods that are released during digestion. However, the quantity of such peptides present or released during digestion is a subject of investigation. Further, the nature of the iron-binding peptides and their precursor proteins remains to be characterized from different food sources. Hence, isolation of these peptides in pure form and characterization of their primary structure shall aid immensely in understanding the nature of iron-binding food peptides and associated mineral binding mechanisms. Eggs, a rich source of protein, are widely consumed. Peptides with antiviral, anti-hypertensive, anti-microbial, anti-cancer and antioxidant activities that have been identified in the enzymatic hydrolysates of egg proteins (Mine, 2007). Studies in humans and animal models have reported the inhibitory effect of eggs on iron bioavailability (Grengard, Sentenac, & Mendelsohn, 1964; Rose, Vahlteich, & Macleod, 1934). The egg yolk phosphovitin, an extensively phosphorylated protein, binds iron and prevents its release during gastric and intestinal phases of digestion (Grengard et al., 1964). However, iron-binding peptides if any, released during the digestion of egg white proteins and their effect on intestinal iron absorption are not studied yet. Recently Abeyrathne, Lee, Jo, Nam, and Ahn (2014) reported the release of iron and copper binding peptides during enzymatic digestion of ovalbumin. However, the nature of the specific peptides involved in mineral binding and associated mechanisms remains to be studied. The present study examined the effect of gastric and intestinal digestion of egg white protein on the iron-binding capacity and attempted to decipher the molecular identity of specific peptide(s) involved in ironbinding. Further, the effect of the isolated iron-binding peptide on iron uptake and iron induced ferritin synthesis was also assessed in intestinal cells. 2. Material and methods 2.1. Materials Porcine pepsin (Cat#P7012), pancreatin (Cat#P7545) and all other chemicals/reagents were obtained from Sigma Chemical Co. (Bangalore, India), unless otherwise specified. 59FeCl3 (carrier free) was obtained from Board of Radiation and Isotope Technology (BRIT), Mumbai, India. Chicken eggs were purchased from the local market. Phosphorylated synthetic peptide (DKLPGFGDS(PO)4IEAQ) was obtained from GenScript (NJ, USA) and its purity (N 98%) was confirmed by MALDI-MS. 2.2. In vitro digestion and isolation of iron-binding peptides 2.2.1. Simulated gastrointestinal digestion Chicken eggs (Gallus gallus, 4 Nos) were boiled in water for 20 min and the solid whites were collected, and washed with Milli Q water. The egg whites (112 g) were cut into small pieces, suspended in 250 mL normal saline and homogenized in a kitchen blender for 2 min at the maximum speed (18,000 rpm). The simulated in vitro gastric and intestinal digestion was carried out as described previously (Glahn, Lee, Yeung, Goldman, & Miller, 1998; Pullakhandam, Nair, Pamini, & Punjal, 2011), except that the digestion was carried out for 6 h to ensure complete digestion of the egg white protein. Briefly, the pH of the egg white suspension was adjusted to 2 with 6 N HCl followed by the addition of 1 g of porcine pepsin (≥ 2500 units/mg protein, dissolved in 10 mL of 1 N HCl). The pepsin digestion was carried out
309
for 6 h at 37 °C in a shaking water bath. At the end of this step, the pH of sample was raised to 6.5 with 2 M NaHCO3 followed by the addition of 1 g of porcine pancreatin (8X USP equivalents, dissolved in 0.01 mM NaHCO3). The samples were incubated at 37 °C for 6 h to complete digestion. During different stages of digestion, an aliquot of the digestion mixture was drawn for testing the iron solubilization activity as described below. At the end of simulated gastrointestinal digestion, the samples were clarified by centrifugation at 10,000 g for 15 min at 4 °C and the supernatant is referred to as the ‘digesta’. 2.2.2. Ultrafiltration The digesta was subjected to ultrafiltration through 5 kDa cut-off centrifugal filters (Millipore) to collect b 5 kDa peptides in the filtrate (5kF) and N5 kDa proteins/peptides in the retentate (5kR). The 5kF and 5kR fractions were frozen and stored at − 20 °C, until further analysis. 2.2.3. Solid phase extraction Solid phase extraction was performed on an octadecyl-silica (C-18) matrix (Discovery DSC 18, Sigma). Briefly, the glass column consisting 5 g of C-18 matrix was equilibrated with 2% acetonitrile (v/v) containing 0.1% trifluoroacetic acid (TFA). The 5kF fraction (10 mL containing 0.1% TFA) was loaded on the column, washed with 100 mL of 2% acetonitrile to remove unbound proteins. The bound proteins were then eluted with 10 mL of 50% acetonitrile, concentrated in a vacuum evaporator (Vacufuge Plus, Eppendorf). The dried fractions were stored at −20 °C, until further analysis. They were reconstituted in saline (0.9% NaCl, containing 0.1% TFA) prior to the analysis. 2.2.4. Iron solubilization assay Ferric iron is not soluble at neutral pH unless it is reduced to ferrous form or is bound to soluble components. Therefore, estimation of soluble iron in the absence of reduction provides a direct measure of ironbinding activity. The ferric iron solubilization assay was performed as described previously (Palika et al., 2013). Briefly, 100 μL aliquots of the digesta, 5kF, 5kR, C-18 bound and flow-through fractions or synthetic peptide (0 to 50 μM) were diluted with 50 mM 2-(N-Morpholino) ethanesulfonic acid sodium salt (MES) buffer pH 6.5 to 1 mL and supplemented with 25 μM FeCl3 (traced with 50 nCi 59FeCl3), and incubated for 30 min at 37 °C. At the end of incubation, samples were centrifuged at 10,000 g at 4 °C for 15 min and the supernatant solution (800 μL) was mixed with 5 mL of Bray's mixture and counted in liquid scintillation counter (PerkinElmer; TRICARB 2900TR). The percent binding of iron was calculated considering the solubility of 59Fe iron in 6 N HCl as 100%. 2.2.5. Ferric iron reduction assay The reduction of ferric iron was measured as described previously (Palika et al., 2013). Briefly, the reaction mixture (500 μL) contained egg peptide fractions (100 μL), 25 μM FeCl3 and 500 μM ferrozine (a chromogen for ferrous iron) in 50 mM MES pH 6.5. The reaction mixture was incubated for 120 min at 37 °C, and the absorbance was measured at 562 nm in a micro-plate reader (BioTek; Powerwave HT-1). 2.2.6. Gel filtration chromatography A superdex-peptide (10/30, HR) column (Amersham Biosciences) connected to Akta-purifier module (Amersham Biosciences) was equilibrated with 5 column volumes of normal saline. The C-18 bound fraction reconstituted in 500 μL of saline containing 0.1% TFA was loaded on the above column and eluted with the equilibration buffer at a flow rate of 0.5 mL/min, while monitoring the absorption at 215 nm. A total of 29 fractions (1 mL each) were collected and used immediately in the iron solubilization assay. The fractions with iron solubilizing activity were concentrated by solid phase extraction on C-18 column as described above.
310
R. Palika et al. / Food Research International 67 (2015) 308–314
2.2.7. Reverse phase (RP)-HPLC Reverse phase chromatography of peptides was performed on a C-18 column (Thermo-Hypersil ODS, 5 μ, 250 × 4.6 mm) connected to a HPLC equipped with an online UV monitor and controlled by the Chemstation software (Agilent 1100, Pala Alto, CA, USA). The column was eluted with a binary gradient of solvent A (2% acetonitrile (v/v), 0.1% TFA) from 0 to 5 min followed by linear gradient of 50% solvent B (acetonitrile, 0.1% TFA) from 5 to 25 min at a flow rate of 1 mL/min and the eluent was monitored at 215 nm. The individual peaks were collected, concentrated and tested for iron-binding activity as described above.
2.4.2. 59Fe uptake in Caco-2 cells Ferric chloride (2.5 mM stock in 10 mM HCl) was diluted to a final concentration of 25 μM (traced with 0.5 μCi of 59Fe) with MES buffer (50 mM, pH 6.5) in the absence (blank) or presence of ascorbic acid (250 μM), 0, 1:0.5, 1:1 and 1:2 molar ratios of iron to synthetic egg peptide, and then fed to the differentiated Caco-2 cells for a period of 2 h. After incubation the monolayers were washed thrice with ice-cold phosphate buffer saline containing 10 mM bathophenanthroline (to remove non-specifically bound iron), harvested by scraping, and counted in a liquid scintillation counter (PerkinElmer, USA). The percent uptake was calculated considering the uptake with ascorbic acid as 100%.
2.3. Characterization of iron-binding peptide
2.4.3. Ferritin expression in Caco-2 cells The coupled in vitro digestion/Caco-2 cell model was used for measuring the ferritin induction as described previously (Glahn et al., 1998; Pullakhandam et al., 2011). Briefly, 25 μM FeCl3 in the presence and absence of ascorbic acid (1:10 molar ratio) and egg peptide (1:1) was subjected to in vitro gastric and intestinal digestion, followed by feeding the digesta to differentiated Caco-2 cells for a period of 24 h. At the end of incubation, the cells were washed, lysed and ferritin content in the cell lysate was estimated as described previously (Pullakhandam et al., 2011). A human ferritin sandwich ELISA method developed in-house and validated against recombinant ferritin (94/ 572, NIBSC, UK) was used for ferritin estimation in cell lysates. Briefly, ferritin content was estimated in 5 μL of Caco-2 cell lysates using human liver ferritin Ig-G conjugated to horse-radish peroxidase and the substrate system orthophenylenediamine-H2O2. The color intensity was measured using an ELISA plate reader (BioTek, Powerwave HT-1).
2.3.1. Matrix assisted laser desorption ionization–time of flight (MALDI–TOF) spectrometry The dried HPLC fraction with iron solubilizing activity was reconstituted in 20 μL of 2% acetonitrile (with 0.1% TFA). An aliquot (1 μL) of this sample was mixed with equal volume of 50% acetonitrile, 0.1% TFA containing 5 mg/ml α-cyano-4-hydroxycinnamic acid (CHCA), spotted on a stainless steel target plate and dried. All MS and MS/MS spectra were acquired on an ABI 5800 TOF–TOF (AB Sciex, Foster City, CA, USA) instrument that was equipped with a nitrogen laser and operated in a positive-ion delayed extraction reflector mode. Usually, 250 individual spectra of each spot were averaged to produce a mass spectrum. The MS/MS spectra were acquired in 1 kV positive ion mode. Peptide identification was performed by the ProteinPilot™ Software (AB Sciex, USA) against the G. gallus proteins from SwissProt databank (www.expasy.org) using the following search criteria: oxidation of methionine as variable modification and phosphorylation of serine/threonine as fixed modifications, no proteolytic enzyme specified and mass accuracy tolerance was set at 50 ppm for parent and 0.5 Da for fragments.
2.3.2. Dephosphorylation To test the effect of dephosphorylation on iron-binding activity, the synthetic peptide (200 μg in 50 mM ammonium bicarbonate buffer) was incubated in the presence and absence of alkaline phosphatase (ALP, 100 units) for a period of 12 h at 37 °C. At the end of incubation the iron-binding activity was assessed as described above.
2.3.3. Methyl-esterification Methyl-esterification of carboxyl groups of the peptide was performed as described previously (Krishnamurthy et al., 1989). Briefly, peptide (200 μg) was dried and incubated in methanol or anhydrous methanolic-HCl (1.25 M HCl) for a period of 12 h at room temperature. At the end of incubation the solvent was dried in a centrifugal evaporator (Vacufuge Plus, Eppendorf, USA), the dried peptide dissolved in 200 μL of Milli Q water (0.1% TFA) and tested for the iron-binding activity as described above.
2.4. Effect of iron-binding peptide on iron absorption in intestinal cells 2.4.1. Caco-2 cell culture Caco-2 cells were obtained from National Centre for Cell Sciences (Pune, India). Cells were seeded at a density of 50,000 cells/cm2 in 6-well plates (Corning, India). The cells were grown in Dulbecco's modified Eagle's medium with 10% fetal bovine serum (FBS), 1% nonessential amino acids, 2 mM glutamine and 1% antibiotic-antimycotic solution. The cells were maintained at 37 °C in an incubator with a 5% CO2/95% air atmosphere at constant humidity. Once the cells reached confluence, the spent media was changed every alternate day. The cells were used for iron uptake experiments 12–14 days post confluence as described previously (Palika et al., 2013; Pullakhandam et al., 2011).
2.5. Statistics All the experiments were performed in triplicates and repeated at least twice to generate 6 observations. Mean and SD were calculated using Microsoft Excel and the data was analyzed using one-way ANOVA, followed by the post hoc least significant differences (LSD) test, using SPSS software (Version 11.0). The results were considered significant if the P b 0.05. 3. Results 3.1. Ferric iron solubilizing activity of egg peptide fractions The iron solubility with egg white protein increased from 17% ± 2 to 42% ± 4.3 as a function of time up to 100 min of pepsin digestion (Fig. 1A). The addition of pancreatin further increased the iron solubility, reaching maximum (102% ± 6.3) between 100–150 min (Fig. 1A). However, no reduction of ferric iron was observed during digestion with either pepsin or pancreatin (data not shown), implying that the increased iron solubilization could be due to binding. The iron-binding activity was higher and similar in the presence of acid or egg protein digesta compared to blank (Fig. 1B). Further, iron-binding activity was significantly higher in 5kF (108% ± 5.6) than 5kR (55.2% ± 6.3) (Fig. 1B). Also the iron-binding activity of C-18 bound fraction (106% ± 5.4) was significantly higher than that of the flow through fraction (36% ± 4.2, Fig. 1B). 3.2. Purification and characterization of iron-binding peptide from egg white digesta The gel filtration chromatography of C-18 bound egg peptide fraction showed 2 major peaks and few minor peaks between 20–60 min (Fig. 2A). Among all the fractions tested, the iron-binding activity was found exclusively in a minor peak eluting between 26–28 min (corresponding approximately to 2000 Da mass; marked with * in Fig. 2A), whereas in the HPLC analysis, the iron-binding activity was found exclusively in the peak eluting at 16.2 min (Fig. 2B).
R. Palika et al. / Food Research International 67 (2015) 308–314
311
Fig. 1. Effect of gastrointestinal digestion on iron solubilizing activity of egg white proteins. Panel A — Iron solubilizing activity of egg white protein digested with pepsin and pancreatin was assessed as a function of time. Panel B — Ferric iron solubilizing activity of egg white digesta during fractionation and purification. The % solubility was calculated considering the solubility of iron in 6 N HCl as 100%. The data represent mean + SD and bars with different superscripts are significantly different (P b 0.05).
3.3. MALDI-MS and de novo sequencing of iron-binding peptide
3.4. Mechanism of iron-binding
The MALDI-MS spectra of iron-binding peptide fraction isolated by RP-HPLC showed a major peak at 1456.74 amu apart from its Na + K adduct showing the peak at 1516.66 amu (Fig. 3A). Interestingly, a peak with a mass of 1358.75 amu (− 98 amu loss) was observed in the mass spectrum, indicating phosphorylated serine or threonine residues. MALDI-MS/MS followed by de novo sequencing analysis identified the peptide as DKLPGFGDS(PO)4IEAQ (Fig. 3B), which is an exact match to the internal fragment of ovalbumin (residues 61–73, GenBank AAB59956.1) with a mascot score of 190 and fragment ion score of 75% (please see supplementary data for de novo sequencing/data base search results).
In order to confirm the iron-binding activity of the egg peptide, further experiments were conducted with the synthetic peptide. The iron-binding activity of the synthetic peptide increased dose dependently and saturated at 1:1 molar ratio of iron to peptide (Fig. 4A). Dephosphorylation of the synthetic peptide with alkaline phosphatase completely inhibited (N92% inhibition) and methyl-esterification of acidic amino acids (aspartic acid and glutamic acid) partially inhibited (40%) its iron-binding activity (Fig. 4A). The molecular model of the isolated egg-phosphopeptide generated from the PDB (ID: 1OVA) crystal structure co-ordinates revealed the extension of serine-phosphate and asp-carboxyl groups outward of the peptide back bone in the similar plane and with a proximity of 5.7 A° (Fig. 4B). 3.5. Effect of synthetic egg peptide on iron absorption and ferritin expression in Caco-2 cells Compared to control (17% ± 3.9), ascorbic acid (1:10 molar ratio) markedly increased (100% ± 5.9) 59Fe uptake in differentiated Caco-2 cells. 59Fe uptake also increased dose dependently in the presence of 1:0.5 (35% ± 5.5) and 1:1 (80 ± 6.9) ratio of iron to synthetic peptide compared to control. However, 59Fe uptake remained unchanged beyond 1:1 molar ratio of iron to peptide (Fig. 5A). Although, 59Fe uptake was significantly inhibited upon dephosphorylation or methylesterification of the synthetic peptide (Fig. 5A), inhibition was significantly higher upon dephosphorylation than methyl-esterification. The ferritin content of Caco-2 cells was similar in the absence or presence of ferric iron. Interestingly, the addition of ascorbic acid (1:10 molar ratio) or the synthetic peptide (1:1 molar ratio) significantly induced the ferritin synthesis in Caco-2 cells, compared to controls albeit, the extent of ferritin induction by ascorbic acid was significantly higher than that of the synthetic peptide (Fig. 5B). 4. Discussion
Fig. 2. Purification of iron-binding peptide released during the digestion of egg white proteins. The egg white peptides released during the gastrointestinal digestion were isolated and enriched by ultrafiltration followed by solid phase extraction. Panel A — An aliquot of this peptide sample was fractionated on Superdex-peptide column as described in the Material and methods section. Panel B — The fractions showing iron-binding activity were further fractionated on RP-HPLC as described in the Material and methods section. The peaks marked with * indicate iron-binding activity.
Bioavailability of iron is reported to be high from protein rich flesh foods. It is believed that cryptic peptides released during the gastrointestinal digestion of food proteins bind and facilitate intestinal iron absorption. In the present study, we have isolated and characterized an iron-binding peptide released during gastrointestinal digestion of egg white proteins. The isolated peptide was identified as an internal fragment of ovalbumin corresponding to the amino acids 61–73, with a phosphoserine at position 69 (DKLPGFGDS(PO4)IEAQ). Further, we substantiated these observations using a synthetic peptide corresponding to this sequence. The synthetic peptide also solubilized and increased the iron absorption in intestinal Caco-2 cells. In addition, we
312
R. Palika et al. / Food Research International 67 (2015) 308–314
Fig. 3. MALDI-MS/MS and de novo sequence analysis of egg white iron-binding peptide. The HPLC peak fraction showing the iron-binding activity was subjected to MALDI-MS and MS/MS analysis. Panel A — MS spectrum of peptide in the positive ion mode. Panel B — Sequence annotation of MS/MS spectrum.
demonstrated that phosphorylation of serine is a prerequisite for ironbinding activity, while the acidic amino acids also play a key role. Pepsin and pancreatin digests but not the total egg protein homogenate significantly increased the solubility of iron at neutral pH (Fig. 1A). Further, there was no reduction of ferric iron with these fractions, indicating that the ferric iron solubilization was indeed due to the binding of specific peptides to ferric iron. Enzymatic hydrolysis of variety of food sources has been reported to release peptides with varied
biological activities such as mineral binding, inhibition of angiotensin-1-converting enzyme (ACE) and antioxidant activity (Mine, 2007; Moller, Scholz-Ahrens, Roos, & Schrezenmeir, 2008). Further, enzymatic hydrolysis of milk protein, chicken and beef had been reported to release iron-binding peptides (Bouhallab & Bougle, 2004; de la Hoz et al., 2014; Hurrell et al., 2006; Lv et al., 2009; Storcksdieck & Hurrell, 2007; Swain et al., 2002; Taylor et al., 1986). It is also known that binding of proteins or peptides
Fig. 4. Effect of dephosphorylation and methyl-esterification on iron solubilization activity of synthetic egg peptide: Panel A — The ferric iron (25 μM) was incubated with synthetic peptide (25 μM) treated with alkaline phosphatase (ALP) or anhydrous methanolic-HCl in 50 mM MES buffer pH 6.5. The iron solubility was assessed as described in the Material and methods section. The % iron solubility was computed considering the iron solubility with peptide as 100%. The data represent mean + SD and the bars that do not share common superscript differ significantly (P b 0.01). Panel B — The molecular model of the peptide was generated using the crystal structure co-ordinates of ovalbumin (PDB: 1OVA) corresponding to 61–73 amino acid residues and the model was visualized using Discovery Studio Visualizer (Version 3.1) to demonstrate the extension of phosphate group of Ser-69 and carboxyl group of Asp-68.
R. Palika et al. / Food Research International 67 (2015) 308–314
313
Fig. 5. Effect of synthetic peptide on 59Fe iron uptake and iron induced ferritin synthesis in differentiated Caco-2 cells: Panel A — Ferric chloride (25 μM, traced with 0.5 μCi of 59Fe) was fed to Caco-2 cells in the presence of ascorbic acid (Vit C) or synthetic egg peptide (Pep) for a period of 2 h. The radioactivity associated with the cells was assessed as described in the Material and methods section. The % uptake was computed considering the uptake with Vit C as 100%. Panel B — Iron, Vit C and peptide mixtures were subjected to simulated in vitro digestion and fed to the Caco-2 cells for a period of 24 h as described in the Material and methods section. Iron induced ferritin expression was estimated by ELISA method. The data represent mean + SD and bars that do not share common superscript differ significantly (P b 0.05).
solubilizes the otherwise insoluble ferric iron at neutral pH. For instance, binding of transferrin and lactoferrin facilitates the solubilization and transport of ferric iron in serum and milk, respectively (Baker et al., 2003). It is therefore likely that peptides generated during the digestion of egg white protein bound and solubilized the ferric iron. Higher iron-binding activity in 5kF than 5kR fraction indicates the low molecular weight (b5 kDa) nature of the ironbinding peptide released during the digestion of egg white protein. It thus appears that the residual iron-binding activity in 5kR fractions could be because of incomplete filtration or the presence of other high molecular weight (N5 kDa) iron-binding peptides. Further, the much higher protein content of 5kR fraction (12 times higher than 5kF) could also contribute to the residual iron-binding activity. Interestingly, the iron solubilizing activity of 5kF fraction was predominant in C-18 bound fraction, implying the hydrophobic nature of iron-binding peptide(s). The low molecular weight nature of the iron-binding fraction is indeed corroborated by gel filtration chromatography of the C-18 bound fraction. Further, RP-HPLC analysis led to the elution of iron-binding activity in a single peak probably suggesting that a specific peptide was involved in iron binding. MALDI-MS analysis followed by de novo sequencing of iron-binding fraction obtained from the RP-HPLC identified that the peptide corresponds to amino acid residues 61–73 of ovalbumin with a phosphoserine at 69th position. In agreement with these results, phosphorylation of Ser-69 was reported in ovalbumin previously (Kinoshita-Kikuta, Kinoshita, & Koike, 2012; Nisbet, Saundry, Moir, Fothergill, & Fothergill, 1981). It was interesting that the synthetic peptide corresponding to this amino acid sequence also solubilized the ferric iron in a dose dependent manner and the solubility was saturated at equimolar concentration of iron and peptide, implying one iron-binding site on the peptide. These results together confirm the identity of primary structure of the peptide and its parent protein. Previous studies reported that phosphopeptides and peptide fractions rich in acidic amino acids bind to metal ions including iron (Stensballe, Andersen, & Jensen, 2001; Storcksdieck & Hurrell, 2007). In line with these reports, dephosphorylation of the synthetic peptide led to the complete inhibition of iron solubilizing activity, while methyl-esterification of carboxyl groups inhibited the activity partially. In agreement with these results, dephosphorylation of casein phosphopeptides is reported to inhibit their Ca binding activity completely (Ferraretto, Gravaghi, Fiorilli, & Tettamanti, 2003). Further, acidic residues in the CPP are predicted to be involved in the stabilization of Ca-phosphate complexes in solution (Cross, Huq, Palamara, Perich, & Reynolds, 2005). Indeed, peptide fractions rich in acidic amino acids are also reported to bind iron in enzymatic hydrolysate of meat and shrimp (Huang, Ren, & Jiang, 2011; Hurrell et al., 2006; Storcksdieck & Hurrell, 2007). In aqueous solution, ferric iron undergoes hydrolytic
polymerization to form insoluble precipitates (Spiro et al., 1966). Since ferric iron is a strong lewis acid, ligation of strong bases such as phosphate, carboxylate and amine functional groups prevents the hydrolytic polymerization and precipitation. It is therefore possible that ligation of peptide phosphoseryl group with ferric iron might prevent its polymerization resulting in solubilization of ferric iron at neutral pH. Further, carboxylate groups of aspartic acid flanking the phosphoserine residue might stabilize the iron–phosphate interaction. Interestingly, molecular model generated using the crystal structure coordinates of ovalbumin showed extension of phosphate and carboxyl group of adjacent aspartic acid from the peptide back bone on a similar plane, supporting the role of both phosphate and carboxylate groups in iron-binding. Nevertheless, further studies are needed to understand the exact role of carboxyl groups in iron binding. The solubility and oxidation state of iron are two major determinants of intestinal iron absorption (Sharp & Srai, 2007). The higher absorption of ferrous iron compared to ferric iron could be explained by its higher solubility at intestinal pH. The solubility of iron salt is positively associated with its absorption by the enterocyte, regardless of the oxidation state (Hurrell et al., 2004). For instance, iron absorption from a soluble ferric iron compound (NaFe (III) EDTA) is similar to that of ferrous iron salt (Hurrell et al., 2004). The addition of either ascorbic acid or peptide significantly increased the uptake of 59Fe in differentiated Caco-2 cells. However, the extent of increase in iron uptake was significantly higher with ascorbic acid, consistent with its ferric iron reducing and solubilizing activity. The fact that the peptide only solubilized but not reduced the ferric iron is in agreement with the iron uptake in intestinal cells. Previously we and others have shown that EDTA and citric acid induced iron uptake in intestinal cells requires the prior reduction of ferric iron and that DcytB could be involved in reducing chelated ferric iron (Palika et al., 2013; Zhu, Yeung, Glahn, & Miller, 2006). It is therefore possible that peptide binding increases the accessibility of ferric iron to the DcytB, for reduction and subsequent uptake by Caco-2 cells. Further, our finding that dephosphorylation of the peptide completely inhibited the 59Fe uptake, while methyl-esterification partially inhibited the uptake, appears to support the role of binding/solubilizing activity in stimulating intestinal iron uptake. The iron absorption stimulating activity of the isolated peptide in vivo depends on its stability during gastrointestinal digestion. Although the peptide was isolated after exhaustive digestion of egg white protein, in silico digestion analysis of the peptide sequence (DKLPGFGDS(PO)4IEAQ) using MS-Digest software indeed identified several proteolytic cleavage sites for pepsin and trypsin (data not shown). In order to rule out further digestion of the peptide, we studied the bioavailability of iron using coupled in vitro digestion/Caco-2 cell model (Glahn et al., 1998; Pullakhandam et al., 2011). The ferritin content of
314
R. Palika et al. / Food Research International 67 (2015) 308–314
Caco-2 cells exposed to ferric iron is similar to that of control suggesting a poor absorption of ferric iron in intestinal cells. However, ferritin content of the cells exposed to ferric iron in the presence of either ascorbic acid or peptide was significantly higher compared to control. Similar to that of iron uptake, the ferritin induction was also higher in the presence of ascorbic acid than the peptide, which in turn is due to reduction of iron by ascorbic acid but not by peptide. In agreement with these results, extensive digestion of the synthetic peptide with pepsin, trypsin and chymotrypsin for a period of 24 h also did not influence its ironbinding activity (data not shown). These results suggest that the isolated peptide was resistant to further digestion by gastric and intestinal proteases. Recently, we have demonstrated that oxytocin, a peptide hormone; though possesses specific proteolytic cleavage sites for pepsin, is resistant to peptic digestion unless its internal disulfide bond is oxidized (Pullakhandam, Palika, Vemula, Polasa, & Boindala, 2014). Thus it appears likely that subtle structural constraints in the peptide may render it resistant to digestion. To the best of our knowledge this is the first systematic study wherein the iron-binding peptide released during the digestion of food protein is isolated, characterized and proof of concept for its activity is provided with its synthetic counterpart. These results strengthen the hypothesis that food derived peptides modulate intestinal iron absorption and that the isolated iron-binding egg peptide could be exploited as a potential nutraceutical for improving iron nutrition. Acknowledgments Supported by Department of Biotechnology (BT/PR5228/FNS/20/ 549/2012), Govt. of India. Ravi and Purna are supported by a Junior Research Fellowship from University Grants Commission, Govt. of India. We thank Dr. P. Ravinder for the assistance with ferritin ELISA and Mr. K. Suresh for the technical assistance with in vitro digestion. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.foodres.2014.11.049. References Abeyrathne, E.D., Lee, H.Y., Jo, C., Nam, K.C., & Ahn, D.U. (2014). Enzymatic hydrolysis of ovalbumin and the functional properties of the hydrolysates. Poultry Science, 93, 2678–2686. Baker, H.M., Anderson, B.F., & Baker, E.N. (2003). Dealing with iron: Common structural principles in proteins that transport iron and heme. Proceedings of the National Academy of Sciences of the United States of America, 100, 3579–3583. Bouhallab, S., & Bougle, D. (2004). Biopeptides of milk: Caseinophosphopeptides and mineral bioavailability. Reproductive Nutrition Development, 44, 493–498. Casanueva, E., & Viteri, F.E. (2003). Iron and oxidative stress in pregnancy. Journal of Nutrition, 133, 1700S–1708S. Cook, J.D., & Monsen, E.R. (1976). Food iron absorption in human subjects. III. Comparison of the effect of animal proteins on nonheme iron absorption. American Journal of Clinical Nutrition, 29, 859–867. Cross, K.J., Huq, N.L., Palamara, J.E., Perich, J.W., & Reynolds, E.C. (2005). Physicochemical characterization of casein phosphopeptide–amorphous calcium phosphate nanocomplexes. Journal of Biological Chemistry, 280, 15362. de la Hoz, L., Ponezi, A.N., Milani, R.F., Nunes da Silva, V.S., Sonia de Souza, A., & BertoldoPacheco, M.T. (2014). Iron-binding properties of sugar cane yeast peptides. Food Chemistry, 142, 166–169.
Douglas, F.W., Rainey, N.H., Wong, N.P., Edmonson, L.F., & LaCroix, D.E. (1981). Color, flavor, and iron bioavailability in iron-fortified chocolate milk. Journal of Dairy Science, 64, 1785–1793. Ferraretto, A., Gravaghi, C., Fiorilli, A., & Tettamanti, G. (2003). Casein-derived bioactive phosphopeptides: Role of phosphorylation and primary structure in promoting calcium uptake by HT-29 tumor cells. FEBS Letters, 551, 92–98. Glahn, R.P., Lee, O.A., Yeung, A., Goldman, M.I., & Miller, D.D. (1998). Caco-2 cell ferritin formation predicts nonradiolabeled food iron availability in an in vitro digestion/ Caco-2 cell culture model. Journal of Nutrition, 128, 1555–1561. Grengard, O., Sentenac, A., & Mendelsohn, N. (1964). Phosvitin, the iron carrier of egg yolk. Biochimica Biophysica Acta, 90, 406–407. Huang, G., Ren, Z., & Jiang, J. (2011). Separation of iron-binding peptides from shrimp processing by-products hydrolysates. Food and Bioprocess Technology, 8, 1527–1532. Hurrell, R.F. (2002). Fortification: Overcoming technical and practical barriers. Journal of Nutrition, 132, 806S–812S. Hurrell, R.F., Lynch, S., Bothwell, T., Cori, H., Glahn, R., Hertrampf, E., et al. (2004). Enhancing the absorption of fortification iron. A SUSTAIN task force report. International Journal of Vitamin and Nutrition Research, 74, 387–401. Hurrell, R.F., Reddy, M.B., Juillerat, M., & Cook, J.D. (2006). Meat protein fractions enhance nonheme iron absorption in humans. Journal of Nutrition, 136, 2808–2812. Kinoshita-Kikuta, E., Kinoshita, E., & Koike, T. (2012). Separation and identification of four distinct serine-phosphorylation states of ovalbumin by Phos-tag affinity electrophoresis. Electrophoresis, 33, 849–855. Krishnamurthy, T., Szafraniec, L., Hunt, D.F., Shabanowitz, J., Yates, J.R., Hauer, C.R., et al. (1989). Structural characterization of toxic cyclic peptides from blue-green algae by tandem mass spectrometry. Proceedings of the National Academy of Sciences of the United States of America, 86, 770–774. Lv, Y., Liu, Q., Bao, X., Tang, W., Yang, B., & Guo, S. (2009). Identification and characteristics of iron-chelating peptides from soybean protein hydrolysates using IMAC-Fe3+. Journal of Agriculture and Food Chemistry, 57, 4593–4597. Mine, Y. (2007). Egg proteins and peptides in human health—Chemistry, bioactivity and production. Current Pharmaceutical Design, 13, 875–884. Moller, N.P., Scholz-Ahrens, K.E., Roos, N., & Schrezenmeir, J. (2008). Bioactive peptides and proteins from foods: Indication for health effects. European Journal of Nutrition, 47, 171–182. Nair, K.M., & Iyengar, V. (2009). Iron content, bioavailability & factors affecting iron status of Indians. Indian Journal of Medical Research, 130, 634–645. Nisbet, A.D., Saundry, R.H., Moir, A.J., Fothergill, L.A., & Fothergill, J.E. (1981). The complete amino-acid sequence of hen ovalbumin. European Journal of Biochemistry, 115, 335–345. Palika, R., Mashurabad, P.C., Kilari, S., Kasula, S., Nair, K.M., & Pullakhandam, R. (2013). Citric acid mediates the iron absorption from low molecular weight human milk fractions. Journal of Agriculture and Food Chemistry, 61, 11151–11157. Pullakhandam, R., Nair, K.M., Pamini, H., & Punjal, R. (2011). Bioavailability of iron and zinc from multiple micronutrient fortified beverage premixes in Caco-2 cell model. Journal of Food Science, 76, H38–H42. Pullakhandam, R., Palika, R., Vemula, S.R., Polasa, K., & Boindala, S. (2014). Effect of oxytocin injection to milching buffaloes on its content & stability in milk. Indian Journal of Medical Research, 139, 933–939. Rose, M.S., Vahlteich, E.M., & Macleod, G. (1934). Factors in food influencing hemoglobin regeneration III. Eggs in comparison with whole wheat, prepared bran, oatmeal, beef liver, and beef muscle. Journal of Biological Chemistry, 104, 217–229. Sharp, P., & Srai, S.K. (2007). Molecular mechanisms involved in intestinal iron absorption. World Journal of Gastroenterology, 13, 4716–4724. Spiro, T.G., Allerton, S.E., Renner, J., Terzis, A., Bils, R., & Saltman, P. (1966). The hydrolytic polymerization of iron(III). Journal of the American Chemical Society, 88, 2721–2726. Stensballe, A., Andersen, S., & Jensen, O.N. (2001). Characterization of phosphoproteins from electrophoretic gels by nanoscale Fe(III) affinity chromatography with off-line mass spectrometry analysis. Proteomics, 1, 207–222. Storcksdieck, B.S., & Hurrell, R.F. (2007). Iron-binding properties, amino acid composition, and structure of muscle tissue peptides from in vitro digestion of different meat sources. Journal of Food Science, 72, S019–S029. Swain, J.H., Tabatabai, L.B., & Reddy, M.B. (2002). Histidine content of low-molecularweight beef proteins influences non-heme iron bioavailability in Caco-2 cells. Journal of Nutrition, 132, 245–251. Taylor, P.G., Martinez-Torres, C., Romano, E.L., & Layrisse, M. (1986). The effect of cysteine-containing peptides released during meat digestion on iron absorption in humans. American Journal of Clinical Nutrition, 43, 68–71. Zhu, L., Yeung, C.K., Glahn, R.P., & Miller, D.D. (2006). Iron dissociates from the NaFeEDTA complex prior to or during intestinal absorption in rats. Journal of Agriculture and Food Chemistry, 54, 7929–7934.
Contents lists available at ScienceDirect
Food Research International journal homepage: www.elsevier.com/locate/foodres
Characterization of iron-binding phosphopeptide released by gastrointestinal digestion of egg white Ravindranadh Palika a, Purna Chandra Mashurabad a, Madhavan K. Nair a, G. Bhanuprakash Reddy b, Raghu Pullakhandam a,⁎ a b
Biophysics Division, National Institute of Nutrition, Indian Council of Medical Research, Jamai Osmania, Hyderabad 500 007, India Biochemistry Division, National Institute of Nutrition, Indian Council of Medical Research, Jamai Osmania, Hyderabad 500 007, India
a r t i c l e
i n f o
Article history: Received 8 October 2014 Accepted 25 November 2014 Available online 3 December 2014 Chemical compounds studied in this article:: Ferric chloride (PubChem CID: 23673676) Ferrozine (PubChem CID: 34127) α-Cyano-4-hydroxycinnamic acid (PubChem CID: 5328791) 2-(N-Morpholino) ethanesulfonic acid sodium salt (PubChem CID: 78165) Acetonitrile (PubChem CID: 6342) Ascorbic acid (PubChem CID: 54670067)
a b s t r a c t Binding and solubilization of ferric iron by food peptides, released during digestion, facilitate intestinal iron absorption. In the present study, we investigated the release of iron-binding peptides during in vitro gastrointestinal digestion of chicken (Gallus gallus) egg white. The iron-binding activity of the egg white protein increased upon gastrointestinal digestion. The iron-binding fraction of egg white digesta was purified by gel filtration chromatography followed by reverse phase HPLC. Subsequently, this fraction was identified as an internal fragment of ovalbumin (DKLPGFGDS(PO)4IEAQ, 61–73 residues, GenBank AAB59956.1) by MALDI-MS/MS followed by de novo sequencing. The synthetic peptide corresponding to the identified iron-binding peptide sequence bound and increased the 59Fe-iron uptake. Further, the synthetic peptide also stimulated the iron-induced ferritin synthesis in intestinal Caco-2 cells. While, dephosphorylation of synthetic peptide completely inhibited the iron-binding activity, methyl-esterification of its carboxyl groups partially inhibited the activity. These results suggest that food derived peptides modulate intestinal iron absorption and that the isolated iron-binding egg peptide could be a potential nutraceutical for improving iron absorption. © 2014 Elsevier Ltd. All rights reserved.
Keywords: Bioavailability Caco-2 cells De novo sequencing Iron-binding peptides In vitro digestion
1. Introduction Iron exists in ferric (Fe3+) and ferrous (Fe2+) forms in nature. Dietary iron is available mainly in two forms, the heme- and nonheme iron. Heme-iron is better absorbed than non-heme iron (Baker, Anderson, & Baker, 2003). Dietary non-heme iron present predominantly in ferric form is reduced by duodenal cytochrome-B (DcytB) prior to its intestinal absorption via divalent metal ion transporter 1 (DMT1) (Sharp & Srai, 2007). The intestinal absorption of ferric-iron is limited by its poor solubility at near neutral pH, and chelation improves its solubility and intestinal absorption (Hurrell et al., 2004; Palika et al., 2013). Phytic acid (inositol-hexaphosphate, IP6) and polyphenols, the major secondary metabolites of plant foods, are potent inhibitors of non-heme iron absorption (Hurrell, 2002; Nair & Iyengar, 2009). Therefore, the observed high prevalence of iron deficiency anemia in ⁎ Corresponding author at: Biophysics Division, National Institute of Nutrition, Jamai Osmania, Hyderabad 500 604, India. Tel.: +91 40 27197323; fax: +91 40 27019074. E-mail address: [email protected] (R. Pullakhandam).
http://dx.doi.org/10.1016/j.foodres.2014.11.049 0963-9969/© 2014 Elsevier Ltd. All rights reserved.
vegetarians could be due to poor density and bioavailability of iron (Nair & Iyengar, 2009). Therapeutic iron supplementation and food fortification are potential strategies to prevent iron deficiency anemia (Hurrell, 2002; Hurrell et al., 2004; Nair & Iyengar, 2009). However, oral iron therapy is associated with free radical damage, and the addition of iron to food results in unacceptable organoleptic properties due to the pro-oxidant nature of iron (Casanueva & Viteri, 2003; Douglas, Rainey, Wong, Edmonson, & LaCroix, 1981). Although more inert (encapsulated) or chelated forms of iron increase the shelf life of products, the iron bioavailability from such sources remains a serious concern (Hurrell et al., 2004). Therefore, alternate strategies such as selection of iron fortificants with higher iron solubility/bioavailability or addition of components that increase iron solubility and hence absorption are needed. Studies over the past two decades have demonstrated that iron bioavailability from protein rich foods, particularly from animal sources, is high (Cook & Monsen, 1976; Hurrell, 2002; Hurrell et al., 2004; Nair & Iyengar, 2009). Histidine or cysteine-rich peptides released during the
R. Palika et al. / Food Research International 67 (2015) 308–314
digestion of dietary protein are reported to stimulate intestinal iron absorption (Hurrell, Reddy, Juillerat, & Cook, 2006; Storcksdieck & Hurrell, 2007; Swain, Tabatabai, & Reddy, 2002; Taylor, Martinez-Torres, Romano, & Layrisse, 1986). Indeed, casein phosphopeptides (CPP) isolated from in vitro digests of milk have been demonstrated to improve mineral bioavailability including that of iron (Bouhallab & Bougle, 2004). Ironbinding peptides have also been isolated from soybeans, whey protein and yeast (de la Hoz et al., 2014; Lv et al., 2009). These observations suggest wide-spread occurrence of iron-binding peptides in foods that are released during digestion. However, the quantity of such peptides present or released during digestion is a subject of investigation. Further, the nature of the iron-binding peptides and their precursor proteins remains to be characterized from different food sources. Hence, isolation of these peptides in pure form and characterization of their primary structure shall aid immensely in understanding the nature of iron-binding food peptides and associated mineral binding mechanisms. Eggs, a rich source of protein, are widely consumed. Peptides with antiviral, anti-hypertensive, anti-microbial, anti-cancer and antioxidant activities that have been identified in the enzymatic hydrolysates of egg proteins (Mine, 2007). Studies in humans and animal models have reported the inhibitory effect of eggs on iron bioavailability (Grengard, Sentenac, & Mendelsohn, 1964; Rose, Vahlteich, & Macleod, 1934). The egg yolk phosphovitin, an extensively phosphorylated protein, binds iron and prevents its release during gastric and intestinal phases of digestion (Grengard et al., 1964). However, iron-binding peptides if any, released during the digestion of egg white proteins and their effect on intestinal iron absorption are not studied yet. Recently Abeyrathne, Lee, Jo, Nam, and Ahn (2014) reported the release of iron and copper binding peptides during enzymatic digestion of ovalbumin. However, the nature of the specific peptides involved in mineral binding and associated mechanisms remains to be studied. The present study examined the effect of gastric and intestinal digestion of egg white protein on the iron-binding capacity and attempted to decipher the molecular identity of specific peptide(s) involved in ironbinding. Further, the effect of the isolated iron-binding peptide on iron uptake and iron induced ferritin synthesis was also assessed in intestinal cells. 2. Material and methods 2.1. Materials Porcine pepsin (Cat#P7012), pancreatin (Cat#P7545) and all other chemicals/reagents were obtained from Sigma Chemical Co. (Bangalore, India), unless otherwise specified. 59FeCl3 (carrier free) was obtained from Board of Radiation and Isotope Technology (BRIT), Mumbai, India. Chicken eggs were purchased from the local market. Phosphorylated synthetic peptide (DKLPGFGDS(PO)4IEAQ) was obtained from GenScript (NJ, USA) and its purity (N 98%) was confirmed by MALDI-MS. 2.2. In vitro digestion and isolation of iron-binding peptides 2.2.1. Simulated gastrointestinal digestion Chicken eggs (Gallus gallus, 4 Nos) were boiled in water for 20 min and the solid whites were collected, and washed with Milli Q water. The egg whites (112 g) were cut into small pieces, suspended in 250 mL normal saline and homogenized in a kitchen blender for 2 min at the maximum speed (18,000 rpm). The simulated in vitro gastric and intestinal digestion was carried out as described previously (Glahn, Lee, Yeung, Goldman, & Miller, 1998; Pullakhandam, Nair, Pamini, & Punjal, 2011), except that the digestion was carried out for 6 h to ensure complete digestion of the egg white protein. Briefly, the pH of the egg white suspension was adjusted to 2 with 6 N HCl followed by the addition of 1 g of porcine pepsin (≥ 2500 units/mg protein, dissolved in 10 mL of 1 N HCl). The pepsin digestion was carried out
309
for 6 h at 37 °C in a shaking water bath. At the end of this step, the pH of sample was raised to 6.5 with 2 M NaHCO3 followed by the addition of 1 g of porcine pancreatin (8X USP equivalents, dissolved in 0.01 mM NaHCO3). The samples were incubated at 37 °C for 6 h to complete digestion. During different stages of digestion, an aliquot of the digestion mixture was drawn for testing the iron solubilization activity as described below. At the end of simulated gastrointestinal digestion, the samples were clarified by centrifugation at 10,000 g for 15 min at 4 °C and the supernatant is referred to as the ‘digesta’. 2.2.2. Ultrafiltration The digesta was subjected to ultrafiltration through 5 kDa cut-off centrifugal filters (Millipore) to collect b 5 kDa peptides in the filtrate (5kF) and N5 kDa proteins/peptides in the retentate (5kR). The 5kF and 5kR fractions were frozen and stored at − 20 °C, until further analysis. 2.2.3. Solid phase extraction Solid phase extraction was performed on an octadecyl-silica (C-18) matrix (Discovery DSC 18, Sigma). Briefly, the glass column consisting 5 g of C-18 matrix was equilibrated with 2% acetonitrile (v/v) containing 0.1% trifluoroacetic acid (TFA). The 5kF fraction (10 mL containing 0.1% TFA) was loaded on the column, washed with 100 mL of 2% acetonitrile to remove unbound proteins. The bound proteins were then eluted with 10 mL of 50% acetonitrile, concentrated in a vacuum evaporator (Vacufuge Plus, Eppendorf). The dried fractions were stored at −20 °C, until further analysis. They were reconstituted in saline (0.9% NaCl, containing 0.1% TFA) prior to the analysis. 2.2.4. Iron solubilization assay Ferric iron is not soluble at neutral pH unless it is reduced to ferrous form or is bound to soluble components. Therefore, estimation of soluble iron in the absence of reduction provides a direct measure of ironbinding activity. The ferric iron solubilization assay was performed as described previously (Palika et al., 2013). Briefly, 100 μL aliquots of the digesta, 5kF, 5kR, C-18 bound and flow-through fractions or synthetic peptide (0 to 50 μM) were diluted with 50 mM 2-(N-Morpholino) ethanesulfonic acid sodium salt (MES) buffer pH 6.5 to 1 mL and supplemented with 25 μM FeCl3 (traced with 50 nCi 59FeCl3), and incubated for 30 min at 37 °C. At the end of incubation, samples were centrifuged at 10,000 g at 4 °C for 15 min and the supernatant solution (800 μL) was mixed with 5 mL of Bray's mixture and counted in liquid scintillation counter (PerkinElmer; TRICARB 2900TR). The percent binding of iron was calculated considering the solubility of 59Fe iron in 6 N HCl as 100%. 2.2.5. Ferric iron reduction assay The reduction of ferric iron was measured as described previously (Palika et al., 2013). Briefly, the reaction mixture (500 μL) contained egg peptide fractions (100 μL), 25 μM FeCl3 and 500 μM ferrozine (a chromogen for ferrous iron) in 50 mM MES pH 6.5. The reaction mixture was incubated for 120 min at 37 °C, and the absorbance was measured at 562 nm in a micro-plate reader (BioTek; Powerwave HT-1). 2.2.6. Gel filtration chromatography A superdex-peptide (10/30, HR) column (Amersham Biosciences) connected to Akta-purifier module (Amersham Biosciences) was equilibrated with 5 column volumes of normal saline. The C-18 bound fraction reconstituted in 500 μL of saline containing 0.1% TFA was loaded on the above column and eluted with the equilibration buffer at a flow rate of 0.5 mL/min, while monitoring the absorption at 215 nm. A total of 29 fractions (1 mL each) were collected and used immediately in the iron solubilization assay. The fractions with iron solubilizing activity were concentrated by solid phase extraction on C-18 column as described above.
310
R. Palika et al. / Food Research International 67 (2015) 308–314
2.2.7. Reverse phase (RP)-HPLC Reverse phase chromatography of peptides was performed on a C-18 column (Thermo-Hypersil ODS, 5 μ, 250 × 4.6 mm) connected to a HPLC equipped with an online UV monitor and controlled by the Chemstation software (Agilent 1100, Pala Alto, CA, USA). The column was eluted with a binary gradient of solvent A (2% acetonitrile (v/v), 0.1% TFA) from 0 to 5 min followed by linear gradient of 50% solvent B (acetonitrile, 0.1% TFA) from 5 to 25 min at a flow rate of 1 mL/min and the eluent was monitored at 215 nm. The individual peaks were collected, concentrated and tested for iron-binding activity as described above.
2.4.2. 59Fe uptake in Caco-2 cells Ferric chloride (2.5 mM stock in 10 mM HCl) was diluted to a final concentration of 25 μM (traced with 0.5 μCi of 59Fe) with MES buffer (50 mM, pH 6.5) in the absence (blank) or presence of ascorbic acid (250 μM), 0, 1:0.5, 1:1 and 1:2 molar ratios of iron to synthetic egg peptide, and then fed to the differentiated Caco-2 cells for a period of 2 h. After incubation the monolayers were washed thrice with ice-cold phosphate buffer saline containing 10 mM bathophenanthroline (to remove non-specifically bound iron), harvested by scraping, and counted in a liquid scintillation counter (PerkinElmer, USA). The percent uptake was calculated considering the uptake with ascorbic acid as 100%.
2.3. Characterization of iron-binding peptide
2.4.3. Ferritin expression in Caco-2 cells The coupled in vitro digestion/Caco-2 cell model was used for measuring the ferritin induction as described previously (Glahn et al., 1998; Pullakhandam et al., 2011). Briefly, 25 μM FeCl3 in the presence and absence of ascorbic acid (1:10 molar ratio) and egg peptide (1:1) was subjected to in vitro gastric and intestinal digestion, followed by feeding the digesta to differentiated Caco-2 cells for a period of 24 h. At the end of incubation, the cells were washed, lysed and ferritin content in the cell lysate was estimated as described previously (Pullakhandam et al., 2011). A human ferritin sandwich ELISA method developed in-house and validated against recombinant ferritin (94/ 572, NIBSC, UK) was used for ferritin estimation in cell lysates. Briefly, ferritin content was estimated in 5 μL of Caco-2 cell lysates using human liver ferritin Ig-G conjugated to horse-radish peroxidase and the substrate system orthophenylenediamine-H2O2. The color intensity was measured using an ELISA plate reader (BioTek, Powerwave HT-1).
2.3.1. Matrix assisted laser desorption ionization–time of flight (MALDI–TOF) spectrometry The dried HPLC fraction with iron solubilizing activity was reconstituted in 20 μL of 2% acetonitrile (with 0.1% TFA). An aliquot (1 μL) of this sample was mixed with equal volume of 50% acetonitrile, 0.1% TFA containing 5 mg/ml α-cyano-4-hydroxycinnamic acid (CHCA), spotted on a stainless steel target plate and dried. All MS and MS/MS spectra were acquired on an ABI 5800 TOF–TOF (AB Sciex, Foster City, CA, USA) instrument that was equipped with a nitrogen laser and operated in a positive-ion delayed extraction reflector mode. Usually, 250 individual spectra of each spot were averaged to produce a mass spectrum. The MS/MS spectra were acquired in 1 kV positive ion mode. Peptide identification was performed by the ProteinPilot™ Software (AB Sciex, USA) against the G. gallus proteins from SwissProt databank (www.expasy.org) using the following search criteria: oxidation of methionine as variable modification and phosphorylation of serine/threonine as fixed modifications, no proteolytic enzyme specified and mass accuracy tolerance was set at 50 ppm for parent and 0.5 Da for fragments.
2.3.2. Dephosphorylation To test the effect of dephosphorylation on iron-binding activity, the synthetic peptide (200 μg in 50 mM ammonium bicarbonate buffer) was incubated in the presence and absence of alkaline phosphatase (ALP, 100 units) for a period of 12 h at 37 °C. At the end of incubation the iron-binding activity was assessed as described above.
2.3.3. Methyl-esterification Methyl-esterification of carboxyl groups of the peptide was performed as described previously (Krishnamurthy et al., 1989). Briefly, peptide (200 μg) was dried and incubated in methanol or anhydrous methanolic-HCl (1.25 M HCl) for a period of 12 h at room temperature. At the end of incubation the solvent was dried in a centrifugal evaporator (Vacufuge Plus, Eppendorf, USA), the dried peptide dissolved in 200 μL of Milli Q water (0.1% TFA) and tested for the iron-binding activity as described above.
2.4. Effect of iron-binding peptide on iron absorption in intestinal cells 2.4.1. Caco-2 cell culture Caco-2 cells were obtained from National Centre for Cell Sciences (Pune, India). Cells were seeded at a density of 50,000 cells/cm2 in 6-well plates (Corning, India). The cells were grown in Dulbecco's modified Eagle's medium with 10% fetal bovine serum (FBS), 1% nonessential amino acids, 2 mM glutamine and 1% antibiotic-antimycotic solution. The cells were maintained at 37 °C in an incubator with a 5% CO2/95% air atmosphere at constant humidity. Once the cells reached confluence, the spent media was changed every alternate day. The cells were used for iron uptake experiments 12–14 days post confluence as described previously (Palika et al., 2013; Pullakhandam et al., 2011).
2.5. Statistics All the experiments were performed in triplicates and repeated at least twice to generate 6 observations. Mean and SD were calculated using Microsoft Excel and the data was analyzed using one-way ANOVA, followed by the post hoc least significant differences (LSD) test, using SPSS software (Version 11.0). The results were considered significant if the P b 0.05. 3. Results 3.1. Ferric iron solubilizing activity of egg peptide fractions The iron solubility with egg white protein increased from 17% ± 2 to 42% ± 4.3 as a function of time up to 100 min of pepsin digestion (Fig. 1A). The addition of pancreatin further increased the iron solubility, reaching maximum (102% ± 6.3) between 100–150 min (Fig. 1A). However, no reduction of ferric iron was observed during digestion with either pepsin or pancreatin (data not shown), implying that the increased iron solubilization could be due to binding. The iron-binding activity was higher and similar in the presence of acid or egg protein digesta compared to blank (Fig. 1B). Further, iron-binding activity was significantly higher in 5kF (108% ± 5.6) than 5kR (55.2% ± 6.3) (Fig. 1B). Also the iron-binding activity of C-18 bound fraction (106% ± 5.4) was significantly higher than that of the flow through fraction (36% ± 4.2, Fig. 1B). 3.2. Purification and characterization of iron-binding peptide from egg white digesta The gel filtration chromatography of C-18 bound egg peptide fraction showed 2 major peaks and few minor peaks between 20–60 min (Fig. 2A). Among all the fractions tested, the iron-binding activity was found exclusively in a minor peak eluting between 26–28 min (corresponding approximately to 2000 Da mass; marked with * in Fig. 2A), whereas in the HPLC analysis, the iron-binding activity was found exclusively in the peak eluting at 16.2 min (Fig. 2B).
R. Palika et al. / Food Research International 67 (2015) 308–314
311
Fig. 1. Effect of gastrointestinal digestion on iron solubilizing activity of egg white proteins. Panel A — Iron solubilizing activity of egg white protein digested with pepsin and pancreatin was assessed as a function of time. Panel B — Ferric iron solubilizing activity of egg white digesta during fractionation and purification. The % solubility was calculated considering the solubility of iron in 6 N HCl as 100%. The data represent mean + SD and bars with different superscripts are significantly different (P b 0.05).
3.3. MALDI-MS and de novo sequencing of iron-binding peptide
3.4. Mechanism of iron-binding
The MALDI-MS spectra of iron-binding peptide fraction isolated by RP-HPLC showed a major peak at 1456.74 amu apart from its Na + K adduct showing the peak at 1516.66 amu (Fig. 3A). Interestingly, a peak with a mass of 1358.75 amu (− 98 amu loss) was observed in the mass spectrum, indicating phosphorylated serine or threonine residues. MALDI-MS/MS followed by de novo sequencing analysis identified the peptide as DKLPGFGDS(PO)4IEAQ (Fig. 3B), which is an exact match to the internal fragment of ovalbumin (residues 61–73, GenBank AAB59956.1) with a mascot score of 190 and fragment ion score of 75% (please see supplementary data for de novo sequencing/data base search results).
In order to confirm the iron-binding activity of the egg peptide, further experiments were conducted with the synthetic peptide. The iron-binding activity of the synthetic peptide increased dose dependently and saturated at 1:1 molar ratio of iron to peptide (Fig. 4A). Dephosphorylation of the synthetic peptide with alkaline phosphatase completely inhibited (N92% inhibition) and methyl-esterification of acidic amino acids (aspartic acid and glutamic acid) partially inhibited (40%) its iron-binding activity (Fig. 4A). The molecular model of the isolated egg-phosphopeptide generated from the PDB (ID: 1OVA) crystal structure co-ordinates revealed the extension of serine-phosphate and asp-carboxyl groups outward of the peptide back bone in the similar plane and with a proximity of 5.7 A° (Fig. 4B). 3.5. Effect of synthetic egg peptide on iron absorption and ferritin expression in Caco-2 cells Compared to control (17% ± 3.9), ascorbic acid (1:10 molar ratio) markedly increased (100% ± 5.9) 59Fe uptake in differentiated Caco-2 cells. 59Fe uptake also increased dose dependently in the presence of 1:0.5 (35% ± 5.5) and 1:1 (80 ± 6.9) ratio of iron to synthetic peptide compared to control. However, 59Fe uptake remained unchanged beyond 1:1 molar ratio of iron to peptide (Fig. 5A). Although, 59Fe uptake was significantly inhibited upon dephosphorylation or methylesterification of the synthetic peptide (Fig. 5A), inhibition was significantly higher upon dephosphorylation than methyl-esterification. The ferritin content of Caco-2 cells was similar in the absence or presence of ferric iron. Interestingly, the addition of ascorbic acid (1:10 molar ratio) or the synthetic peptide (1:1 molar ratio) significantly induced the ferritin synthesis in Caco-2 cells, compared to controls albeit, the extent of ferritin induction by ascorbic acid was significantly higher than that of the synthetic peptide (Fig. 5B). 4. Discussion
Fig. 2. Purification of iron-binding peptide released during the digestion of egg white proteins. The egg white peptides released during the gastrointestinal digestion were isolated and enriched by ultrafiltration followed by solid phase extraction. Panel A — An aliquot of this peptide sample was fractionated on Superdex-peptide column as described in the Material and methods section. Panel B — The fractions showing iron-binding activity were further fractionated on RP-HPLC as described in the Material and methods section. The peaks marked with * indicate iron-binding activity.
Bioavailability of iron is reported to be high from protein rich flesh foods. It is believed that cryptic peptides released during the gastrointestinal digestion of food proteins bind and facilitate intestinal iron absorption. In the present study, we have isolated and characterized an iron-binding peptide released during gastrointestinal digestion of egg white proteins. The isolated peptide was identified as an internal fragment of ovalbumin corresponding to the amino acids 61–73, with a phosphoserine at position 69 (DKLPGFGDS(PO4)IEAQ). Further, we substantiated these observations using a synthetic peptide corresponding to this sequence. The synthetic peptide also solubilized and increased the iron absorption in intestinal Caco-2 cells. In addition, we
312
R. Palika et al. / Food Research International 67 (2015) 308–314
Fig. 3. MALDI-MS/MS and de novo sequence analysis of egg white iron-binding peptide. The HPLC peak fraction showing the iron-binding activity was subjected to MALDI-MS and MS/MS analysis. Panel A — MS spectrum of peptide in the positive ion mode. Panel B — Sequence annotation of MS/MS spectrum.
demonstrated that phosphorylation of serine is a prerequisite for ironbinding activity, while the acidic amino acids also play a key role. Pepsin and pancreatin digests but not the total egg protein homogenate significantly increased the solubility of iron at neutral pH (Fig. 1A). Further, there was no reduction of ferric iron with these fractions, indicating that the ferric iron solubilization was indeed due to the binding of specific peptides to ferric iron. Enzymatic hydrolysis of variety of food sources has been reported to release peptides with varied
biological activities such as mineral binding, inhibition of angiotensin-1-converting enzyme (ACE) and antioxidant activity (Mine, 2007; Moller, Scholz-Ahrens, Roos, & Schrezenmeir, 2008). Further, enzymatic hydrolysis of milk protein, chicken and beef had been reported to release iron-binding peptides (Bouhallab & Bougle, 2004; de la Hoz et al., 2014; Hurrell et al., 2006; Lv et al., 2009; Storcksdieck & Hurrell, 2007; Swain et al., 2002; Taylor et al., 1986). It is also known that binding of proteins or peptides
Fig. 4. Effect of dephosphorylation and methyl-esterification on iron solubilization activity of synthetic egg peptide: Panel A — The ferric iron (25 μM) was incubated with synthetic peptide (25 μM) treated with alkaline phosphatase (ALP) or anhydrous methanolic-HCl in 50 mM MES buffer pH 6.5. The iron solubility was assessed as described in the Material and methods section. The % iron solubility was computed considering the iron solubility with peptide as 100%. The data represent mean + SD and the bars that do not share common superscript differ significantly (P b 0.01). Panel B — The molecular model of the peptide was generated using the crystal structure co-ordinates of ovalbumin (PDB: 1OVA) corresponding to 61–73 amino acid residues and the model was visualized using Discovery Studio Visualizer (Version 3.1) to demonstrate the extension of phosphate group of Ser-69 and carboxyl group of Asp-68.
R. Palika et al. / Food Research International 67 (2015) 308–314
313
Fig. 5. Effect of synthetic peptide on 59Fe iron uptake and iron induced ferritin synthesis in differentiated Caco-2 cells: Panel A — Ferric chloride (25 μM, traced with 0.5 μCi of 59Fe) was fed to Caco-2 cells in the presence of ascorbic acid (Vit C) or synthetic egg peptide (Pep) for a period of 2 h. The radioactivity associated with the cells was assessed as described in the Material and methods section. The % uptake was computed considering the uptake with Vit C as 100%. Panel B — Iron, Vit C and peptide mixtures were subjected to simulated in vitro digestion and fed to the Caco-2 cells for a period of 24 h as described in the Material and methods section. Iron induced ferritin expression was estimated by ELISA method. The data represent mean + SD and bars that do not share common superscript differ significantly (P b 0.05).
solubilizes the otherwise insoluble ferric iron at neutral pH. For instance, binding of transferrin and lactoferrin facilitates the solubilization and transport of ferric iron in serum and milk, respectively (Baker et al., 2003). It is therefore likely that peptides generated during the digestion of egg white protein bound and solubilized the ferric iron. Higher iron-binding activity in 5kF than 5kR fraction indicates the low molecular weight (b5 kDa) nature of the ironbinding peptide released during the digestion of egg white protein. It thus appears that the residual iron-binding activity in 5kR fractions could be because of incomplete filtration or the presence of other high molecular weight (N5 kDa) iron-binding peptides. Further, the much higher protein content of 5kR fraction (12 times higher than 5kF) could also contribute to the residual iron-binding activity. Interestingly, the iron solubilizing activity of 5kF fraction was predominant in C-18 bound fraction, implying the hydrophobic nature of iron-binding peptide(s). The low molecular weight nature of the iron-binding fraction is indeed corroborated by gel filtration chromatography of the C-18 bound fraction. Further, RP-HPLC analysis led to the elution of iron-binding activity in a single peak probably suggesting that a specific peptide was involved in iron binding. MALDI-MS analysis followed by de novo sequencing of iron-binding fraction obtained from the RP-HPLC identified that the peptide corresponds to amino acid residues 61–73 of ovalbumin with a phosphoserine at 69th position. In agreement with these results, phosphorylation of Ser-69 was reported in ovalbumin previously (Kinoshita-Kikuta, Kinoshita, & Koike, 2012; Nisbet, Saundry, Moir, Fothergill, & Fothergill, 1981). It was interesting that the synthetic peptide corresponding to this amino acid sequence also solubilized the ferric iron in a dose dependent manner and the solubility was saturated at equimolar concentration of iron and peptide, implying one iron-binding site on the peptide. These results together confirm the identity of primary structure of the peptide and its parent protein. Previous studies reported that phosphopeptides and peptide fractions rich in acidic amino acids bind to metal ions including iron (Stensballe, Andersen, & Jensen, 2001; Storcksdieck & Hurrell, 2007). In line with these reports, dephosphorylation of the synthetic peptide led to the complete inhibition of iron solubilizing activity, while methyl-esterification of carboxyl groups inhibited the activity partially. In agreement with these results, dephosphorylation of casein phosphopeptides is reported to inhibit their Ca binding activity completely (Ferraretto, Gravaghi, Fiorilli, & Tettamanti, 2003). Further, acidic residues in the CPP are predicted to be involved in the stabilization of Ca-phosphate complexes in solution (Cross, Huq, Palamara, Perich, & Reynolds, 2005). Indeed, peptide fractions rich in acidic amino acids are also reported to bind iron in enzymatic hydrolysate of meat and shrimp (Huang, Ren, & Jiang, 2011; Hurrell et al., 2006; Storcksdieck & Hurrell, 2007). In aqueous solution, ferric iron undergoes hydrolytic
polymerization to form insoluble precipitates (Spiro et al., 1966). Since ferric iron is a strong lewis acid, ligation of strong bases such as phosphate, carboxylate and amine functional groups prevents the hydrolytic polymerization and precipitation. It is therefore possible that ligation of peptide phosphoseryl group with ferric iron might prevent its polymerization resulting in solubilization of ferric iron at neutral pH. Further, carboxylate groups of aspartic acid flanking the phosphoserine residue might stabilize the iron–phosphate interaction. Interestingly, molecular model generated using the crystal structure coordinates of ovalbumin showed extension of phosphate and carboxyl group of adjacent aspartic acid from the peptide back bone on a similar plane, supporting the role of both phosphate and carboxylate groups in iron-binding. Nevertheless, further studies are needed to understand the exact role of carboxyl groups in iron binding. The solubility and oxidation state of iron are two major determinants of intestinal iron absorption (Sharp & Srai, 2007). The higher absorption of ferrous iron compared to ferric iron could be explained by its higher solubility at intestinal pH. The solubility of iron salt is positively associated with its absorption by the enterocyte, regardless of the oxidation state (Hurrell et al., 2004). For instance, iron absorption from a soluble ferric iron compound (NaFe (III) EDTA) is similar to that of ferrous iron salt (Hurrell et al., 2004). The addition of either ascorbic acid or peptide significantly increased the uptake of 59Fe in differentiated Caco-2 cells. However, the extent of increase in iron uptake was significantly higher with ascorbic acid, consistent with its ferric iron reducing and solubilizing activity. The fact that the peptide only solubilized but not reduced the ferric iron is in agreement with the iron uptake in intestinal cells. Previously we and others have shown that EDTA and citric acid induced iron uptake in intestinal cells requires the prior reduction of ferric iron and that DcytB could be involved in reducing chelated ferric iron (Palika et al., 2013; Zhu, Yeung, Glahn, & Miller, 2006). It is therefore possible that peptide binding increases the accessibility of ferric iron to the DcytB, for reduction and subsequent uptake by Caco-2 cells. Further, our finding that dephosphorylation of the peptide completely inhibited the 59Fe uptake, while methyl-esterification partially inhibited the uptake, appears to support the role of binding/solubilizing activity in stimulating intestinal iron uptake. The iron absorption stimulating activity of the isolated peptide in vivo depends on its stability during gastrointestinal digestion. Although the peptide was isolated after exhaustive digestion of egg white protein, in silico digestion analysis of the peptide sequence (DKLPGFGDS(PO)4IEAQ) using MS-Digest software indeed identified several proteolytic cleavage sites for pepsin and trypsin (data not shown). In order to rule out further digestion of the peptide, we studied the bioavailability of iron using coupled in vitro digestion/Caco-2 cell model (Glahn et al., 1998; Pullakhandam et al., 2011). The ferritin content of
314
R. Palika et al. / Food Research International 67 (2015) 308–314
Caco-2 cells exposed to ferric iron is similar to that of control suggesting a poor absorption of ferric iron in intestinal cells. However, ferritin content of the cells exposed to ferric iron in the presence of either ascorbic acid or peptide was significantly higher compared to control. Similar to that of iron uptake, the ferritin induction was also higher in the presence of ascorbic acid than the peptide, which in turn is due to reduction of iron by ascorbic acid but not by peptide. In agreement with these results, extensive digestion of the synthetic peptide with pepsin, trypsin and chymotrypsin for a period of 24 h also did not influence its ironbinding activity (data not shown). These results suggest that the isolated peptide was resistant to further digestion by gastric and intestinal proteases. Recently, we have demonstrated that oxytocin, a peptide hormone; though possesses specific proteolytic cleavage sites for pepsin, is resistant to peptic digestion unless its internal disulfide bond is oxidized (Pullakhandam, Palika, Vemula, Polasa, & Boindala, 2014). Thus it appears likely that subtle structural constraints in the peptide may render it resistant to digestion. To the best of our knowledge this is the first systematic study wherein the iron-binding peptide released during the digestion of food protein is isolated, characterized and proof of concept for its activity is provided with its synthetic counterpart. These results strengthen the hypothesis that food derived peptides modulate intestinal iron absorption and that the isolated iron-binding egg peptide could be exploited as a potential nutraceutical for improving iron nutrition. Acknowledgments Supported by Department of Biotechnology (BT/PR5228/FNS/20/ 549/2012), Govt. of India. Ravi and Purna are supported by a Junior Research Fellowship from University Grants Commission, Govt. of India. We thank Dr. P. Ravinder for the assistance with ferritin ELISA and Mr. K. Suresh for the technical assistance with in vitro digestion. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.foodres.2014.11.049. References Abeyrathne, E.D., Lee, H.Y., Jo, C., Nam, K.C., & Ahn, D.U. (2014). Enzymatic hydrolysis of ovalbumin and the functional properties of the hydrolysates. Poultry Science, 93, 2678–2686. Baker, H.M., Anderson, B.F., & Baker, E.N. (2003). Dealing with iron: Common structural principles in proteins that transport iron and heme. Proceedings of the National Academy of Sciences of the United States of America, 100, 3579–3583. Bouhallab, S., & Bougle, D. (2004). Biopeptides of milk: Caseinophosphopeptides and mineral bioavailability. Reproductive Nutrition Development, 44, 493–498. Casanueva, E., & Viteri, F.E. (2003). Iron and oxidative stress in pregnancy. Journal of Nutrition, 133, 1700S–1708S. Cook, J.D., & Monsen, E.R. (1976). Food iron absorption in human subjects. III. Comparison of the effect of animal proteins on nonheme iron absorption. American Journal of Clinical Nutrition, 29, 859–867. Cross, K.J., Huq, N.L., Palamara, J.E., Perich, J.W., & Reynolds, E.C. (2005). Physicochemical characterization of casein phosphopeptide–amorphous calcium phosphate nanocomplexes. Journal of Biological Chemistry, 280, 15362. de la Hoz, L., Ponezi, A.N., Milani, R.F., Nunes da Silva, V.S., Sonia de Souza, A., & BertoldoPacheco, M.T. (2014). Iron-binding properties of sugar cane yeast peptides. Food Chemistry, 142, 166–169.
Douglas, F.W., Rainey, N.H., Wong, N.P., Edmonson, L.F., & LaCroix, D.E. (1981). Color, flavor, and iron bioavailability in iron-fortified chocolate milk. Journal of Dairy Science, 64, 1785–1793. Ferraretto, A., Gravaghi, C., Fiorilli, A., & Tettamanti, G. (2003). Casein-derived bioactive phosphopeptides: Role of phosphorylation and primary structure in promoting calcium uptake by HT-29 tumor cells. FEBS Letters, 551, 92–98. Glahn, R.P., Lee, O.A., Yeung, A., Goldman, M.I., & Miller, D.D. (1998). Caco-2 cell ferritin formation predicts nonradiolabeled food iron availability in an in vitro digestion/ Caco-2 cell culture model. Journal of Nutrition, 128, 1555–1561. Grengard, O., Sentenac, A., & Mendelsohn, N. (1964). Phosvitin, the iron carrier of egg yolk. Biochimica Biophysica Acta, 90, 406–407. Huang, G., Ren, Z., & Jiang, J. (2011). Separation of iron-binding peptides from shrimp processing by-products hydrolysates. Food and Bioprocess Technology, 8, 1527–1532. Hurrell, R.F. (2002). Fortification: Overcoming technical and practical barriers. Journal of Nutrition, 132, 806S–812S. Hurrell, R.F., Lynch, S., Bothwell, T., Cori, H., Glahn, R., Hertrampf, E., et al. (2004). Enhancing the absorption of fortification iron. A SUSTAIN task force report. International Journal of Vitamin and Nutrition Research, 74, 387–401. Hurrell, R.F., Reddy, M.B., Juillerat, M., & Cook, J.D. (2006). Meat protein fractions enhance nonheme iron absorption in humans. Journal of Nutrition, 136, 2808–2812. Kinoshita-Kikuta, E., Kinoshita, E., & Koike, T. (2012). Separation and identification of four distinct serine-phosphorylation states of ovalbumin by Phos-tag affinity electrophoresis. Electrophoresis, 33, 849–855. Krishnamurthy, T., Szafraniec, L., Hunt, D.F., Shabanowitz, J., Yates, J.R., Hauer, C.R., et al. (1989). Structural characterization of toxic cyclic peptides from blue-green algae by tandem mass spectrometry. Proceedings of the National Academy of Sciences of the United States of America, 86, 770–774. Lv, Y., Liu, Q., Bao, X., Tang, W., Yang, B., & Guo, S. (2009). Identification and characteristics of iron-chelating peptides from soybean protein hydrolysates using IMAC-Fe3+. Journal of Agriculture and Food Chemistry, 57, 4593–4597. Mine, Y. (2007). Egg proteins and peptides in human health—Chemistry, bioactivity and production. Current Pharmaceutical Design, 13, 875–884. Moller, N.P., Scholz-Ahrens, K.E., Roos, N., & Schrezenmeir, J. (2008). Bioactive peptides and proteins from foods: Indication for health effects. European Journal of Nutrition, 47, 171–182. Nair, K.M., & Iyengar, V. (2009). Iron content, bioavailability & factors affecting iron status of Indians. Indian Journal of Medical Research, 130, 634–645. Nisbet, A.D., Saundry, R.H., Moir, A.J., Fothergill, L.A., & Fothergill, J.E. (1981). The complete amino-acid sequence of hen ovalbumin. European Journal of Biochemistry, 115, 335–345. Palika, R., Mashurabad, P.C., Kilari, S., Kasula, S., Nair, K.M., & Pullakhandam, R. (2013). Citric acid mediates the iron absorption from low molecular weight human milk fractions. Journal of Agriculture and Food Chemistry, 61, 11151–11157. Pullakhandam, R., Nair, K.M., Pamini, H., & Punjal, R. (2011). Bioavailability of iron and zinc from multiple micronutrient fortified beverage premixes in Caco-2 cell model. Journal of Food Science, 76, H38–H42. Pullakhandam, R., Palika, R., Vemula, S.R., Polasa, K., & Boindala, S. (2014). Effect of oxytocin injection to milching buffaloes on its content & stability in milk. Indian Journal of Medical Research, 139, 933–939. Rose, M.S., Vahlteich, E.M., & Macleod, G. (1934). Factors in food influencing hemoglobin regeneration III. Eggs in comparison with whole wheat, prepared bran, oatmeal, beef liver, and beef muscle. Journal of Biological Chemistry, 104, 217–229. Sharp, P., & Srai, S.K. (2007). Molecular mechanisms involved in intestinal iron absorption. World Journal of Gastroenterology, 13, 4716–4724. Spiro, T.G., Allerton, S.E., Renner, J., Terzis, A., Bils, R., & Saltman, P. (1966). The hydrolytic polymerization of iron(III). Journal of the American Chemical Society, 88, 2721–2726. Stensballe, A., Andersen, S., & Jensen, O.N. (2001). Characterization of phosphoproteins from electrophoretic gels by nanoscale Fe(III) affinity chromatography with off-line mass spectrometry analysis. Proteomics, 1, 207–222. Storcksdieck, B.S., & Hurrell, R.F. (2007). Iron-binding properties, amino acid composition, and structure of muscle tissue peptides from in vitro digestion of different meat sources. Journal of Food Science, 72, S019–S029. Swain, J.H., Tabatabai, L.B., & Reddy, M.B. (2002). Histidine content of low-molecularweight beef proteins influences non-heme iron bioavailability in Caco-2 cells. Journal of Nutrition, 132, 245–251. Taylor, P.G., Martinez-Torres, C., Romano, E.L., & Layrisse, M. (1986). The effect of cysteine-containing peptides released during meat digestion on iron absorption in humans. American Journal of Clinical Nutrition, 43, 68–71. Zhu, L., Yeung, C.K., Glahn, R.P., & Miller, D.D. (2006). Iron dissociates from the NaFeEDTA complex prior to or during intestinal absorption in rats. Journal of Agriculture and Food Chemistry, 54, 7929–7934.