Article pubs.acs.org/JAFC
Enhancement of Non-heme Iron Absorption by Anchovy (Engraulis japonicus) Muscle Protein Hydrolysate Involves a NanoparticleMediated Mechanism Haohao Wu,† Suqin Zhu,† Mingyong Zeng,*,† Zunying Liu,† Shiyuan Dong,† Yuanhui Zhao,† Hai Huang,§ and Y. Martin Lo# †
College of Food Science and Engineering, Ocean University of China, 5 Yushan Road, Qingdao, Shandong Province 266003, China School of Aquatic Products, Rizhao Polytechnic, No. 16 North of Yantai Road, Rizhao, Shandong Province 276826, China # Department of Nutrition and Food Science, University of Maryland, College Park, Maryland 20742, United States §
ABSTRACT: The mechanisms by which meat enhances human absorption of non-heme iron remain unknown. Recently, anchovy (Engraulis japonicus) muscle protein hydrolysate (AMPH) was found to mediate the formation of nanosized ferric hydrolysis products in vitro. The current paper evaluates the effects of AMPH on the bioavailability and the intestinal speciation of non-heme iron in rats, followed by an investigation of cellular uptake pathways of in vitro-formed AMPH-stabilized nanosized ferric hydrolysis products (ANPs) by polarized human intestinal epithelial (Caco-2) cells. The hemoglobin regeneration efficiencies in anemic rats followed the order ferric citrate (9.79 ± 2.02%) < commercial bare α-Fe2O3 nanoparticles (16.37 ± 6.65%) < mixture of ferric citrate and AMPH (40.33 ± 6.36%) ≈ ferrous sulfate (40.88 ± 7.67%) < ANPs (56.25 ± 11.35%). Percentage contents of intestinal low-molecular-weight iron in the groups of FC+AMPH, FeSO4, and ANPs were significantly lower than the corresponding hemoglobin regeneration efficiencies (P < 0.05), providing strong evidence for the involvement of nanosized iron in intestinal iron absorption from FC+AMPH, FeSO4, and ANPs. Calcein-fluorescence measurements of the labile iron pool of polarized Caco-2 cells revealed the involvement of both divalent transporter 1 and endocytosis in apical uptake of ANPs, with endocytosis dominating at acidic extracellular pH. Overall, AMPH enhancement of non-heme iron absorption involves a nanoparticle-mediated mechanism. KEYWORDS: iron bioavailability, muscle protein, anchovy, rats, Caco-2 cell, nanoparticle
■
INTRODUCTION After birth, humans have to acquire iron from the diet to support growth and compensate for iron loss; over 2 billion people worldwide suffer from iron deficiency. Although dietary iron can be found in the form of heme or non-heme iron, interventions to reduce iron deficiency usually focus on increasing non-heme iron absorption, which is greatly influenced by various dietary components. Inhibitors of nonheme iron absorption in humans include phytates, oxalates, polyphenols, fibers, calcium, copper, and zinc; ascorbic acid and muscle foods (i.e., meat, poultry, and fish) enhance non-heme iron absorption.1 The enhancing effect of muscle foods on nonheme iron absorption is usually attributed to certain “meat factors” (e.g., peptides, acidic polysaccharides, and L-αglycerophosphocholine), which may react with non-heme iron in the intestinal lumen, thereby increasing iron solubilization and absorption.2−4 However, the chemical nature of such soluble non-heme iron has still not been fully elucidated. One economical meat source that could be used to combat iron deficiency in developing countries is the anchovy, a small marine pelagic fish harvested in huge quantity worldwide, mainly for nonfood uses.5 Recently, we demonstrated that anchovy (Engraulis japonicus) muscle protein hydrolysate (AMPH) could mediate the formation of nanosized ferric hydrolysis products within a pH range similar to that of human digestive lumen;6 thus, AMPH might enhance non-heme iron © 2014 American Chemical Society
absorption in humans via a nanoparticle-mediated mechanism. The current research examined the effects of AMPH on the bioavailability and the intestinal speciation of non-heme iron in rats, followed by an investigation of cellular uptake pathways of in vitro-formed AMPH-stabilized nanosized ferric hydrolysis products (ANPs) in polarized human intestinal epithelial (Caco-2) cells.
■
MATERIALS AND METHODS
Materials. Japanese anchovies, E. japonicus (5−8 cm total length), caught in the Yellow Sea by Allen Ship Service Co. Ltd. (Shandong, China) were frozen immediately at −35 °C on board and transported in a Styrofoam-insulated box to the laboratory refrigerator (−20 °C) within 4 h. Trypsin from bovine pancreas with a declared activity of 250 U/mg was purchased from Sinopharm Chemical Co. Ltd. (Shanghai, China). Nano-α-Fe2O3 powder with declared diameters of 20−30 nm was purchased from Beijing Nachen Co. Ltd. (Beijing, China). Bathophenanthrolinedisulfonic acid (BPDS), deferoxamine mesylate (DFB), ethylenediaminetetraacetic acid (EDTA), nitrilotriacetic acid (NTA), phytic acid sodium salt, diethylenetriaminepentaacetic acid (DTPA), trans-1,2-cyclohexanediaminetetraacetic acid (CDTA), sodium selenite, hydrocortisone, and triiodothyronine were obtained from Sigma-Aldrich Co. (St. Louis, MO, USA). Calcein Received: Revised: Accepted: Published: 8632
April 18, 2014 July 27, 2014 July 29, 2014 July 29, 2014 dx.doi.org/10.1021/jf5018719 | J. Agric. Food Chem. 2014, 62, 8632−8639
Journal of Agricultural and Food Chemistry
Article
acetoxymethylester, cell culture mediums, insulin, and epidermal growth factor were purchased from Gibco (Grand Island, NY, USA). All other reagents used in this study were commercially available and of analytical grade. Preparation of AMPH and ANPs. AMPH was obtained as described previously.7 Briefly, 25% (w/v) homogenates of anchovy meat were hydrolyzed by trypsin with a 25:1 substrate to enzyme ratio (w/w) at 37 °C and pH 8 for 4 h with continuous magnetic stirring, followed by heat inactivation of the enzyme in a boiling water bath for 10 min, and the reaction mixture was then centrifuged at 10000g for 20 min. The supernatant was stored at 4 °C for no more than 48 h and filtered through 0.22 μm cellulose acetate filters before use. Total protein concentration in the supernatant was determined according to Lowry’s method using bovine serum albumin as the standard protein.8 ANPs were prepared according to the procedure described by Wu et al.6 with slight modification. Briefly, 1 volume of freshly prepared ferric chloride (20 mM) was added dropwise into 10 volumes of magnetically stirred AMPH solution (20 g peptide/L) with the solution pH maintained at 7.0 ± 0.2 using 1 M sodium hydroxide. The obtained sample solution was stored at 4 °C for no more than 24 h and filtered through 0.22 μm cellulose acetate filters before use. For the structural characterization of ANPs, sample solutions were dropped onto carbon-coated copper grids, allowed to air-dry, and then examined in a JEM 1200 EX (JEOL, Tokyo, Japan) electron microscope. Iron content in the sample solution was determined by flame atomic absorption spectroscopy using a Shimadzu AA-6800 spectrometer (Kyoto, Japan). Animal Experiments. All animal handling was conducted according to the principles and guidelines outlined in the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals and approved by the Ethical Committee of Animal Care and Use at Ocean University of China (Permit 20001013). Sprague− Dawley rats were obtained from Shandong Lukang Pharmaceutical Group Co., Ltd. (Shandong, China). Pelletized purified AIN-93Gbased diets were prepared by TROPHIC Animal Feed High-tech Co., Ltd. (Nantong, China). The Fe-deficient diet contained ∼10 mg/kg iron derived from raw materials. The Fe-sufficient diet contained ∼25 mg/kg iron in the form of ferric citrate (FC) and ∼10 mg/kg iron derived from raw materials. The Fe-repletion diets contained ∼10 mg/ kg iron derived from raw materials and ∼20 mg/kg iron in the forms of ferrous sulfate (FeSO4), ANPs, FC, mixture of FC and AMPH (FC +AMPH), and commercial bare α-Fe2O3 nanoparticles (Nano-αFe2O3). The iron compounds were incorporated into the diets during preparation of the wet ingredient mixtures. FeSO4, FC, or Nano-αFe2O3 equivalent to 20 mg iron was mixed with 1 kg of the other dry ingredients and 200 mL of Milli-Q water to obtain the wet ingredient mixture. To incorporate the solution form of AMPH (18 g peptide/L) and ANPs (1.8 mM iron and 18 g peptide/L), 200 mL of either of them was mixed with 1 kg of the other dry ingredients to make the wet ingredient mixture. After being pelletized, the wet ingredient mixture was dried overnight at 40 °C in a drying oven. Iron contents in the dried pelletized diets were determined by flame atomic absorption spectroscopy using a Shimadzu AA-6800 spectrometer. Rats received Milli-Q water ad libitum throughout the experiments. Body weight was recorded at the beginning and end of each experimental period. The iron bioavailability was measured in anemic rats using a hemoglobin regeneration bioassay.9,10 Female Sprague−Dawley rats (n = 12) breastfeeding male pups (16 days old; n = 52) were housed in individual stainless steel cages with four to five pups each and fed a Fedeficient diet ad libitum for a week. On the weaning day, 52 male rats (23 days old) were transferred into individual stainless steel cages under controlled conditions at temperatures of 22 ± 2 °C and relative humidities of 55 ± 10% with a light/dark cycle of 12/12 h. The justweaned rats were stratified according to body weight and within each stratum were randomly allocated to two groups: the Fe-depleted group (n = 48) fed a Fe-deficient diet ad libitum and a control group (n = 4) fed a Fe-sufficient diet ad libitum. After a Fe-depletion period of 27 days, 20 μL of fresh rat blood sampled by tail vein incision was diluted with EDTA and subsequently analyzed using a Cell-Dyn 1600 hematology analyzer (Abbott Laboratories, Abbott Park, IL, USA).
The mean hemoglobin concentrations of the Fe-depleted group and the control group were 71.1 ± 13.8 and 153.8 ± 13.9 g/L, respectively. Rats in the Fe-depleted group were stratified by the hemoglobin concentrations and randomized into six groups (eight rats each). The rats in each group consumed one of the five Fe-repletion diets containing FeSO4, ANPs, FC, FC+AMPH, or Nano-α-Fe2O3 or the Fe-deficient diet for 13 days (the repletion period) ad libitum. Rats in the control group remained on the Fe-sufficient diet during the repletion period. Periodically, the daily food intake of each rat was manually recorded. A known and adequate amount of a diet was given to a rat in the feeding container between 9:00 and 10:00 a.m. After 24 h, the diet remaining in the feeding container and at the bottom of the cage was reweighed; calculating the differences showed the rat’s daily food intake. At the end of the repletion period, blood samples were once again taken by tail vein incision. The rats were anesthetized with ether before their livers were taken, rinsed with saline, and stored at −20 °C. Iron concentration in the liver was measured by flame atomic absorption spectroscopy using a Shimadzu AA-6800 spectrometer. Hemoglobin (Hb) iron pool (mg), assuming a total rat blood volume of 6.7% body weight and an iron content in hemoglobin of 0.335%, was calculated as follows:10 Hb iron pool (mg) = body wt (kg) × 0.067 × Hb concn (g/L) × 3.35
The hemoglobin regeneration efficiency was calculated by using the following equation: Hb regeneration efficiency (%) = [final Hb iron pool (mg) − initial Hb iron pool (mg)] /total iron consumed (mg) × 100
The relative biological value was calculated on the basis of the hemoglobin regeneration efficiencies: relative biological value (%) = Hb regeneration efficiency of each animal /av Hb regeneration efficiency of the FeSO4 group × 100
Iron speciation analysis of rat intestinal contents was carried out as previously described by Kapsokefalou and Miller11 and Simpson et al.12 with some modifications. Male Sprague−Dawley rats (53 days old; n = 40) weighing 290−310 g were stratified according to body weight and within each stratum were randomly distributed into five groups (eight rats each). The rats were fed the Fe-deficient diet for a week before they were daytime fasted for 9 h. The fasted rats in each group then consumed one of the five Fe-repletion diets for 30 min ad libitum and were anesthetized with ether 90 min after consumption of the meal. The rats were killed by cervical dislocation before an abdominal midline incision was performed. After the small intestine was removed from the carcass, the content in the proximal one-third segment was flushed with saline into a centrifuge tube and further fractionated into two fractions using Millipore filters (Göttingen, Germany): the 0.22 μm filtrate and the 3 kDa (≈1 nm) filtrate. Iron concentrations in the intestinal content, 0.22 μm filtrate, and 3 kDa filtrate were determined by flame atomic absorption spectroscopy using a Shimadzu AA-6800 spectrometer. Cellular Uptake Experiment. Caco-2 cells were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China) and seeded at a density of 5 × 104 cells/cm2 in collagen-treated 24-well plates (BD Biosciences, San Jose, CA, USA). The cells were routinely maintained in high-glucose (4.5 g/L) Dulbecco’s modified Eagle medium supplemented with 10% heat-inactivated fetal bovine serum, 25 mM HEPES, 4 mM glutamine, and 1 mM sodium pyruvate at 37 °C in an incubator with a 95% air/5% CO2 atmosphere at constant humidity, and the medium was renewed every 2 days. After reaching confluence (2−3 days postseeding), the cells were maintained in 8633
dx.doi.org/10.1021/jf5018719 | J. Agric. Food Chem. 2014, 62, 8632−8639
Journal of Agricultural and Food Chemistry
Article
Figure 1. Structural characterization of in vitro-formed nanosized ferric hydrolysis products stabilized by anchovy muscle protein hydrolysate (ANPs): (A) transmission electron microscopy images collected from a drop of ANPs; (B) whole area energy dispersive X-ray analysis of one of the particles in panel A showing elemental compositions (peaks for sodium (Na), copper (Cu), potassium (K), silicon (Si), phosphorus (P), sulfur (S), chlorine (Cl), carbon (C), and oxygen (O) are identified); (C) high-resolution transmission electron microscopy; (D) selected area electron diffraction. 2.5 h, and calcein fluorescence (485 nm excitation, 530 nm emission) was monitored every 5 min. To trap and detect ferrous iron, 100 μM BPDS was used, and the absorbance of the ferrous tris(BPDS) complex was measured at 536 nm (molar extinction coefficient, ε = 1.927 × 104 L mol−1 cm−1). Stock solution of FeSO4 (1 mM) was freshly prepared before use. Stock solution of ferric chelates (1:1 molar ratio) of NTA, phytate, EDTA, DTPA, CDTA, and DFB was prepared by mixing equal volumes of chelators (2 mM) and freshly prepared FeCl3 (2 mM). Statistical Analysis. Data are expressed as the mean ± standard deviation (SD). Differences between groups were tested for significance by the least significant difference (LSD) mean comparison using the SPSS software program (SPSS Inc., Chicago, IL, USA). The relationship between variables was determined by the Pearson
complete Dulbecco’s modified Eagle medium for another 12 days to differentiate completely and further incubated for 24 h in serum-free minimum essential medium supplemented with 4 mg/L hydrocortisone, 5 mg/L insulin, 5 μg/L selenium, 34 μg/L triiodothyronine, and 20 μg/L epidermal growth factor. Following this, the medium was aspirated, and the cells were stained with 1 μM calcein acetoxymethylester in Hanks’ balanced salt solution (HBSS) for 30 min at 37 °C before being washed three times with HBSS. The cells were then incubated in Tyrode solution (137 mM NaCl, 2.7 mM KCl, 1 mM MgCl2, 1.8 mM CaCl2, 5.5 mM D-glucose) buffered with 20 mM MES (pH 5.5) or MOPS (pH 7.0) containing FeSO4, ANPs, or any of 1:1 ferric chelates of NTA, phytate, EDTA, DTPA, CDTA, and DFB at the concentration equivalent to 20 μM iron in a Synergy H4 hybrid microplate reader (Bio-Tek, Winooski, VT, USA) at 37 °C for 8634
dx.doi.org/10.1021/jf5018719 | J. Agric. Food Chem. 2014, 62, 8632−8639
Journal of Agricultural and Food Chemistry
Article
Table 1. Details and Results of the Hemoglobin Regeneration Bioassay groupa N diet iron (mg/kg) diet intake (g) baseline BW (g) BW gainb (g) baseline Hb (g/L) Hb changec (g/L)
Fe-deficient
FeSO4
ANPs
FC
FC+AMPH
Nano-α-Fe2O3
Fe-sufficient
8 10.5 ± 1.5 265 ± 30.5a 222.7 ± 21.5 62.4 ± 27.8a 71.1 ± 11.9 −14.4 ± 11.1a
8 32.4 ± 2.3 419.5 ± 70.3c 221.0 ± 20.1 101.7 ± 32.9bc 71.4 ± 16.8 50 ± 19.8b
8 30.5 ± 0.7 332.1 ± 32.7ab 221.4 ± 39.2 109.7 ± 37.2c 73.7 ± 10.5 52.8 ± 19.8b
8 29.8 ± 2.0 268.2 ± 43.7a 222.4 ± 38.8 79.7 ± 25.9ab 70.5 ± 15.3 −6.9 ± 21.7a
8 28.8 ± 0.2 380.6 ± 70.0bc 215.0 ± 17.2 88.7 ± 28.9ab 71.9 ± 12.7 42.8 ± 21.0b
8 31.8 ± 2.4 267.2 ± 16.9a 219.9 ± 11.8 75.6 ± 26.7ab 71.1 ± 19.0 0.125 ± 17.2a
4 36.8 ± 2.5 383.9 ± 37.9b 297.5 ± 33.9 85.0 ± 3.6ab 153.8 ± 13.9 −25.3 ± 23.8
a
Abbreviations: FeSO4, ferrous sulfate; ANPs, in vitro-formed AMPH-stabilized nanosized ferric hydrolysis products; FC, ferric citrate; FC+AMPH, mixture of ferric citrate and anchovy muscle protein hydrolysate; Nano-α-Fe2O3, commercial bare α-Fe2O3 nanoparticles. Values are means ± SD. Means in a column without a common letter (a−c) are statistically different, P < 0.05. bBW, body weight. cHb, hemoglobin. correlation test. The probability level 30% in all five groups, suggesting that the AIN-93G diet could induce a high basal level of intestinal nanosized iron. This might be attributed to the digestion products of maltodextrin, starch, and casein in the basal formulation of AIN-93G diet or gastrointestinal secretions. Interestingly, the percentage of nanosized iron for the FeSO4 group was also significantly higher than that of its LMW counterpart (P < 0.05), which might be explained by the formation of nanosized iron via rapid in situ oxidation and hydrolysis of ferrous iron bound to soluble macromolecules in the presence of swallowed oxygen in the intestinal tract. Due to their tendency to aggregate and sediment in aqueous solutions, bare iron oxide nanoparticles are difficult to disperse into stable suspensions within the nanoscale range unless very efficient dispersing techniques (e.g., probe ultrasonication) are employed,28 so it is unsurprising that Nano-α-Fe2O3 induced a significantly lower level of intestinal nanosized iron than ANPs (P < 0.05). The involvement of intestinal nanosized iron in iron absorption from FC+AMPH, FeSO4, and ANPs raises questions about how gut epithelial cells absorb intraluminal nanosized iron. Usually, exogenous iron enters the cellular labile iron pool before it is stored in ferritin, exported from the cell, or used to synthesize iron proteins.29 Calcein, a fluorescence probe for the labile iron pool, has been frequently used to monitor cellular iron uptake in real time.30,31 We thus used calcein-loaded polarized Caco-2 cell monolayers as a gut epithelial model to investigate the intestinal iron absorption from ANPs, a mimetic for intraluminal nanosized iron. As shown in Figure 3, iron uptake from FeSO4 and ANPs at concentrations equivalent to 20 μM iron was monitored by the quenching of calcein fluorescence, that is, the incorporation of exogenous iron into the cellular labile iron pool, under two pH conditions (5.5 and 7.0) simulating those in the proximal and distal small intestine.32 BPDS, a strong ferrous chelator, was used to block iron absorption via the divalent metal transporter 1 (DMT1). As expected, cellular iron uptake from FeSO4 was completely blocked by BPDS at pH 5.5 and 7.0 (data for the latter pH are not shown), whereas iron uptake from ANPs was only partially blocked at both pH values, suggesting that both DMT1-dependent and DMT1-independent mechanisms are involved in iron absorption from ANPs. The two best known DMT1-independent iron uptake mechanisms are by means of heme carrier protein 1 and endocytosis. Heme carrier protein 1 is specific to heme iron uptake,33 so endocytosis, a very common mechanism for cellular entry of nanoparticles, attracted much attention in 8636
dx.doi.org/10.1021/jf5018719 | J. Agric. Food Chem. 2014, 62, 8632−8639
Journal of Agricultural and Food Chemistry
Article
Figure 4. Ferrous concentrations after 2 h of incubation of 20 μM different forms of ferric iron with polarized Caco-2 cell monolayers at an extracellular pH of 5.5. Abbreviations: ANPs, in vitro-formed nanosized ferric hydrolysis products stabilized by anchovy muscle protein hydrolysate; CDTA, trans-1,2-cyclohexanediaminetetraacetic acid; DFB, deferoxamine mesylate; DTPA, diethylene triamine pentaacetic acid; EDTA, ethylenediaminetetraacetic acid; NTA, nitrilotriacetic acid. Ferric stability constants were obtained from the literature.40 Data are presented as means ± standard deviations of three independent experiments. Groups not sharing a common letter are significantly different (P < 0.05).
Figure 3. Iron uptake by polarized Caco-2 cell monolayers in the presence of 20 μM iron at extracellular pH values of 5.5 and 7.0. Abbreviations: FeSO4, ferrous sulfate; ANPs, in vitro-formed AMPHstabilized nanosized ferric hydrolysis products; BPDS, bathophenanthrolinedisulfonic acid. Cellular uptake of exogenous iron into the labile iron pool was measured as the quenching of calcein fluorescence. BPDS is a strong ferrous chelator to block the divalent metal transportor-1 pathway. Data are means of three independent experiments with standard deviations of
