Main

Oxidative stress occurs when there is an imbalance between the amount of antioxidant properties and reactive oxygen species. In the neonatal period, especially in preterm infants injury to the GIT by oxidative stress is thought to be involved in the pathogenesis of serious diseases such as necrotizing enterocolitis (1,2).

HM is the ideal food during infancy and is known to contain various types of bioactive substances, some of which are reported to be antioxidants (3). We demonstrated the antioxidative effects of HM in two previous studies. The level of urinary 8-hydroxy-2′-deoxyguanosine (a marker of oxidative DNA damage) excretions of breast-fed infants was significantly lower than that of formula-fed infants in 1 mo of age (4). HM-treated IEC-6 cells showed a significantly higher survival rate after H2O2-induced oxidative damage than cells treated with infant formula (5). Despite these findings, it is still unclear which components of HM play the most important antioxidative roles in the GIT of infants.

Lf is an iron-binding glycoprotein that belongs to the transferrin (Tf) family. Although it has been identified in many secretions from various species, it mainly exists in mammalian milk and colostrum. The amino acid compositions of bovine Lf and human Lf show 69% sequence homology (6). Lf appears to play a role in many biologic processes, including the proliferation and differentiation of enterocytes (7), regulation of iron absorption (8), antimicrobial activities (9), immunomodulation and anti-inflammatory responses (10), and antivirus or anticancer activities (11). Lf is considered to be an antioxidant because of its ability to bind two atoms of iron, which is important in the formation of free radicals (12). Interestingly, Lf concentration and iron saturation in milk vary considerably with species; bovine milk contains approximately 0.1 mg/mL of Lf that is 22% saturated with iron compared with an average Lf of 1.4 mg/mL (10%–20% of the total protein) in mature HM that is only 4% iron saturated (13).

Recently, a fluorescent probe is one of the most straightforward techniques used in the assessment of intracellular oxidative stress. The fluorescent reporter molecule DCFH-DA is a commonly used intracellular fluorescent probe that diffuses rapidly through cell membranes and is subsequently hydrolyzed to DCFH by intracellular esterases. The nonfluorescent DCFH subsequently trapped inside the cell, whereupon oxidation changes it to the highly fluorescent dichlorofluorescein (DCF). It may serve as a sensitive cytosolic marker of oxidative stress. It has been proposed that DCFH detects a broad range of oxidation reactions that occur during intracellular oxidative stress (14).

In the present study, we attempt to determine the antioxidative capacities of apo-hLf in comparison with those of holo-hLf, apo-bTf, and the iron chelator DFX in an IEC to clarify whether apo-hLf exhibits its iron-chelating capacity to reduce oxidative stress. Furthermore, we used DCFH-DA to demonstrate the cell protection mechanisms of apo-hLf during H2O2-induced oxidative damage.

METHODS

Cell lines.

For cell viability experiments, we used IEC-6 cells (ATCC, Rockville, MD, obtained from the Riken Cell Bank, Ibaraki, Japan, passages between 15 and 25 after acquisition) derived from a nontransformed rat intestinal crypt cell. The cells were maintained in Dulbecco's modified Eagle's medium (DMEM) (GIBCO Laboratories, Grand Island, NY) containing 5% fetal calf serum (FCS) (GIBCO), 4.5 g/L glucose, 0.1 U/mL insulin (Wako Pure Chemical Industries, Osaka, Japan), 5 mmol/L l-glutamine (GIBCO), 100 U/mL penicillin, and 100 μg/mL streptomycin (GIBCO).

For fluorescent intensity experiments, we used Caco-2 cells (ATCC, obtained from the official distributor in Japan, passages between 40 and 50) derived from a human colon adenocarcinoma. These cells spontaneously differentiated in culture resemble small intestinal enterocytes in both morphology and biochemical function. The cells were maintained in stock cultures in DMEM supplemented with 2 mmol/L glutamine, 10% FCS, 0.1 mmol/L nonessential amino acids, 100 U/mL penicillin, and 100 μg/mL streptomycin (GIBCO). Cells were cultured in a humidified incubator with an atmosphere of 5% CO2 at 37°C and were supplied fresh medium every 2 d.

Oxidative stress and cell viability evaluation.

H2O2 (Wako) was added to the culture medium to induce oxidative stress. Neutral red (NR), which is only taken up by living cells, was used to measure cell survival rates (15). Cells were incubated with 0.33 g/L NR (Sigma Chemical Co. St. Louis, MO) in DMEM for 1 h. Subsequently, the cell layer was rinsed twice with phosphate-buffered saline (PBS) (GIBCO), and NR was extracted with 1% (vol/vol) acetic acid in a 1:1 water:ethanol mixture. Absorbance at 560 nm was measured spectrophotometrically.

Preliminary experiments.

Two preliminary experiments were performed to determine suitable experimental conditions.

H2O2 concentration.

To determine the optimal concentration of H2O2 for the oxidative challenge, confluent IEC-6 cells that were preincubated with or without 50 μg/mL apo-hLf (Sigma Chemical Co.) for 24 h were subsequently treated with 0.1, 0.25, or 0.5 mmol/L H2O2 for 30 min at 37°C.

apo-hLf concentration.

To define the most effective concentration of apo-hLf, confluent IEC-6 cells were examined for 24 h with 0, 5, 25, or 50 μg/mL apo-hLf at 37°C and then challenged with 0.25 mmol/L H2O2 for 30 min at 37°C.

Cell viability experiment.

IEC-6 cells were cultivated with FCS-containing medium in a 48-well culture plate (Coster Corp., Cambridge, MA) at an initial cell density of 5 × 104/cm2. On d 5, the medium was replaced by medium without FCS and cells were preincubated for 24 h either with 50 μg/mL of apo-hLf, holo-hLf (Sigma Chemical Co.), apo-bTf (Sigma Chemical Co.), or 800 ng/mL DFX (Sigma Chemical Co.) in minimal essential medium (MEM) (GIBCO), which does not contain iron, to minimize the effect of free iron ion. On d 6, H2O2 (final concentration of 0.25 mmol/L) was applied as a challenge for 30 min. Cells were washed three times with PBS and then incubated for 24 h with complete medium. On d 7, an NR uptake assay was performed.

Fluorescence intensity experiment.

Caco-2 cells at an initial cell density of 5 × 104/cm2 were grown in 35-mm glass bottom culture dishes with FCS-containing medium. After differentiated cells were confluent (d 21), Caco-2 cells were preincubated for 24 h in MEM (without FCS) with or without the addition of 50 μg/mL of apo-hLF. Then cells were treated with 1 mmol/L H2O2 in MEM for 45 min, followed by incubation with 10 μmol/L DCFH-DA at 37°C in a dark environment for 10 min, and washed with PBS three times.

Fluorescent intensity of images.

The images of Caco-2 cells exhibiting DCF fluorescence were taken using a confocal laser scanning microscope (CLSM) (Nikon). The images of five different, randomly selected fields (60–80 min after the DCFH-DA treatment), were stored in a personal computer. Quantitative analysis of fluorescent intensity was performed using NIH Image (version 1.61).

Fluorescent intensity of cell homogenates.

After DCFH-DA treatment, the cells were removed from the culture dishes by scraping and placed on ice. Using a Potter-type homogenizer, the cells were disrupted in ice-cold phosphate buffer containing 0.1% (vol/vol) Triton X-100. The homogenate was centrifuged at 15,000 × g for 15 min at 4°C. Fluorescent intensities were quantified with a fluorometer. Total protein concentrations of cell homogenates were determined using a Bio-Rad protein assay with bovine serum albumin as the standard. Results were expressed as a ratio of fluorescence intensity per milligram of total protein.

Data analysis.

All data were expressed as the mean ± standard deviation (SD). Differences between groups were analyzed with a t test. Differences from baseline were tested by one-way analysis of variance and, if indicated, by post hoc Tukey-Kramer test. A p value <0.05 was considered statistically significant. All statistical analyses were performed with StatView 5.0 (Abacus Concepts, Inc., Berkeley, CA).

RESULTS

Optimal concentration of H2O2.

Cell viability decreased as the concentration of H2O2 increased. Survival rates of IEC-6 cells preincubated with 50 μg/mL of apo-hLf after the 0-, 0.1-, 0.25-, or 0.5-mmol/L H2O2 challenge were 102.6 ± 8.1%, 97.9 ± 4.6%, 92.0 ± 2.0%, and 49.2 ± 3.1%, respectively (n = 6, mean ± SD). In contrast, survival rates of the cells after the H2O2 challenge without apo-hLf preincubation were 100.1 ± 5.4%, 98.1 ± 3.6%, 77.2 ± 5.3%, and 24.4 ± 3.8%, respectively (n = 6, mean ± SD) (Fig. 1). After the 0.25- and 0.5-mmol/L H2O2 challenge, significantly higher survival rates were observed in the cells preincubated with apo-hLf than those preincubated without apo-hLf. Therefore, 0.25 mmol/L H2O2 was considered to be optimal.

Figure 1
figure 1

Effect of apo-hLf on H2O2-induced oxidative damage in IEC-6 cells. Confluent IEC-6 cells were preincubated with (solid line) or without (dashed line) 50 μg/mL of apo-hLf followed by increasing concentrations of H2O2 challenge for 30 min at 37°C. Surviving cell rates are expressed as a percentage of control (0 mmol/L H2O2). Data represent the means ± SD of six replicate wells for a single experiment that is representative of three comparable experiments. *p < 0.05 compared with value of no apo-hLf group.

Optimal concentration of apo-hLf for preincubation.

Increasing concentrations of apo-hLf were added and preincubated for 24 h before the 0.25-mmol/L H2O2 challenge. The survival rates of no pretreatment cells (H2O2 challenge only) or the cells preincubated with 5, 25, and 50 μg/mL apo-hLf were 69.5 ± 1.7%, 70.8 ± 1.7%, 78.5 ± 5.2%, and 82.1 ± 7.3%, respectively (n = 6, mean ± SD) (Fig. 2). Cell viabilities of the cells preincubated with 25 and 50 μg/mL apo-hLf were significantly higher than those of untreated cells. Therefore, the concentration of 50 μg/mL was chosen.

Figure 2
figure 2

Effect of pretreatment with apo-hLf on H2O2-induced oxidative damage in IEC-6 cells. Confluent IEC-6 cells were preincubated for 24 h with different concentrations of apo-hLf at 37°C and then challenged H2O2 (0.25 mmol/L, 30 min). No pretreatment samples received only H2O2. Surviving cell rates are expressed as a percentage of control (no H2O2 challenge) cells. Data represent the means ± SD of six replicate wells for a single experiment that is representative of three comparable experiments. The data were analyzed by one-way analysis of variance with post hoc analysis by the Tukey-Kramer test. *p < 0.05 compared with value of no pretreatment group.

Cell viability experiment.

The cell survival rates of untreated IEC-6 cells and the rates for cells preincubated with apo-hLf, holo-hLf, apo-bTf, or DFX were 76.8 ± 2.5%, 95.7 ± 5.8%, 80.5 ± 4.4%, 88.0 ± 1.7%, 99.7 ± 5.1%, respectively (n = 6, mean ± SD) (Fig. 3). Survival rates of the cells preincubated with apo-hLf, apo-bTf, or DFX were significantly higher than those of untreated cells or cells preincubated with holo-hLf. Survival rates of the cells preincubated with apo-bTf were also significantly higher than those of untreated cells.

Figure 3
figure 3

Antioxidative property of apo-hLf, holo-hLf, apo-bTf, and DFX on H2O2-induced oxidative damage in IEC-6 cells. Confluent cells were preincubated with 50 μg/mL of apo-hLf, holo-hLf, or apo-bTf or 800 ng/mL of DFX for 24 h at 37°C. Then 0.25 mmol/L H2O2 was applied as a challenge for 30 min. Survived cell rate are expressed as a percentage of control (no H2O2 challenge) cells. Data represent the means ± SD of six replicate wells for a single experiment that is representative of three comparable experiments. The data were analyzed by one-way analysis of variance with post hoc analysis by the Tukey-Kramer test. *p < 0.05 compared with value of no pretreatment group. †p < 0.05 compared with value of the holo-hLf group.

Fluorescence intensities of images.

Fluorescence of DCF was most evident along the cell membrane (Fig. 4a). In the CLSM images, the fluorescent intensities of Caco-2 cells preincubated with apo-hLF were lower compared with those of untreated cells. Fluorescent intensities, as measured by NIH Image analysis software, were significantly lower in the cells preincubated with apo-hLF than in untreated cells (33.4 ± 3.5 and 47.0 ± 10.0 relative intensity unit, respectively; n = 5, mean ± SD) (Fig. 4b).

Figure 4
figure 4

Effect of apo-hLf on H2O2 in DCF fluorescence in Caco-2 cells. Cells were preincubated with or without 50 μg/mL of apo-hLF for 24 h. Cells were treated with 1 mmol/L H2O2 in culture medium for 45 min, followed by incubation with 10 μmol/L DCFH-DA at 37°C (5% CO2-air:atmosphere) for 10 min. Fluorescent DCF images were taken using a CLSM. (a) Representative DCF fluorescent images from no preincubated cells (56 min after DCFH-DA treatment) and preincubated with apo-hLf cells (55 min after DCFH-DA treatment). Scale bars = 20 μm. (b) Fluorescent intensity of CLSM images. The images of five different, randomly selected fields were stored in a personal computer. Quantitative analysis of fluorescent intensity was performed using NIH Image, expressed as relative intensity unit. (c) Fluorescent intensity of cell homogenates. After DCFH-DA treatment, the cells were removed from the culture dishes by scraping and were disrupted in ice-cold phosphate buffer containing 0.1% (vol/vol) Triton X-100 using a homogenizer. Fluorescent intensities were quantified with a fluorometer. Results were expressed as a ratio of fluorescence intensity per milligram of total protein (relative light unit/milligram of protein). Values are means + SD (n = 6). The data were analyzed by t test. *Statistically different (p < 0.05).

Fluorescent intensities of cell homogenates. Fluorescent intensities of cell homogenates (expressed as protein ratio) were significantly lower in the cells preincubated with apo-hLf than in untreated cells (13.3 ± 3.2 and 15.4 ± 3.6 relative light unit/milligram of protein, respectively; n = 6, mean ± SD) (Fig. 4c).

DISCUSSION

We evaluated the antioxidative effects of apo-hLf in intestinal epithelial cell lines. Significantly higher survival rates were observed in the cells preincubated with the iron-unsaturated form of iron binding proteins and/or an iron-chelating agent.

Lf binds two atoms of ferric iron with high affinity (300 times higher than Tf) and still maintains function as an iron chelator. Huang et al. (16) reported that iron-unsaturated Lf has an antioxidant effect on a corn oil emulsion and on a lecithin liposome system, although iron-saturated Lf was unable to inhibit oxidation in a liposome system. Furthermore, iron-unsaturated Lf decreased the generation of hydroxyl radical by phagocytes and inhibited iron-catalyzed auto-oxidation of monocyte membranes (17).

Lf-specific receptors have been identified in brush-border membranes of fetal and infant intestines (18) and have been suggested to contribute to the cellular iron uptake. Lf receptor is also identified in IEC-6 cells (19) and Caco-2 cells (20). Ashida et al. (21) also reported that human Lf was internalized by Caco-2 cells from the apical side and localized to the nuclei. These studies showed that Caco-2 and IEC-6 cell lines are useful models to assess the effect of Lf in the intestine.

The results of fluorescence experiments indicate that oxidative stress levels of apo-hLf treated cells were significantly lower than those of untreated cells. Previous studies implicated a role for redox-active iron in cellular oxidation of DCFH to DCF (22,23). In contrast, H2O2-induced intracellular DCF fluorescence was inhibited by iron chelators (24). We speculate that apo-hLf in culture medium chelates intra- and extracellular free ferric iron, which subsequently decreases intracellular free ferrous iron and results in decreased oxidative stress. However, we have no evidence of whether Lf receptors in intestinal cells can uptake both holo and apo forms of Lf. Additional studies are needed, but previous studies reported the recognition of Lf by the Lf receptor of the protozoan parasite Leishmania donovani being independent of whether it is metal loaded (25). Conversely, the affinity of metal-loaded Tf for the Tf receptor is much higher than that of apo-Tf (26). Zhang et al. also demonstrated that apo-hLf can be recognized as well as holo-hLf in IEC-6 cells (19).

Supplementation of iron in infant formula or breast milk has the potential of exacerbating free radical oxidation via its role as a catalyst in oxidative stress (27). However, the use of iron-fortified infant formulas also prevents iron deficiency anemia among full-term infants (28). Lf is known to be partially resistant to proteolysis by addition of trypsin (29), and it has been shown that a significant proportion of Lf can survive digestion in infants (30). Bovine Lf has been added to infant formula since 1986, but clinical trials have not shown any enhancing effect on iron absorption or iron status (31,32) because the binding of hLf receptor is considered specific for hLF (33). Suzuki and colleagues (34) recently demonstrated that recombinant hLf expressed in rice shows stable biologic activities similar to those of native hLf. This suggests that recombinant hLf may be a valuable source when added to infant formula.

The present findings indicate that apo-hLf and other iron-binding compounds alleviate H2O2-induced oxidative damage in IEC-6 cells and apo-hLf cells. Pretreatment of Caco-2 cells with apo-Lf appears to attenuate the intracellular oxidative stress caused by exogenous H2O2. We speculate that the iron-chelating capacity of apo-Lf plays an important role in reducing oxidative stress in these systems. In conclusion, our findings suggest that iron-unsaturated Lf in human breast milk acts as an antioxidant in the GIT.