Dietary Iron Deficiency and Oversupplementation Increase Intestinal Permeability, Ion Transport, and Inflammation in Pigs^,,

Abstract

Background: Understanding the influence of dietary iron deficiency and dietary iron oversupplementation on intestinal health is important for both animal production and human health.

Objective: The aim of this study was to determine whether dietary iron concentration influences intestinal physiology, morphology, and inflammation in the porcine duodenum.

Methods: Twenty-four male pigs (21 d old) were fed diets containing either 20 mg Fe/kg [low dietary iron (L-Fe)], 120 mg Fe/kg [adequate dietary iron (A-Fe); control], or 520 mg Fe/kg [high dietary iron (H-Fe)] by FeSO₄ supplement (dry matter basis). After 32–36 d, the duodenum was harvested from pigs and mounted in Ussing chambers for the measurement of transepithelial electrical resistance (TER), short-circuit current, and ³H-mannitol permeability. Intestinal morphology and inflammation were assessed by histologic examination, and proinflammatory gene expression was assessed by real-time polymerase chain reaction.

Results: Compared with A-Fe–fed pigs, pigs fed L-Fe diets exhibited reduced TER (by 30%; P < 0.05). Compared with that of A-Fe–fed controls, the paracellular flux of ³H-mannitol across the duodenal mucosa was higher (P < 0.05) in L-Fe–fed (>100%) and H-Fe–fed (∼4-fold) pigs; the L-Fe–fed and H-Fe–fed groups did not differ significantly from one another. Compared with the L-Fe–fed pigs, the A-Fe–fed and H-Fe–fed pigs had malondialdehyde concentrations 1.4- and 2.5-fold higher in the duodenum and 4.4- and 6.6-fold higher in the liver, respectively (P < 0.05). Neutrophil counts were higher in both the L-Fe–fed (by 3-fold) and H-Fe–fed (by 3.3-fold) groups than in the A-Fe–fed group; the L-Fe–fed and H-Fe–fed groups did not significantly differ from one another. Duodenal mucosal tumor necrosis factor α (TNFA), interleukin (IL) 1β, and IL6 relative gene expression was upregulated by 36%, 28%, and 45%, respectively, in H-Fe pigs (P < 0.05), but not in L-Fe pigs, compared with A-Fe pigs.

Conclusion: These data suggest that adequate but not oversupplementation of dietary iron in pigs is required to maintain intestinal barrier health and function.

Introduction

Iron deficiency is one of the most common nutritional disorders worldwide. The WHO (2006) estimated that >2 billion people worldwide suffer from iron-deficient anemia (1). In childhood, iron deficiency and subsequent anemia increase the risk of stunted growth and mental retardation (2). In addition to insufficient iron intake, impaired iron absorption and chronic blood loss in patients with inflammatory bowel disease (IBD)⁹ also could cause iron-deficient anemia (3, 4). Therefore, dietary iron fortification is crucial to the prevention and treatment of iron deficiency. However, it has been shown that iron supplementation itself could result in an increased risk of infection-related mortality and morbidity from diarrhea (5–7). Therefore, the conflict between iron-deficiency anemia and the health issue of overdose from iron supplementation has spurned greater interest in the role of dietary iron intake on human intestinal health. In agricultural pig production, iron supplementation of young pigs via dietary and parenteral routes is a standard practice to prevent iron deficiency in the rapidly growing pig. Weaned nursery pigs require 80–100 mg Fe/kg dry matter (DM); however, many commercial starter diets contain amounts of iron in considerable excess of this requirement. This is due to the fact that iron is supplemented to the diet, and the fact that several swine feed ingredients contain excessive amounts of iron, including blood meal (3000 mg Fe/kg DM), dicalcium phosphate (10,000 mg Fe/kg DM), and limestone (3500 mg Fe/kg DM) (8).

Accumulating evidence suggests that iron has a substantial impact on intestinal function, potentially via different mechanisms. High dietary iron contributes to oxidative stress through the production of free radicals, which leads to lipid peroxidation, inflammation, and intestinal injury (9, 10). Increased dietary iron intake in rodents also increased bacterial proliferation in vivo (11). In humans, oral iron supplementation increased the clinical inflammation score of patients with IBD (9, 12). In addition, several lines of evidence indicate that inflammatory status is also induced by iron deficiency. In mouse models, iron deficiency induced iron depletion of the macrophage associated with the synthesis of cytokines (13). Moreover, iron deficiency severely reduced hepcidin transcription, which exaggerated proinflammatory cytokine production of macrophages (14). Despite the importance and prevalence of iron deficiency and the common use of oral iron supplementation, and the associated gastrointestinal clinical signs, very little is known about the influence of deficient or excess dietary iron on intestinal function. The objective of this study was to determine whether dietary iron concentration (low, adequate, and high) influences intestinal physiology, morphology, and inflammation in the pig. We hypothesized that feeding low- or high-iron diets to pigs would be disruptive to duodenal barrier function and morphology relative to pigs fed adequate-iron diets, and that feeding high-iron diets would induce intestinal inflammation.

Methods

Pigs and experimental design.

All pig protocols were approved by the North Carolina State University Animal Care and Use Committee before initiation of the 32-d trial. The pigs and experimental design used for this study also were used for an independent objective to investigate the effects of dietary iron on manganese metabolism, which was published previously (15). Briefly, 24 male weanling pigs (aged 21 d; body weight 5.5 ± 0.4 kg) were blocked by litter and weight and randomly assigned to 1 of 3 treatments (2 pigs/pen, 4 pens/treatment). At birth, piglets were injected with a one-half dose of iron dextran (100 mg) to minimize body iron stores before initiation of the trial while still preventing anemia. Treatments included an iron-deficient basal diet [low dietary iron (L-Fe)], a basal diet supplemented with FeSO₄ to provide 100 mg supplemental Fe/kg DM to meet the requirement of iron for young pigs [adequate dietary iron (A-Fe)], and a basal diet with 500 mg supplemental Fe/kg DM to represent the concentration of iron often found in commercial starter diets [high dietary iron (H-Fe)] (15). The basal diet was analyzed to contain 20 mg Fe/kg DM, which fulfilled ∼25% of the iron requirement of young pigs (8).

Intestinal sample collection.

Two pigs per treatment per day were randomly selected to be killed on days 33, 34, 35, and 36. To minimize stress, pigs were sedated before killing by using a combination of Telazol, ketamine, and xylazine at a dose of 0.03 mL/kg body weight. Pigs were killed with 2.2 mL/kg body weight Fatal Plus (pentobarbital sodium) given intravenously via a catheterized ear vein. Immediately after pigs were killed, a 13-cm segment of the proximal duodenum beginning ∼10 cm distal to the pyloric sphincter was harvested and placed in cold oxygenated (95% O₂:5% CO₂) Ringer solution (16). The next 25 cm of duodenum was removed from the pig, cut open longitudinally, and rinsed generously with 0.87% saline to remove any digesta before sample collection. Duodenal scrapings of the exposed mucosa were collected, immediately flash-frozen in liquid N₂, and stored at −80°C until gene and protein analysis.

Ussing chamber procedures.

Segments of proximal duodenum were prepared for Ussing chamber experiments as described previously (16). Briefly, the mucosa was stripped from the seromuscular layer in oxygenated (95% O₂:5% CO₂) Ringer solution. Tissues were then mounted in 1.13-cm²–aperture Ussing chambers (17). Tissues were bathed on the serosal and mucosal sides with 10 mL oxygenated Ringer solution maintained at 37°C. The spontaneous potential difference (PD) was measured with the use of Ringer–agar bridges connected to calomel electrodes, and the PD was short-circuited through Ag–AgCl electrodes with the use of a voltage clamp that corrected for fluid resistance. Tissues were maintained in the short-circuited state, except for brief intervals to record the open-circuit PD. Transepithelial electrical resistance (TER; Ω · cm²) was calculated from the spontaneous PD and short-circuit current (I_sc), as previously described (18). After a 30-min equilibration period in Ussing chambers, TER was recorded at 15-min intervals over a 2-h period and then averaged to derive the basal TER values for a given pig.

Mucosal to serosal fluxes of ³H-mannitol.

To assess dietary iron effect on mucosal permeability, 0.2 microcuries ³H-mannitol/mL was placed on the mucosal side of Ussing chamber–mounted tissues. After a 15-min equilibration period, standards were taken from the mucosal side of each chamber and a 60-min flux period was established by taking 0.5-mL samples from the serosal compartment. The presence of ³H was established by measuring β emission in a liquid scintillation counter (LKB Wallac, model 1219 Rack Beta; Perkin Elmer Life and Analytical Sciences). Unidirectional ³H-mannitol fluxes from mucosa to serosa were evaluated by determining mannitol-specific activity added to the mucosal bathing solution and calculating the net appearance of ³H over time in the serosal bathing solution on a chamber unit area basis.

Histopathology.

Duodenal tissues were harvested from pigs and fixed in 10% neutral buffered formalin until processing. Paraffin-embedded intestinal samples were sectioned (5 μm) and stained with either Prussian blue (iron staining) or hematoxylin and eosin for histologic analysis. Histologic analysis was performed by a trained individual who was blinded to experimental groups. Photomicrographs were acquired with 20× and 40× magnifications at a resolution of 1360 × 1024 with the use of imaging software (Olympus DP2-BSW, version 2.2) while operating a high-resolution digital camera (Olympus DP72) equipped with a clinical light microscope (Olympus BX45). Before imaging, the system was calibrated at each magnification with the use of a stage micrometer. Measurements were taken by using the arbitrary line tool and exported onto a spreadsheet program (Excel 2010; Microsoft). Villi were measured with the use of the 20× objective. For each duodenal-stained histologic slide prepared for each pig, we located 4 different areas on the slide within the 20× field view that contained ≥3 well-oriented villi based on the criteria that 1) the entire crypt and villi be captured in cross-section and 2) the central lacteal be present. Therefore, a minimum of 12 individual villi measurements/pig were measured and then averaged to derive the mean villi height for each pig. Neutrophil counts were performed at 40× magnification with the use of a 0.043-mm² grid and were expressed as number of neutrophils/mm² for each pig. All histopathology data were analyzed as n = 4 pens/treatment.

Oxidative status analysis.

Isolation of proteins from snap-frozen duodenal scrapings was performed as previously described (15). The extracted mucosal proteins were analyzed in duplicate for malondialdehyde with the use of a commercially available kit (Lipid Peroxidation Assay Kit; Oxford Biomedical).

Gene expression analysis.

Total RNA samples were isolated from frozen duodenal mucosal scrapings with the use of the Qiagen RNeasy Mini kit. First-strand cDNA was synthesized from 3-μg RNA with the use of a Thermo Scientific Maxima First Strand cDNA Synthesis Kit for RT-qPCR with dsDNase according to the manufacturer’s instructions. Semiquantitative real-time PCR was used to determine the relative quantities of transcripts of the genes of interest. The genes encoding the ribosomal protein large subunit 4 (RPL4) and β-actin (ACTB) were selected and validated as suitable internal reference genes (19). The relative gene expressions of mast cell–specific tryptase 7 (MCT7) and chymase 1 (CMA1), proinflammatory cytokines, tumor necrosis factor α (TNFA), IL1B1, IL6, IL8, IL10, and interferon γ (IFNG) were determined (Supplemental Table 1). All PCR reactions were subjected to a melt-curve analysis to validate the absence of nonspecific products. The data are presented as 1000 × 2^−∆CT in gene expression normalized to the geometric mean of RPL4 and ACTB (20) before statistical analysis.

Statistical analysis.

Data were analyzed by using the general linear model procedure of SAS, version 9.4, with dietary treatment considered as a main effect. The mean of pigs per pen was used for statistical analysis, and the pig pen served as the experimental unit (n = 4/treatment). Tukey’s post hoc adjustment was used for multiple comparisons between treatments. Differences were considered to be significant at P ≤ 0.05, and tendencies were discussed at P > 0.05–0.1. Values reported are least-squares means ± SEs. When unequal variances were observed (e.g., ³H-mannitol data), data were analyzed with the use of a PROC MIXED procedure in SAS to adjust for unequal variance (GROUP = option) and df (DDFM = SATTERTHWAITE option).

Results

Effects of dietary iron concentrations on pig iron status and growth performance.

Data regarding feed intake, performance, and iron status of the pigs used for this study were published previously by our group (15). To summarize, we demonstrated that pigs fed the L-Fe diet were anemic, whereas those fed the H-Fe diet had excessive iron absorption by the end of the study (day 32), as evidenced by hemoglobin concentrations, intestinal and hepatic iron concentrations, and mRNA expression of ferroportin, divalent metal transporter 1, and hepcidin (15).

Effects of dietary iron on gross and histopathologic appearance of porcine duodenum.

In the small intestine, the duodenum accounts for the majority of dietary iron absorption; thus, it is thought to be most affected by dietary iron concentration. Therefore, in the present study, we focused our experiment on the effects of iron on the duodenal mucosa. Upon gross examination, dietary iron concentration had an impact on the gross appearance of the duodenal mucosa. Duodenal mucosa from pigs fed the L-Fe diet exhibited a pale appearance compared with that of pigs fed the A-Fe control (Figure 1F, left), supporting the clinical anemia demonstrated in our previously published study (15). In contrast, duodenal mucosa from pigs fed the H-Fe diet exhibited a bright red hyperemic appearance (Figure 1F, right).

FIGURE 1.

Staining with Prussian blue (iron staining) of the duodenum of pigs fed L-Fe (20 mg Fe/kg DM) (A), A-Fe (120 mg Fe/kg DM) (B), or H-Fe (520 mg Fe/kg DM) (C) diets for 32 d (representative samples from n = 3 histologic stained sections). Hematoxylin and eosin staining of the duodenum of pigs fed A-Fe (D) and H-Fe (E) diets (arrows indicate neutrophil infiltration). Mucosal gross picture (from left to right: from pigs fed L-Fe, A-Fe, and H-Fe diets) (F), villus height (G), and neutrophil count (H). Values are least-squares means ± SEs, n = 4/treatment. Labeled means without a common letter differ, P < 0.05. A-Fe, adequate dietary iron; DM, dry matter; H-Fe, high dietary iron; L-Fe, low dietary iron.

To determine whether dietary iron concentrations were associated with histopathologic changes in the duodenum, we performed histologic analysis on hematoxylin and eosin–stained duodenal mucosal sections. Mucosal architecture was affected in H-Fe–fed pigs compared with those fed the A-Fe and L-Fe diets, indicated by reductions (P < 0.05) in villus height (Figure 1C, G). In addition to hematoxylin and eosin staining, mucosal tissue sections were stained with Prussian blue to determine the localization of iron within the intestinal mucosa and to confirm our previous studies regarding the concentration of iron within the mucosa (15). Prussian blue staining revealed a dose-dependent effect of dietary iron concentrations on iron staining within the duodenal mucosa. Compared with the A-Fe control diet, duodenum from pigs fed the H-Fe diet exhibited increased iron staining within the intestinal epithelial cells and subepithelial lamina propria cells, presumably macrophages (Figure 1A– C). In contrast, there was minimal staining of iron in the duodenum in the L-Fe diet.

Effect of dietary iron on oxidative stress and inflammation in the duodenum.

Excess iron reacts with hydrogen peroxide produced by neutrophils, resulting in hydroxyl radical production and mucosal damage by lipid peroxidation. Therefore, we investigated the effects of dietary iron concentration on neutrophil numbers and lipid peroxidation, measured as malondialdehyde. Duodenal mucosa from pigs fed the H-Fe diet exhibited increased (P < 0.05) neutrophil infiltration compared with those fed the A-Fe control (Figure 1E, H). Although a histologic evaluation of the duodenal mucosa of L-Fe–fed pigs revealed no observable changes in intestinal morphology, there were increased (P < 0.05) numbers of mucosal neutrophils observed compared with that of pigs fed the A-Fe diet (Figure 1H). Both the liver and the duodenal mucosal concentrations of malondialdehyde, a product of lipid peroxidation, increased in a dose-dependent manner with increased dietary iron intake (Figure 2A, B).

FIGURE 2.

Oxidative stress assessed as MDA concentration of the duodenum (A) and liver (B) from pigs fed L-Fe (20 mg Fe/kg DM), A-Fe (120 mg Fe/kg DM), or H-Fe (520 mg Fe/kg DM) diets for 32 d. Values are least-squares means ± SEs, n = 4/treatment. Labeled means without a common letter differ, P < 0.05. A-Fe, adequate dietary iron; DM, dry matter; H-Fe, high dietary iron; L-Fe, low dietary iron; MDA, malondialdehyde.

To further evaluate the influence of dietary iron concentration on intestinal inflammatory status, we assessed the relative gene expression of major proinflammatory cytokines. Compared with the A-Fe diet, the H-Fe treatment promoted gene expression of TNFA, IL1B, and IL6 (Figure 3A– C), and the L-Fe diet tended to increase the expression of IL1B and IL6 (P < 0.1). Gene expression of IL8, IL10, IFNG (Figure 3D– F), MCT7, and CMA1 (Supplemental Figure 1) were not affected by treatment. We also assessed the effects of dietary iron on proinflammatory cytokine gene expression in the liver and found no differences between treatment groups (Supplemental Figure 2).

FIGURE 3.

Relative mRNA expressions of TNFA (A), IL1B (B), IL6 (C), IL8 (D), IL10 (E), and IFNG (F) in the duodenal mucosa of pigs fed L-Fe (20 mg Fe/kg DM), A-Fe (120 mg Fe/kg DM), or H-Fe (520 mg Fe/kg DM) diets for 32 d. Values are least-squares means ± SEs, n = 4/treatment. Labeled means without a common letter differ, P < 0.05. A-Fe, adequate dietary iron; DM, dry matter; H-Fe, high dietary iron; IFNG, interferon γ; L-Fe, low dietary iron; TNFA, tumor necrosis factor α.

Effects of dietary iron on barrier function and baseline electrogenic ion transport in the porcine duodenum.

To determine whether dietary iron concentration affected intestinal barrier function, we mounted duodenal mucosa from pigs fed different dietary iron concentrations in Ussing chambers. We assessed intestinal permeability via the measurement of TER and mucosal to serosal flux of ³H-mannitol. Duodenal TER was reduced in pigs fed both the L-Fe (P < 0.05) and H-Fe (P < 0.1) diets compared with pigs fed the A-Fe diet (Figure 4A). In agreement with the TER measurement, the duodenum from pigs fed the L-Fe and H-Fe diets exhibited increased mucosal to serosal fluxes of ³H-mannitol compared with those fed the A-Fe control diet (P < 0.05; Figure 4B), thus confirming an increase in paracellular permeability associated with either iron deficiency or iron excess. Electrogenic ion transport, determined by transepithelial I_sc, was also measured in porcine duodenum. The duodenum from pigs fed the L-Fe and H-Fe diets exhibited higher baseline I_sc values than did those fed the A-Fe control diet (P < 0.05; Figure 4C), indicating that electrogenic ion transport processes were affected by dietary iron.

FIGURE 4.

Intestinal permeability assessed as TER (A) and ³H-mannitol permeability (B) and transepithelial I_sc (C) in the duodenum of pigs fed L-Fe (20 mg Fe/kg DM), A-Fe (120 mg Fe/kg DM), or H-Fe (520 mg Fe/kg DM) diets for 32 d. Values are least-squares means ± SEs, n = 4/treatment. Labeled means without a common letter differ, P < 0.05. A-Fe, adequate dietary iron; DM, dry matter; H-Fe, high dietary iron; I_sc, short-circuit current; L-Fe, low dietary iron; TER, transepithelial electrical resistance.

Discussion

Dietary iron plays a number of important roles in the body, including DNA synthesis and oxygen transport, which are essential for the growth and development of young animals. However, accumulating clinical evidence indicates that excessive iron supplementation can be detrimental to gastrointestinal health. The underlying mechanisms by which iron influences gastrointestinal health and growth remain poorly understood. In the present study, we demonstrated that both a deficiency and excess of dietary iron had substantial deleterious impacts on intestinal morphology, inflammation, and function.

In the present study, feeding pigs an H-Fe diet but not an L-Fe diet had a deleterious influence on the morphology of the proximal small intestine, as evidenced by a 25% reduction in villus height. These data might indicate that excess dietary iron more dramatically decreases the absorption capability of the duodenum relative to iron deficiency. Given that the inflammatory responses (cytokine gene expression and neutrophil numbers) were similar between the L-Fe and H-Fe treatments, this suggests that inflammation did not appear to be a common mechanism for altered morphology. However, the heightened lipid peroxidation and increased accumulation of iron within the enterocytes observed in the present study could have contributed to more epithelial damage and loss and thus reduced villi heights. Studies in rodent models have reported increases in duodenal and/or jejunal villus height in iron deficiency as a compensatory mechanism to increase intestinal iron absorption (21–23). This was not observed in pigs fed the L-Fe diet in the present study. However, experimental differences (e.g., duration of iron deficiency, age of animals, etc.), as well as species differences, might account for the differences between rodent studies and the present study.

In the present study, we observed an increased baseline I_sc in the duodenum of pigs fed either iron-deficient or iron-excess diets, indicative of increased transmucosal net electrogenic ion transport processes in the L-Fe and H-Fe treatment groups. Although an elevated duodenal I_sc most likely reflects the active secretion of anions such as Cl− and/or HCO₃⁻, the reason for an increased I_sc in the L-Fe and H-Fe diets is not clear at this time. There are a number of factors that drive electrogenic ion transport processes, including inflammation, irritant or antigen exposure, bacterial and viral toxins, and nutrient transporter activity (24). The increase of I_sc in pigs fed both the L-Fe and H-Fe diets suggests that this response is not due to a direct effect of iron concentration, but may be due to a common secondary pathway associated with iron deficiency and iron excess. For example, inflammation, cytokine synthesis, and increased intestinal permeability were observed in both the L-Fe and H-Fe diets in the present study, which is in line with the pattern of elevated I_sc. Inflammatory mediators, such as TNF-α, IL3, and IL6, have been shown to induce I_sc responses and thus may explain our current findings (25–27). In addition, excessive translocation of luminal antigens or bacteria as a result of increased intestinal permeability observed in both iron deficiency and iron excess could trigger persistently elevated intestinal secretion, thus explaining the elevated baseline I_sc.

Compared with pigs fed the A-Fe control diet, the L-Fe– and H-Fe–fed pigs had >3-fold and >3.3-fold higher neutrophil numbers (Figure 1H), respectively, in the duodenal mucosa. A similar pattern was observed with the gene expression of proinflammatory cytokines (Figure 3A–C). Together, these data indicate that appropriate dietary iron concentrations are critical for the balance of intestinal immune function and inflammation. In humans, the influence of iron on inflammation has become particularly relevant to the management of patients with IBD, in which iron supplementation is used to prevent iron deficiency (28). In the present study, we showed that excessive iron supplementation induced oxidative stress and lipid peroxidation in the intestine and liver. In agreement with our findings, Bernotti et al. (29) observed a linear increase in malondialdehyde concentrations in Caco-2 cells exposed to increasing amounts of iron (0–400 μM iron). Porres et al. (30) also found that indicators of lipid peroxidation in colonic mucosa were greatly increased in pigs fed diets containing 750 mg Fe/kg DM. The oxidative stress and subsequent lipid peroxidation could trigger intestinal inflammation by activating the NF-κB signaling pathway (31, 32), which may explain the present findings with the H-Fe diet in pigs. However, oxidant status does not appear to explain the inflammation observed in the pigs fed the L-Fe diet in this study, suggesting that there are different mechanisms at play for inflammation induced under iron-deficient and iron-excess states. It has been reported that iron deficiency promotes neutrophil hypersegmentation by the diminution of apoptosis (33), which could initiate a proinflammatory state in the immune system. Moreover, iron deficiency reduced hepcidin transcription, which exaggerated IL6 production of macrophages during LPS injection, because of the lack of the hepcidin-mediated anti-inflammatory response in mice (14). Furthermore, our previous work also indicated that liver hepcidin gene expression in these pigs closely followed liver iron concentration in that extremely low hepcidin mRNA was observed in iron deficient pigs (15).

The permeability of the gastrointestinal tract allows the selective barrier to absorb needed nutrients and prevent the penetration of harmful entities from the external environment, including pathogens and antigens. The breakdown in intestinal barrier function results in the translocation of luminal bacteria, toxins, and antigens into subepithelial tissues, which in turn exacerbate mucosal and systemic inflammatory responses. To our knowledge, only a few studies have investigated the influence of dietary iron on intestinal permeability. By using the ratio of urinary excretion of nonabsorbable sugar probes as an indicator, Berant et al. (34) reported that young children exhibited increased overall intestinal permeability during an iron-deficient state. In line with this study, we found duodenal TER to be decreased in anemic pigs fed the L-Fe diet compared with those fed the A-Fe diet. Moreover, we found that H-Fe–fed pigs also exhibited reduced TER, together demonstrating that iron deficiency or iron excess can impair epithelial barrier of the duodenal mucosa. Accordingly, feeding the L-Fe and H-Fe diets resulted in an increased duodenal permeability to the paracellular probe ³H-mannitol. The precise mechanisms by which iron deficiency or iron excess affected duodenal barrier function in the present study are unclear. Others have reported that the increased intestinal permeability might be due to reduced ATP generation from iron-deficiency hypoxia (35), and the ensuing anemia and ischemia might delay the rapid disposal of injurious substances such as oxygen free radicals, which could result in intestinal epithelial barrier damage (36). Given the anemic status of the pigs fed the L-Fe diet, it is possible that intestinal ischemia may be a potential contributing factor to intestinal injury. However, histologic analysis did not reveal any characteristic lesions associated with ischemic damage (Figure 1). More studies are needed to investigate the mechanisms of intestinal barrier injury triggered by low iron status. On the contrary, under conditions of high iron supplementation, oxidative stress has been linked with damage of cell membranes, mucosal necrosis, and reduction of microvillus height (37), which could directly or indirectly increase intestinal permeability (38, 39). The damaged intestinal barrier with increased permeability leads to epithelial translocation of antigens and pathogens (40), which could incite mucosal inflammatory responses, such as those observed in the present study. Although translocation of luminal antigens might occur with the impairment of barrier function associated with dietary iron concentration in the present study, we did not observe transcriptional changes of proinflammation cytokines the liver (Supplemental Figure 2). However, the concentrations of malondialdehyde were markedly elevated in the liver of H-Fe–fed pigs in the present study, indicating systemic effects from dietary iron concentrations. In addition, high iron concentrations can cause a proliferation of bacteria (41), which in turn could contribute to or exacerbate iron-induced intestinal injury. Taken together, it is likely that the intestinal barrier injury in pigs fed the H-Fe diet is a result of multiple mechanisms including inflammation, oxidative stress, and epithelial injury.

Given the increased supplementation of dietary iron in the diet of young pigs, as well as the translational role of dietary iron supplementation in the exacerbation of intestinal inflammatory disease in humans, results from the current study have important implications for the digestive health of both animals and humans. These data demonstrate that both iron deficiency and iron excess induce intestinal inflammation and increase intestinal permeability and electrogenic ion transport. Given the critical importance of intestinal barrier function to animal and human health, along with the concurrent stressors associated with weaning, the data suggest that dietary iron not only has effects on baseline gut health, but could influence susceptibility to other gastrointestinal diseases associated with intestinal barrier dysfunction in pigs and humans. Therefore, understanding the effects and mechanisms of dietary iron modulation of intestinal permeability and inflammatory response could affect feeding and therapeutic strategies and intestinal health outcomes in both pigs and humans.