Reduced Expression of Ferroportin-1 Mediates Hyporesponsiveness of Suckling Rats to Stimuli That Reduce Iron Absorption

Transcription

1 GASTROENTEROLOGY 2011;141: Reduced Expression of Ferroportin-1 Mediates Hyporesponsiveness of Suckling Rats to Stimuli That Reduce Iron Absorption DEEPAK DARSHAN, SARAH J. WILKINS, DAVID M. FRAZER, and GREGORY J. ANDERSON Iron Metabolism Laboratory, Queensland Institute for Medical Research, PO Royal Brisbane Hospital, Brisbane, Australia BACKGROUND & AIMS: Suckling mammals absorb high levels of iron to support their rapid growth. In adults, iron absorption is controlled by systemic signals that alter expression of the iron-regulatory hormone hepcidin. We investigated whether hepcidin and absorption respond appropriately to systemic stimuli during suckling. METHODS: In Sprague Dawley rats, iron levels increased following administration of iron dextran, and inflammation was induced with lipopolysaccharide. Gene expression was measured by quantitative reverse-transcription polymerase chain reaction; protein levels were measured by immunoblot analyses. Iron absorption was determined based on retention of an oral dose of 59 Fe. RESULTS: Iron absorption was high during suckling and reduced to adult levels upon weaning. In response to iron dextran or lipopolysaccharide, iron absorption in adults decreased substantially, but, in suckling animals, the changes were minimal. Despite this, expression of hepcidin messenger RNA was strongly induced by each agent, before and after weaning. The hyporesponsiveness of iron absorption to increased levels of hepcidin during suckling correlated with reduced or absent duodenal expression of ferroportin 1 (Fpn1), normally a hepcidin target. Fpn1 expression was robust in adults. Predominance of the Fpn1A splice variant, which is under iron-dependent translational control, accounts for the low level of Fpn1 in the iron-deficient intestine of suckling rats. CONCLU- SIONS: Iron absorption during suckling is largely refractory to changes in expression of the systemic iron regulator hepcidin, and this in turn reflects limited expression of Fpn1 protein in the small intestine. Iron absorption is therefore not always controlled by hepcidin. Keywords: Development; Iron Homeostasis; Nutrition; Infant. An adequate supply of iron is essential for good health because this trace element plays a critical role in a wide range of biologic processes. Iron demand is especially high during infancy because of rapid growth and the expansion of the red cell compartment. Consequently, an adequate iron supply during this period is critical for normal development. Iron deficiency at this time of rapid neurodevelopment may have lifelong detrimental effects. 1 This high iron demand is met from a number of sources, including iron stored in the liver in the latter stages of gestation, the catabolism of fetal hemoglobin, and the absorption of iron from the diet. Intestinal iron absorption during suckling is extremely efficient 2 to enable the infant to scavenge as much iron as possible from breast milk, which has a relatively low iron content (from 0.5 g/ml in humans and dairy animals to 13.5 g/ml in rats 3 ). Around the time of weaning, iron absorption drops substantially to adult levels. 4 6 How iron traverses the intestine during suckling is poorly understood. Early studies suggested that iron absorption in neonates is a specific, saturable process and is largely restricted to the proximal small intestine, 4,6 ie, characteristics similar to that of iron absorption in adults. However, it has also been suggested that the general high permeability of the small intestine during suckling may contribute to iron absorption, ie, a nonspecific process. 7 We have recently shown that the distal small intestine makes a proportionally larger contribution to iron absorption during suckling than it does postweaning, 6 and this is the region where the highest permeability of the intestine is seen. 8 However, quantitatively, most absorption still occurs in the proximal small intestine. 6 Whatever the mechanism, a number of studies have shown that, in both rodents and humans, iron absorption during suckling is relatively refractory to stimuli, which would reduce absorption in adults. 7,9 In humans, iron supplementation of 6-month-old infants does not lead to a reduction in absorption, 10 whereas, as in adults, at 9 months, iron supplementation does suppress absorption. 11 Similar data have been obtained in rats where 10-day-old suckling animals show little response to an increased body iron load, whereas 20-day-old animals (around the time of weaning) respond by a reduction in the expression of iron transport proteins in the gut. 9 In adults, the basic components of the pathway by which non-heme iron is absorbed are relatively well defined. 12 Ferric iron in the diet must first be reduced to the ferrous form before it can be utilized. A candidate reductase on the apical or brush border membrane is duodenal cytochrome b. The transport of ferrous iron across the brush border membrane is mediated by divalent metal-ion Abbreviations used in this paper: BTF3, basic transcription factor 3; DcytB, duodenal cytochrome B; DMT1, divalent metal-ion transporter 1; Fpn1, ferroportin1; IRE, iron responsive element; IRP, iron regulatory protein; LPS, lipopolysacchride; mrna, messenger RNA; UTR, untranslated region by the AGA Institute /$36.00 doi: /j.gastro

2 July 2011 REDUCED INTESTINE FERROPORTIN DURING SUCKLING 301 transporter 1 (DMT1). Within the enterocyte, iron is either stored in ferritin or transported across the basolateral membrane by another ferrous iron transporter, ferroportin1 (FPN1). The efficient basolateral efflux of iron also requires an iron oxidase known as hephaestin. It is basolateral transport that is rate limiting for iron absorption, and this step is also the target of systemic signals for the regulation of absorption. These signals direct the expression of the liver-derived peptide hepcidin. Hepcidin in turn binds to ferroportin, leading to its internalization and degradation and thereby reducing absorption. 13 The goal of this study was to determine whether the hepcidin-ferroportin axis is functioning correctly during suckling. We have investigated the mechanism of high iron absorption in suckling rats and in particular how absorption responds to stimuli that repress iron absorption in normal adults. Hepcidin expression increased in response to iron loading and an inflammatory stimulus in both suckling and weaned animals, but only in the older animals did these stimuli decrease iron absorption. The hyporesponsiveness of absorption in the suckling animals can be explained by the demonstration that Fpn1, the target of hepcidin, was either absent or at very low levels during this period. This in turn reflects the pattern of expression of Fpn1 messenger RNA (mrna) splice variants in the intestine during suckling, which favors low Fpn1 synthesis in the iron-deficient conditions of the neonatal intestine. These results indicate that the mechanism of iron absorption during suckling is likely to be multifactorial and, at least in younger neonates, that the transport of iron appears to be largely Fpn1 independent. Materials and Methods Animals and Treatments Sprague Dawley rats were used for all experiments. The rats were maintained on a standard rodent pellet diet (Norco Stockfeeds, South Lismore, Australia; iron content, 160 mg/kg) and were allowed unlimited access to food and deionized water. Animals were weaned at 21 days unless noted otherwise. Some animals were iron loaded by a single injection of 0.3 mg/g iron-dextran (intraperitoneally; Sigma-Aldrich, Sydney, Australia) and killed 4 days later. Inflammation was induced by the injection of lipopolysaccharide (LPS) (0.1 mg/kg, intraperitoneally; Sigma-Aldrich), and the rats were processed for analysis 6 or 10 hours later for Hepcidin antimicrobial peptide (Hamp) or absorption measurements, respectively. Dexamethasone (0.4 g/g body weight/day, intraperitoneally; Sigma-Aldrich) was injected for 3 consecutive days, and animals were killed 24 hours after the last injection. Duodenal enterocytes were isolated as previously described, 14 and enterocytes and liver tissue were snap frozen in liquid nitrogen and stored prior to analysis. All experiments described in this study were approved by the Queensland Institute of Medical Research Animal Ethics Committee. Evaluation of Iron Absorption and Non-Heme Liver Iron Whole animal absorption measurements were carried out by giving rats an oral dose of 59 Fe followed by whole body counting as previously described. 14 Liver tissue was dried at 110 C, and the tissue non-heme iron content was determined colorimetrically as previously described. 15 Quantitative Reverse-Transcription Polymerase Chain Reaction Total RNA was extracted from duodenal enterocytes and liver using TRIzol reagent (Invitrogen, Melbourne, Australia) as per the manufacturer s instructions. Complementary DNA was synthesized using an oligo(dt) primer and Moloney Murine Leukemia Virus reverse transcriptase (Invitrogen). Real-time polymerase chain reaction was performed using LightCycler 480 SYBR Green I Master Mix (Roche, Sydney, Australia) in an LC480 machine (Roche). The data were analyzed by the comparative threshold cycle (C T ) method (2 C T) and were normalized to basic transcription factor 3 (BTF3). The primers used were as follows: Hamp1: forward, GCTGCCTGTCTCCT- GCTTCT; reverse, CTGCAGAGCCGTAGTCTGTCTCGTC. BTF3: forward, TGGCAGCAAACACCTTCACC; reverse, AGC- TTCAGCCAGTCTCCTCAAAC. Fpn1A: forward, AAAGAA- GACCCCGGTGGCAGC; reverse, GGCCAAGGTAGAGGAG- GAATTT. Fpn1B: forward, GTTGGTTGGAGTTTCAATGT- TG; reverse, GGCCAAGGTAGAGGAGGAATTT. Western Blot Analysis Protein was extracted from duodenal enterocytes, and the expression of Fpn1 was determined by Western blotting as described previously. 14,16 Polyclonal antibodies to Fpn1 were raised in rabbits against the peptide CGPDAKEV- RKENQANTSVV that corresponds to the C-terminal amino acids of the human protein and 68% identical to the rat sequence. Two commercially available anti-fpn1 antibodies, one raised to the largest extramembrane loop of the human protein and 100% identical to rat sequence (No. ab58695; Abcam, Cambridge, UK) and the other to the C-terminus of murine Fpn1, which is 90% identical to rat (No. MTP11-A; Alpha Diagnostic International, San Antonio, TX), were also used. A rabbit antiactin antibody (Sigma- Aldrich) was used as a loading control. Statistical Analysis All values are expressed as mean standard error of mean. Statistical differences between means were calculated with PASW Statistics 17.0 (SPSS Inc, Chicago, IL) by using the analysis of variance with Tukey s post hoc test. Results Iron Absorption During Suckling and After Weaning To confirm that iron absorption was showing the expected pattern during the suckling-weaning transition,

3 302 DARSHAN ET AL GASTROENTEROLOGY Vol. 141, No. 1 absorption, but the levels remained lower than those of the 15-day-old pups (70% vs 90%, respectively) (Figure 2C). Thus, in terms of iron status and absorption, the weaned rats on an iron-deficient diet were comparable with suckling rats. We next examined the expression of genes encoding proteins involved in duodenal iron transport because this is the principle site of iron absorption in both suckling animals and adults. In spite of their slightly higher iron absorption and similar hepcidin levels, the expression of Dmt1, duodenal cytochrome B (DcytB), hephaestin, and total Fpn1 mrna was significantly lower in the suckling animals than in iron-deficient weaned rats (Figure 3A D). In fact, for each of the genes studied, expression was lower than that of 25-day-old rats on a Figure Fe absorption during suckling and after weaning. Rats at the postnatal ages indicated were gavaged with 3 Ci of 59 Fe/animal and intestinal iron absorption determined. Iron absorption is presented as the percentage of radioactivity retained by the body after 5 days, relative to the initial dose. Data represent mean standard error of mean; n 5 10 rats/group. iron absorption was measured daily (separate cohorts of animals for each day) in rat pups from postnatal days 15 to 21 then at 23 and 28 days of age. From days 15 to 18, absorption was very high and ranged from 80% to 90%, but, around the time of weaning, it dropped dramatically to reach adult levels of approximately 20% (Figure 1). Rodent pups begin to supplement their milk diet with solid food several days before weaning, and this corresponded with the drop in iron absorption. Comparison of Iron Deficiency in Adults to Iron Metabolism During Suckling Neonatal mammals are in a relatively iron-deficient state as iron is being utilized to satisfy the demands of rapid growth and expansion of the erythroid mass. Because iron absorption is very high under iron-deficient conditions, we first examined whether iron deficiency alone could explain the high iron absorption associated with suckling. To achieve this, we compared 15-day-old suckling rats with 25-day-old rats that had been weaned onto an iron-deficient diet. As a control group, 25-day-old animals were weaned onto a normal diet. The hepatic iron concentration of 15-day-old animals was comparable with that of 25-day-old rats fed an iron-deficient diet but was 11-fold lower than that of weaned animals on a normal diet (Figure 2A). Hepcidin expression in the liver was relatively high in iron replete weaned rats but was undetectable in either iron deficient weanlings or suckling animals (Figure 2B). As seen previously, 59 Fe absorption at 15 days was higher when compared with control adults. Iron-deficient weaned rats also had greatly increased iron Figure 2. Comparison of iron homeostasis in iron deficient weaned rats and suckling animals. (A) Hepatic iron concentration, (B) hepcidin expression, and (C) 59 Fe absorption were measured. Data represent mean standard error of mean. Bars with the same letters are significantly different (P.05); n 5 6 rats/group.

4 July 2011 REDUCED INTESTINE FERROPORTIN DURING SUCKLING 303 control diet. These results indicate that the high iron absorption associated with suckling cannot be explained simply by a state of extreme iron deficiency. Regulation of Iron Absorption in Response to Iron Loading and Inflammation in Neonates The data described above suggest that the expression of iron-related genes in the intestine during suckling is not reflecting body iron demand. To investigate this further, we examined whether hepcidin and iron absorption respond to systemic stimuli during suckling as they do in adults. Because iron absorption is already very high during suckling, we chose 2 stimuli that would be expected to increase hepcidin levels and decrease iron absorption, iron loading, and inflammation. 12 These treatments were effective as shown by the reduction in transferrin saturation following LPS treatment (Figure 4A) and the increase in hepatic iron level following iron dextran treatment (Figure 4B). As expected, hepcidin transcript levels increased dramatically after treating the animals with either iron-dextran or LPS (Figure 4C) in both suckling and weaned rats. Hamp expression peaks at 6 hours after LPS treatment, followed by iron absorption changes several hours later. These data show that, not only is hepcidin able to respond appropriately in suckling animals, but, if anything, the response was more robust than it was in weanlings (Figure 4C). In adults, iron absorption is inversely related to hepcidin expression, and, when iron absorption was measured in weaned animals, it was significantly decreased following both LPS and irondextran treatment as expected (Figure 4D). However, suckling rats showed no change in iron absorption after LPS treatment and only a modest decrease in absorption in response to iron-dextran treatment, even in the face of very high hepcidin levels (Figure 4D). These results demonstrate that, during suckling, the liver is responding to systemic stimuli appropriately by up-regulating hepcidin but that this increased hepcidin is ineffective at blocking iron absorption from the intestine. Figure 3. Intestinal iron transporter gene expression in iron-deficient weaned rats and suckling animals. RNA was extracted from gut epithelial cells isolated from 25-day-old rats that had been weaned onto irondeficient or control diets and 15-day-old suckling animals, and gene expression was determined by quantitative real-time polymerase chain reaction. Expression of (A) total Fpn1, (B) Dmt1, (C) duodenal cytochrome B (DcytB), and (D) Hephaestin are shown relative to basic transcription factor 3 (BTF3). Data represent mean standard error of mean. Bars with the same letters are significantly different (P.05); n 6 rats/group. Expression and Regulation of Ferroportin During Suckling Because hepcidin responds normally during suckling but iron absorption does not, we investigated the major link between hepcidin and the absorptive machinery, the hepcidin target Fpn1. In the duodenum, Fpn1 mrna showed little or no response to either iron-dextran or LPS treatment in both weanling and suckling animals (Figure 5A). Because Fpn1 is regulated predominantly at the protein level in response to hepcidin binding, we next examined Fpn1 protein expression. Surprisingly, we were unable to detect any Fpn1 protein in the intestine of animals up to 15 days old. In the last few days of suckling, weak expression was detected, but, following weaning, the protein was abundant (Figure 5B). The initial results were obtained using an Fpn1 polyclonal antibody directed against the C-terminus of the protein. Using a second C-terminal antibody as well as a third antibody directed

5 304 DARSHAN ET AL GASTROENTEROLOGY Vol. 141, No. 1 against the major extramembrane loop of Fpn1, identical results were obtained. The absence of its principal target for the majority of the suckling period explains why high hepcidin levels are not translated into an absorption response during suckling. Fpn1 mrna is readily detected in the suckling intestine so what is the basis for the lack of protein expression? To address this, we examined the splice variants of Fpn1. Two Fpn1 splice variants were recently described that differ in the presence or absence of an iron responsive element (IRE) in the 5= untranslated region (5=UTR) of the transcripts 17 but have an identical coding region and produce a single product. The IRE-containing transcript (Fpn1A) was the dominant variant in neonates and constituted more than 90% of total Fpn1 transcript until day 18 (Figure 6A). Thereafter, the Fpn1 transcript lacking an IRE (Fpn1B), which was present at only very low levels in younger animals, increased in expression and maintained a steady level of about 30% of the total Fpn1 mrna into adulthood (Figure 6A). When mrnas have an IRE in their 5=UTR, translation is blocked under iron deficient conditions. 18 Because suckling animals are iron deficient, their intestinal epithelium would also be predicted to be iron deficient. This was confirmed by the very low expression of ferritin protein, another gene translationally repressed via a 5= IRE sequence (Figure 5C). Thus, translation from the dominant Fpn1A splice variant would be blocked. Translation from the Fpn1B variant could continue under these conditions, but its levels are so low that the amount of Fpn1 protein produced would be very low. Consistent with this result, we observed that Fpn1 protein levels in the enterocytes correlated with the Fpn1B transcript level. When Fpn1B mrna levels rose around day 19, we began to detect Fpn1 protein in the enterocytes (Figures 5B and 6). Ferritin protein expression also increased with age in parallel with that of Fpn1 (Figure 5C). Figure 4. Regulation of iron absorption and homeostasis in response to iron loading and inflammation in suckling and weaned animals. Suckling animals (15 days old) and 25-day-old weaned animals were treated with either iron dextran (0.3 mg/g) or LPS (0.1 mg/kg) and (A) transferrin saturation, (B) hepatic iron concentration, (C) hepatic Hamp messenger RNA levels, and (D) intestinal 59 Fe absorption were determined as described in the Materials and Methods section. Data represent mean standard error of mean. Bars with the same letters are significantly different (P.05); n 4 6 rats/group. Gray, neonate; black, adult animals. Intestinal Iron Transporters and Iron Absorption Can Be Modulated by Corticosteroid Treatment Our results and those of others indicate that there is a rapid transition in the gut from an immature phenotype to an adult-like phenotype around the time of weaning. 19 Our gene expression and iron absorption data indicate a similar transition in iron homeostasis in the intestine. To investigate gut maturity as a factor affecting iron homeostasis during the suckling-weaning period, we treated suckling animals with the synthetic cortisol analogue dexamethasone, which has been used extensively to mature organs in premature infants. 20 After the administration of dexamethasone to 15-day-old animals, there was an increase in the expression of sucraseisomaltase, a marker of gut maturity. Sucrase-isomaltase mrna reached a level similar to that found in 28-day-old weaned animals indicating that the gut was now behaving more like a mature organ (Figure 7A). Paralleling this increase in gut maturity, expression of the Fpn1B splice variant (without an IRE) was increased by dexamethasone

6 July 2011 REDUCED INTESTINE FERROPORTIN DURING SUCKLING 305 treatment, but Fpn1A expression did not change significantly (Figure 7B). Neither splice variant changed in 28- day-old animals in response to the treatment (Figure 7B). Fpn1 protein expression increased significantly in suckling animals but showed no change in adult rats in response to dexamethasone treatment (Figure 7C). Absorption of 59 Fe was not altered in 28-day-old animals after corticosteroid treatment, but we observed a significant decrease in iron absorption in dexamethasone-treated suckling animals (Figure 7D). These results strengthen the relationship between the control of iron absorption and the expression of the Fpn1B splice variant and demonstrate that maturity of the gut plays an important role in regulating iron homeostasis during the suckling-weaning period of infancy. Discussion In early postnatal life, the infant is reliant on its mother s milk to obtain all its nutrients, and its demand for these nutrients is very high during this time of rapid growth and development. As suckling becomes less important and the infant moves to solid foods, there are substantial changes in intestinal absorptive function. One of the nutrients that show significant changes during weaning is iron. During suckling, the small intestine extracts as much iron as possible from the relatively ironpoor breast milk, but iron absorption drops rapidly to adult levels upon weaning. 21 The molecular basis behind the high iron absorption during suckling and the changes that occur on weaning are poorly understood. In this study, we investigated whether the hepcidin-ferroportin axis was functioning normally during suckling. This is the major system responsible for controlling iron absorption in adults. 22 Surprisingly, although hepcidin was regulated normally in infant rats, the changes in hepcidin expression did not correspond with alterations in iron absorption. Subsequent experiments showed that greatly reduced expression of Fpn1, the enterocyte basolateral export protein that is the target of hepcidin, 13 could explain the hyporesponsiveness of iron absorption at this time. This appears to be the first description of intestinal iron absorption occurring through a pathway that is largely independent of the hepcidin-ferroportin axis. 4 Figure 5. Expression and regulation of Fpn1 in suckling animals. (A) RNA was extracted from duodenal enterocytes isolated from suckling or weaned rats following iron dextran or LPS administration, and Fpn1 messenger RNA levels were determined by quantitative real-time polymerase chain reaction. Protein was extracted from duodenal enterocytes obtained from animals of varying ages, and the level of (B) Fpn1 and (C) ferritin protein was determined by Western blotting. The lower panels show quantitation of protein levels normalized to the loading control actin. Data represent mean standard error of mean. Bars with the same letters are significantly different (P.05); n 5 6 rats/group. Gray, neonate; black, adult animals.

7 306 DARSHAN ET AL GASTROENTEROLOGY Vol. 141, No. 1 Figure 6. Relative abundance of Fpn1 splice variants throughout suckling and weaning. RNA was extracted from the intestine of rats of varying ages and Fpn1A( IRE) and Fpn1B( IRE) transcript levels were measured by quantitative real-time polymerase chain reaction. Data represent mean standard error of mean. n 5 12 rats/ group. Gray, Fpn1A( IRE); black, Fpn1B( IRE) transcripts. To understand better the mechanism of iron absorption during suckling, we investigated whether iron absorption could be regulated normally at this time. We chose the rat for these studies because we could precisely control the weaning time, collect samples from the liver (for the analysis of hepcidin expression), and manipulate the dietary iron content. Compared with humans, rat pups are relatively altricial at birth, and the first part of the suckling period corresponds to late gestation in humans. 19 However, the changes that occur during weaning are similar in both species. We first investigated whether simple iron deficiency could explain the high iron absorption of infant rats under similar mechanisms to those operating postweaning, but this was not the case. Infant rats certainly were iron deficient, but they showed relatively low expression of genes encoding key iron transport proteins in the proximal small intestine during suckling. In contrast, in animals that had been weaned onto an iron-deficient diet such that they had similar iron levels to 15-day-old suckling rats, the expression of iron transportrelated genes in the gut was greatly elevated. These data suggest that iron absorption during suckling does not simply represent an adaptation of the intestine to severe iron deficiency, at least not using the same mechanisms that operate in the gut after weaning. The major regulator of body iron traffic, including iron absorption, is the liver-derived peptide hepcidin, 22 but it is not known how hepcidin behaves during suckling or what contribution it makes to dietary iron intake. We found that, during suckling, hepcidin expression was very low, a situation consistent with high iron absorption because hepcidin acts as a repressor of absorption. To determine whether hepcidin could respond normally during suckling, we applied 2 stimuli that increase hepcidin expression in adults, the administration of either iron dextran or LPS. 23 In both cases, these agents were able to increase hepcidin expression in both suckling and weaned rats. In fact, if anything, the response was more robust in the younger animals. These studies showed that the machinery leading to alterations in hepcidin expression was functioning normally in the infants. In adults, an increase in hepcidin is associated with reduced iron absorption, 24 but this was not the case during suckling. This finding is consistent with earlier studies in rodents and humans that showed iron absorption during suckling to be relatively refractory to stimuli that would normally repress absorption in adults However, it does indicate that the small intestine is hyporesponsive to changes in hepcidin in infancy. Hepcidin acts on its target cells by binding to the iron export protein Fpn1 and facilitating its internalization and degradation. 13 The consequence of this is that iron export to the plasma by the target cell is reduced. In the case of the intestinal enterocytes, hepcidin removes Fpn1 from the basolateral membrane, and this is the basis for reduced iron absorption associated with increased hepcidin levels. To investigate why hepcidin did not appear to be having the expected effects on iron absorption, we examined Fpn1 expression in the suckling intestine. Although Fpn1 mrna levels were comparable before and after weaning, we made the surprising discovery that Fpn1 protein was either very low or undetectable during suckling. In time course studies, essentially no Fpn1 protein could be detected in the gut until the animals were around 15 days old. In the following few days, small amounts of the protein were found, and, by the time of weaning, essentially adult levels had been reached. Identical results were obtained whether we used an antibody directed against the C-terminus of the protein or to a major extramembrane loop. The limited expression of Fpn1 in the gut during suckling explains why iron absorption does not respond effectively to alterations in hepcidin expression. The deficiency of Fpn1 protein, even though the mrna is present, can be explained by the relative abundance of the 2 splice variants of the Fpn1 mrna and their mode of

8 July 2011 REDUCED INTESTINE FERROPORTIN DURING SUCKLING 307 Figure 7. Regulation of Fpn1 and iron absorption by corticosteroids. Animals were treated with dexamethasone (0.4 g/g body weight) at the ages indicated, and the expression of (A) sucrose-isomaltase, (B) Fpn1A( IRE) (black bars), and Fpn1B( IRE) (gray bars) was determined by quantitative real-time polymerase chain reaction and (C) Fpn1 protein levels by Western blotting. (D) Absorption of 59 Fe was also determined. Data represent mean standard error of mean. Bars with the same letters are significantly different (P.05); n 3 8 rats/group. regulation. Two Fpn1 transcripts have been described. 17 They both encode the same protein, but they differ in their 5= splicing such that 1 variant (Fpn1A) contains an IRE in its 5=UTR and 1 (Fpn1B) does not. The Fpn1A variant is subject to translational control. Under irondeficient conditions, iron regulatory proteins (IRPs) bind to the IRE and block translation, but, under iron-sufficient conditions, the IRPs do not bind and translation proceeds. 18 The same type of translational control is shown by the iron storage protein ferritin. We investigated the expression of the 2 Fpn1 transcripts during postnatal development and found that Fpn1A mrna dominated at all ages studied, with a relatively uniform expression through suckling and after weaning. The Fpn1B transcript was present at very low levels throughout much of the suckling period, but, from day 18 onward, message abundance increased to reach quite substantial levels at and beyond weaning. Because the intestinal epithelium is very iron deficient during suckling, as confirmed by the low ferritin protein level at this time, the IRE/IRP system would block translation from the dominant Fpn1A mrna. Protein could still be expressed from the Fpn1B transcript, but there is very little of this mrna present, so this explains why the protein is essentially undetectable. The appearance of Fpn1 protein parallels the increasing abundance of the Fpn1B transcript as postnatal development proceeds. Although our data suggest that iron absorption during suckling in the rat, particularly in younger animals, may be largely independent of Fpn1, some role for Fpn1 seems likely. We did detect Fpn1 in the last few days of suckling, and small amounts may have been present in younger animals that were below the limit of detection of the Western blotting procedure employed. The strongest evidence that Fpn1 is playing some role in intestinal iron transport during suckling comes from Fpn1 knockout mice. Global deletion of the Fpn1 gene is embryonic lethal in mice, and even an intestine-specific deletion using the villin promoter did not lead to viable offspring because of some expression of this promoter in critical embryonic tissues. 25 To overcome this, a mouse was generated where the villin-cre was under the control of a tamoxifen-inducible promoter. Induction of the Cre-mediated excision was begun when the animals were 7 days old, and, by day 15, the animals were visibly pale. 25 The same study also provided evidence of iron accumulation in the enterocytes of 12-day-old Fpn1 knockout mice. These data suggest that Fpn1 does indeed play some role in iron absorption during suckling, but no iron absorption experiments were carried out using the knockout mice, so further studies are required to determine whether the mice can absorb iron through an alternative pathway. If Fpn1 cannot fully explain iron transport across the intestine in early postnatal life, what other mechanisms might be involved? The high permeability of the intestine during suckling provides a possible explanation; however, this process is particularly prominent in the distal small intestine 8 and not in the duodenum and proximal jeju-

9 308 DARSHAN ET AL GASTROENTEROLOGY Vol. 141, No. 1 num where most iron absorption takes place. Furthermore, such a high capacity process does not explain why iron absorption during suckling appears saturable. 4 Lactoferrin or transferrin in breast milk could also play a role in iron absorption at this time, but mice in which the lactoferrin gene has been deleted do not become iron deficient, 26 and transferrin receptors are not expressed on the apical surface of enterocytes. 27 It is also feasible that the infant gut utilizes a novel iron transport pathway or expresses a variant (either a splice variant or a conformational variant) of Fpn1 that is not recognized by the antibodies we used in this study. If the latter was the case, this could reconcile our findings with the preliminary Fpn1 knockout mouse studies. 25 In addition to specific transport processes, physical factors such as the engagement of a greater length of the gut for absorption in neonates, 6 reduced intestinal transit time, or reduced turnover of enterocytes 28 could all potentially contribute to enhanced absorption. Overall, it is likely that the intestine of suckling mammals uses multiple mechanisms to ensure that it meets its iron requirements. The striking reduction in iron absorption that occurs upon weaning has yet to be fully explained. In our studies, we noted a close correlation between the switch from breast milk to solid food and changes in absorption. Young rats, even if they are still with their mother, will begin to take solid food from postnatal days 16 to 18, 29 and this correlates very well with the absorption changes and the changes in the expression of Fpn1B that we observed. It is well-known that treatment of animals with glucocorticoids will lead to premature maturation of the gut. 19 In our hands, dexamethasone treatment of suckling but not weaned rats led to a reduction in iron absorption and an increase in the Fpn1B transcript, consistent with steroid hormones playing an important role in the maturation response. We also found that premature weaning or delayed weaning would lead to early and late maturation of iron absorption respectively (data not shown). Early weaning also led to an increase in Fpn1B expression, consistent with our other data. Understanding the factors that govern the maturation of the iron absorption pathway has important implications for the design of effective infant formulae and feeding regimens. In summary, we have demonstrated that, despite appropriate regulation of hepcidin during suckling, intestinal iron absorption at this time is not readily regulated by stimuli that reduce iron absorption during adults. This is an effective mechanism for maximizing body iron supply in the infant when the demand for iron is very high. To our knowledge, this is the first demonstration of a situation in which increased hepcidin is ineffective at reducing iron absorption. Reduced Fpn1 protein expression appears to underlie this hyporesponsivevess, suggesting that at least some component of iron absorption during suckling is Fpn1 independent. References 1. Lozoff B, Georgieff MK. Iron deficiency and brain development. Semin Pediatr Neurol 2006;13: Domellof M. Iron requirements, absorption and metabolism in infancy and childhood. Curr Opin Clin Nutr Metab Care 2007;10: Ezekiel E, Morgan EH. Milk iron and its metabolism in the lactating rat. J Physiol 1963;165: Gallagher ND, Mason R, Foley KE. Mechanisms of iron absorption and transport in neonatal rat intestine. Gastroenterology 1973; 64: Saarinen UM, Siimes MA. Iron absorption from breast milk, cow s milk, and iron-supplemented formula: an opportunistic use of changes in total body iron determined by hemoglobin, ferritin, and body weight in 132 infants. Pediatr Res 1979;13: Frazer DM, Wilkins SJ, Anderson GJ. Elevated iron absorption in the neonatal rat reflects high expression of iron transport genes in the distal alimentary tract. Am J Physiol Gastrointest Liver Physiol 2007;293:G525 G Ezekiel E. Intestinal iron absorption by neonates and some factors affecting it. J Lab Clin Med 1967;70: Clark SL Jr. The ingestion of proteins and colloidal materials by columnar absorptive cells of the small intestine in suckling rats and mice. J Biophys Biochem Cytol 1959;5: Leong WI, Bowlus CL, Tallkvist J, et al. Iron supplementation during infancy effects on expression of iron transporters, iron absorption, and iron utilization in rat pups. Am J Clin Nutr 2003; 78: Domellof M, Lonnerdal B, Abrams SA, et al. Iron absorption in breast-fed infants: effects of age, iron status, iron supplements, and complementary foods. Am J Clin Nutr 2002;76: Hicks PD, Zavaleta N, Chen Z, et al. Iron deficiency, but not anemia, up-regulates iron absorption in breast-fed Peruvian infants. J Nutr 2006;136: Anderson GJ, Frazer DM, McLaren GD. Iron absorption and metabolism. Curr Opin Gastroenterol 2009;25: Nemeth E, Tuttle MS, Powelson J, et al. Hepcidin regulates cellular iron efflux by binding to ferroportin and inducing its internalization. Science 2004;306: Frazer DM, Wilkins SJ, Becker EM, et al. Hepcidin expression inversely correlates with the expression of duodenal iron transporters and iron absorption in rats. Gastroenterology 2002;123: Torrance JD, Bothwell TH. A simple technique for measuring storage iron concentrations in formalinised liver samples. S Afr J Med Sci 1968;33: Frazer DM, Vulpe CD, McKie AT, et al. Cloning and gastrointestinal expression of rat hephaestin: relationship to other iron transport proteins. Am J Physiol Gastrointest Liver Physiol 2001;281: G931 G Zhang DL, Hughes RM, Ollivierre-Wilson H, et al. A ferroportin transcript that lacks an iron-responsive element enables duodenal and erythroid precursor cells to evade translational repression. Cell Metab 2009;9: Muckenthaler MU, Galy B, Hentze MW. Systemic iron homeostasis and the iron-responsive element/iron-regulatory protein (IRE/IRP) regulatory network. Annu Rev Nutr 2008;28: Pacha J. Development of intestinal transport function in mammals. Physiol Rev 2000;80: ACOG committee opinion. Antenatal corticosteroid therapy for fetal maturation. American College of Obstetricians and Gynecologists. Int J Gynaecol Obstet 2002;78: Collard KJ. Iron homeostasis in the neonate. Pediatrics 2009;123:

10 July 2011 REDUCED INTESTINE FERROPORTIN DURING SUCKLING Nemeth E, Ganz T. The role of hepcidin in iron metabolism. Acta Haematol 2009;122: Pigeon C, Ilyin G, Courselaud B, et al. A new mouse liver-specific gene, encoding a protein homologous to human antimicrobial peptide hepcidin, is overexpressed during iron overload. J Biol Chem 2001;276: Nicolas G, Bennoun M, Porteu A, et al. Severe iron deficiency anemia in transgenic mice expressing liver hepcidin. Proc Natl Acad Sci U S A 2002;99: Donovan A, Lima CA, Pinkus JL, et al. The iron exporter ferroportin/slc40a1 is essential for iron homeostasis. Cell Metab 2005; 1: Ward PP, Mendoza-Meneses M, Cunningham GA, et al. Iron status in mice carrying a targeted disruption of lactoferrin. Mol Cell Biol 2003;23: Anderson GJ, Walsh MD, Powell LW, et al. Intestinal transferrin receptors and iron absorption in the neonatal rat. Br J Haematol 1991;77: Al-Nafussi AI, Wright NA. Cell kinetics in the mouse small intestine during immediate postnatal life. Virchows Arch B Cell Pathol Incl Mol Pathol 1982;40: Thiels E, Alberts JR, Cramer CP. Weaning in rats: II. Pup behavior patterns. Dev Psychobiol 1990;23: Received December 24, Accepted April 5, Reprint requests Address requests for reprints to: Gregory J. Anderson, PhD, Iron Metabolism Laboratory, Queensland Institute of Medical Research, PO Royal Brisbane Hospital, Brisbane, Queensland 4029, Australia. Greg.Anderson@qimr.edu.au; fax: (61) Acknowledgments D.D. and S.J.W. contributed equally to this work. Conflicts of interest The authors disclose no conflicts. Funding Support by a Project Grant (552430) from the National Health and Medical Research Council of Australia (NHMRC) and by a Senior Research Fellowship from the NHMRC (to G.J.A.).