Systemic regulation of Hephaestin and Ireg1 revealed in studies of genetic and nutritional iron deficiency

Hephaestin is a membrane-bound multicopper ferroxidase necessary for iron egress from intestinal enterocytes into the circulation. Mice with sex-linked anemia (sla) have a mutant form of Hephaestin and a defect in intestinal basolateral iron transport, which results in iron deficiency and anemia. Ireg1 (SLC11A3, also known as Ferroportin1 or Mtp1) is the putative intestinal basolateral iron transporter. We compared iron levels and expression of genes involved in iron uptake and storage in sla mice and C57BL/6J mice fed iron-deficient, iron-overload, or control diets. Both iron-deficient wild-type mice and sla mice showed increased expression of Heph and Ireg1 mRNA, compared to controls, whereas only iron-deficient wild-type mice had increased expression of the brush border transporter Dmt1. Unlike iron-deficient mice, sla mouse enterocytes accumulated nonheme iron and ferritin. These results indicate that Dmt1 can be modulated by the enterocyte iron level, whereas Hephaestin and Ireg1 expression respond to systemic rather than local signals of iron status. Thus, the basolateral transport step appears to be the primary site at which the small intestine responds to alterations in body iron requirements.     

Influence of short-chain fatty acids on iron absorption by proximal colon

BACKGROUND: Short-chain fatty acids produced by bacterial fermentation in the colon enhance the local absorption of cations, such as calcium, that could be used to improve the bioavailability of iron if a significant colonic absorption of iron were to occur. METHODS: Iron (iron gluconate, 100 microM) absorption by the caecum of the rat was compared with that in proximal sites of the small bowel using the Ussing chamber model; the influence of probiotic bacteria (Propionibacterium freudenreichii) on iron absorption was assessed and compared with that of two of their fermentation products (acetic and propionic acids) using the Ussing chamber and the ligated colon with gamma emitting iron as experimental models. RESULTS: The caecum absorbed less iron than the duodenum, but significantly more than the jejunum and ileum. This occurred mainly through an enhanced mucosal transfer of iron uptake. Propionibacteria enhanced iron absorption from the proximal colon; the same effect was observed in the presence of viable bacteria, or the culture medium free of viable bacteria, or acetate and propionate or propionate alone. CONCLUSIONS: The proximal colon could be a significant site available for iron absorption; this absorption can be enhanced by local production of short-chain fatty acids such as propionate.     

Absorptive and motor components of the antidiarrhoeal action of loperamide: an in vivo study in pigs.

The effects of loperamide (0.1 mg/kg orally) on net colonic water absorption, orocolonic transit time, and intestinal motility were investigated in pigs chronically fitted either with two cannulas in the proximal colon and a catheter in the duodenum and the ileum or with intraparietal electrodes on the duodenum, jejunum, caecum, and proximal colon and a duodenal catheter. Loperamide, given 20 minutes before a meal reduced significantly colonic net water absorption for 10 hours after eating. It also reduced colonic flow rate, increased orocolonic transit time, modified the postprandial intestinal motility by inducing supplementary phase 3 motor complexes and did not affect caecocolonic motility. Intraduodenal infusion of a hypertonic solution of mannitol (900 mOsm/l; 0.6 ml/minute) for the first postprandial hour strongly reduced or reversed net colonic water absorption, increased the colonic flow rate, accelerated the orocolonic transit, and induced profuse diarrhoea. After loperamide administration, all these effects were blocked and the relative colonic water absorption, expressed as the fraction of flow entering the colon, was strongly increased. Mannitol did not modify motility of the small and large intestine, and supplementary phase 3 motor complexes were observed when mannitol infusion was preceded by loperamide administration. It is concluded that in experimental osmotic diarrhoea loperamide causes a reduction in digesta flow entering into the colon, mediated by its action on small intestinal motility, and an increase in colonic water absorption.     

Influence of Short-Chain Fatty Acids on Iron Absorption by Proximal Colon

Background: Short-chain fatty acids produced by bacterial fermentation in the colon enhance the local absorption of cations, such as calcium, that could be used to improve the bioavailability of iron if a significant colonic absorption of iron were to occur. Methods: Iron (iron gluconate, 100 μM) absorption by the caecum of the rat was compared with that in proximal sites of the small bowel using the Ussing chamber model; the influence of probiotic bacteria ( Propionibacterium freudenreichii ) on iron absorption was assessed and compared with that of two of their fermentation products (acetic and propionic acids) using the Ussing chamber and the ligated colon with gamma emitting iron as experimental models. Results: The caecum absorbed less iron than the duodenum, but significantly more than the jejunum and ileum. This occurred mainly through an enhanced mucosal transfer of iron uptake. Propionibacteria enhanced iron absorption from the proximal colon; the same effect was observed in the presence of viable bacteria, or the culture medium free of viable bacteria, or acetate and propionate or propionate alone. Conclusions: The proximal colon could be a significant site available for iron absorption; this absorption can be enhanced by local production of short-chain fatty acids such as propionate.     

A rapid decrease in the expression of DMT1 and Dcytb but not Ireg1 or hephaestin explains the mucosal block phenomenon of iron absorption

BACKGROUND: A large oral dose of iron will reduce the absorption of a subsequent smaller dose of iron in a phenomenon known as mucosal block. Molecular analysis of this process may provide insights into the regulation of intestinal iron absorption. AIMS: To determine the effect of an oral bolus of iron on duodenal expression of molecules associated with intestinal iron transport in rats and to relate this to changes in iron absorption. METHODS: Rats were given an oral dose of iron and duodenal expression of divalent metal transporter 1 (DMT1), Dcytb, Ireg1, and hephaestin (Hp) was determined using the ribonuclease protection assay, western blotting, and immunofluorescence. Iron absorption was measured using radioactive (59)Fe. RESULTS: A decrease in intestinal iron absorption occurred following an oral dose of iron and this was associated with increased enterocyte iron levels, as assessed by iron regulatory protein activity and immunoblotting for ferritin. Reduced absorption was also accompanied by a rapid decrease in expression of the mRNAs encoding the brush border iron transport molecules Dcytb and the iron responsive element (IRE) containing the splice variant of DMT1. No such change was seen in expression of the non-IRE splice variant of DMT1 or the basolateral iron transport molecules Ireg1 and Hp. Similar changes were observed at the protein level. CONCLUSIONS: These data indicate that brush border, but not basolateral, iron transport components are regulated locally by enterocyte iron levels and support the hypothesis that systemic stimuli exert their primary effect on basolateral transport molecules.     

Distribution of the 10,000 molecular weight calcium binding protein along the small and large intestine of man.

The distribution of the 10,000 molecular weight calcium binding protein along the human small and large intestine was studied using an enzyme linked immunoadsorbent assay. Small intestinal mucosal samples were obtained from the duodenal bulb, the second and third part of the duodenum and at about 50 cm intervals from jejunum and ileum of five whole small intestines of necro-kidney donors. Mucosal samples of caecum, colon ascendens, and transversum were also investigated. The amount of calcium binding protein per milligram of cytosolic protein increased throughout duodenum to reach the maximum in the proximal segment of jejunum and then declined steadily to nearly undetectable levels in ileum. In the colon no 10,000 molecular weight CaBP was detectable. The distribution of CaBP along the small and large intestine of man is thus parallel to the efficiency of the active calcium absorption of human intestine.     

Novel genes and functional relationships in the adult mouse gastrointestinal tract identified by microarray analysis

BACKGROUND & AIMS: A genome-level understanding of the molecular basis of segmental gene expression along the anterior-posterior (A-P) axis of the mammalian gastrointestinal (GI) tract is lacking. We hypothesized that functional patterning along the A-P axis of the GI tract could be defined at the molecular level by analyzing expression profiles of large numbers of genes. METHODS: Incyte GEM1 microarrays containing 8638 complementary DNAs (cDNAs) were used to define expression profiles in adult mouse stomach, duodenum, jejunum, ileum, cecum, proximal colon, and distal colon. Highly expressed cDNAs were classified based on segmental expression patterns and protein function. RESULTS: 571 cDNAs were expressed 2-fold higher than reference in at least 1 GI tissue. Most of these genes displayed sharp segmental expression boundaries, the majority of which were at anatomically defined locations. Boundaries were particularly striking for genes encoding proteins that function in intermediary metabolism, transport, and cell-cell communication. Genes with distinctive expression profiles were compared with mouse and human genomic sequence for promoter analysis and gene discovery. CONCLUSIONS: The anatomically defined organs of the GI tract (stomach, small intestine, colon) can be distinguished based on a genome-level analysis of gene expression profiles. However, distinctions between various regions of the small intestine and colon are much less striking. We have identified novel genes not previously known to be expressed in the adult GI tract. Identification of genes coordinately regulated along the A-P axis provides a basis for new insights and gene discovery relevant to GI development, differentiation, function, and disease.     

Duodenal expression of iron transport molecules in untreated haemochromatosis subjects

BACKGROUND AND AIMS: In HFE associated hereditary haemochromatosis, the duodenal enterocyte behaves as if iron deficient and previous reports have shown increased duodenal expression of divalent metal transporter 1 (DMT1) and iron regulated gene 1 (Ireg1) in affected subjects. In those studies, many patients had undergone venesection, which is a potent stimulus of iron absorption. Our study investigated duodenal expression of DMT1 (IRE and non-IRE), Ireg1, hephaestin, and duodenal cytochrome-b (Dyctb) in untreated C282Y homozygous haemochromatosis patients, iron deficient patients, and iron replete subjects. METHODS: Total RNA was extracted from duodenal biopsies and expression of the iron transport genes was assessed by ribonuclease protection assay. RESULTS: Expression of DMT1 (IRE) and Ireg1 was increased 3-5-fold in iron deficient subjects compared with iron replete subjects. Duodenal expression of DMT1 (IRE) and Ireg1 was similar in haemochromatosis patients and iron replete subjects but in haemochromatosis patients with elevated serum ferritin concentrations, both DMT1 (IRE) and Ireg1 expression were inappropriately increased relative to serum ferritin concentration. Hephaestin and Dcytb levels were not upregulated in haemochromatosis. DMT1 (IRE) and Ireg1 levels showed significant inverse correlations with serum ferritin concentration in each group of patients. CONCLUSIONS: These findings are consistent with DMT1 (IRE) and Ireg1 playing primary roles in the adaptive response to iron deficiency. Untreated haemochromatosis patients showed inappropriate increases in DMT1 (IRE) and Ireg1 expression for a given level of serum ferritin concentration, although the actual level of expression of these iron transport genes was not significantly different from that of normal subjects.     

Hepcidin expression inversely correlates with the expression of duodenal iron transporters and iron absorption in rats

BACKGROUND & AIMS: Hepcidin is an antimicrobial peptide thought to be involved in the regulation of intestinal iron absorption. To further investigate its role in this process, we examined hepatic and duodenal gene expression in rats after the switch from a control diet to an iron-deficient diet. METHODS: Adult rats on an iron-replete diet were switched to an iron-deficient diet and the expression of iron homeostasis molecules in duodenal and liver tissue was studied over 14 days. Intestinal iron absorption was determined at these same time-points by measuring the retention of an oral dose of (59)Fe. RESULTS: Iron absorption increased 2.7-fold within 6 days of switching to an iron-deficient diet and was accompanied by an increase in the duodenal expression of Dcytb, divalent metal transporter 1, and Ireg1. These changes precisely correlated with decreases in hepatic hepcidin expression and transferrin saturation. No change in iron stores or hematologic parameters was detected. CONCLUSIONS: This study showed a close relationship between the expression of hepcidin, duodenal iron transporters, and iron absorption. Both hepcidin expression and iron absorption can be regulated before iron stores and erythropoiesis are affected, and transferrin saturation may signal such changes.     

Intestinal over-expression of iron transporters induces iron overload in birds in captivity

Hereditary hemochromatosis (HH) is a frequent genetic disease of older subjects of northern European descent. It is characterized by increased iron absorption and severe iron overloading in parenchymal organs. A similar disturbance of iron metabolism occurs in specific animal species in captivity. To address the key features leading to high absorption and thus to iron overload in these animals, we have studied the two iron transport proteins DMT1 and Ireg1 in the best-known susceptible species, the mynah bird. Here, we show that these birds have a high expression of DMT1 in the duodenum and also a strikingly high expression of Ireg1 along the whole small intestine. We believe that the iron accumulation in susceptible species only occurs in captivity because of a genotypic adaptation to their natural environment, where contrary to captivity, dietary iron is hardly available. The Caucasian population carrying mutations leading to iron overload today may have also benefited from the genetic advantage of up-regulating iron transport millennia ago, when dietary iron was scarce.     

The ceruloplasmin homolog hephaestin and the control of intestinal iron absorption

Hephaestin is the gene affected in the sex-linked anemic (sla) mouse. These animals have a defect in the export of iron from intestinal enterocytes into the circulation and this implicates hephaestin in the basolateral transfer step of iron absorption. Hephaestin is homologous to the plasma copper-containing protein ceruloplasmin, and all residues involved in copper binding and disulfide bond formation in ceruloplasmin are conserved in hephaestin. Unlike ceruloplasmin, hephaestin is an integral membrane protein with a single trans-membrane domain. It is highly expressed throughout the small intestine, to a lesser extent in the colon, and at low levels in several other tissues. Surprisingly, most hephaestin appears to be located intracellularly in a perinuclear distribution. Like ceruloplasmin, hephaestin has a ferroxidase activity which is predicted to underlie its biological function. In addition, its expression is stimulated under iron deficient conditions. Analysis of the sla mouse has supported our model for the regulation of intestinal iron absorption whereby changes in systemic iron requirements alter the levels of basolateral transport components with subsequent regulation of brush border transport.     

Absorption of Short-Chain Fatty Acids from the In-Situ-Perfused Caecum and Colon of the Guinea Pig

Oltmer S, Engelhardt W v. Absorption of short-chain fatty acids from the in-situ-perfused caecum and colon of the guinea pig. Scand J Gastroenterol 1994;29:1009-1016.

Background: Short-chain fatty acids (SCFA) originate from microbial fermentation of carbohydrates in the hindgut. Mechanisms involved in SCFA absorption are not fully understood.

Methods: The caecum and proximal and distal colon of the guinea pig were perfused in situ.

Results: SCFA absorption per gram mucosal dry weight was highest in the proximal colon and lowest in the caecum. Owing to the large surface, quantitatively SCFA absorption was highest from the caecum. In the distal colon clearance of SCFA increased with chain length (Ac < Pr < Bu); in the caecum and proximal colon only a comparatively small or no such influence was observed. Inhibition of the proton antiport systems in the apical membrane and inhibition of the carbonic anhydrase activity diminished SCFA absorption. The diffusive and carrier components of absorption were calculated for propionate at increasing propionate concentrations.

Conclusions: Marked segmental differences in SCFA absorption are apparent in the large intestine of the guinea pig.     

Colonic epithelial hPepT1 expression occurs in inflammatory bowel disease: transport of bacterial peptides influences expression of MHC class 1 molecules

BACKGROUND & AIMS: hPepT1 is an intestinal epithelial apical membrane transporter responsible for uptake of di/tripeptides (including bacterial derived proinflammatory n-formyl peptides). hPepT1 expression normally has a strict axial gradient-highest in the proximal small intestine with no expression in the colon. METHODS: Small intestinal-like cells (Caco2-BBE), and colonic-like cells (HT29-Cl.19A), and colonic mucosa from diseased and control patients were used in the present study. RESULTS: hPepT1 expression occurs aberrantly in the colon with chronic ulcerative colitis (6 patients) and Crohn's disease (4 patients), but not in normal colon (4 patients) or colon with microscopic colitis (4 patients). To model expression of hPepT1 by colonic-like cells in inflamed states, we stably transfected HT29-Cl.19A cells with a modified hPepT1 tagged on the N-terminus with green fluorescence protein. Analysis of transfected cells revealed that: GFP-hPepT1 protein, like the natural protein, is targeted to the apical plasma membrane. In addition, the tagged protein retains the capability of di/tripeptide absorption, and the expression of the tagged protein by HT29-Cl.19A cells permits absorption of N-formyl-methionyl-leucyl-phenylalanine (fMLP), as occurs in hPepT1 expressing Caco2-BBE cells. fMLP uptake by colonic cells expressing GFP-hPepT1 specifically enhances major histocompatibility complex class I surface expression. CONCLUSIONS: These data collectively indicate that, in some states of chronic inflammation, hPepT1 may be anomolously expressed in the colon. Further, transport of fMLP by hPepT1 potentially stimulates expression of key accessory immune molecule, MHC-1.     

Rat intestinal iron transfer capacity and the longitudinal distribution of its adaptation to iron deficiency

The longitudinal gradient of intestinal iron transfer was investigated in normal and iron-deficient male Sprague-Dawley rats in vitro and in vivo. In normal rats in vitro iron transfer in the duodenum was approximately 3 times higher than in the jejunum and decreased in the ileum to approximately half the jejunal values. Compared to the controls in vitro iron transfer was increased 3-4 times in the duodenum and in the first jejunal segment and 2-3 times in the second jejunal segment. No significant adaptation to iron deficiency was found in the rest of the small intestine. Iron transfer rates showed the same longitudinal pattern when iron was chelated with nitrilotriacetic acid (NTA) or with ascorbate. The absorbed iron quantities, however, were approximately 5 times lower when Fe-ascorbate was used, which might be due to differences in bioavailability. Omission of Fe-NTA and Fe-ascorbate had no impact on the vitality of the segments. Glucose transfer was used as vitality criterion. It was not significantly different between corresponding iron-deficient and control segments. To control these results in vivo mesenteric blood was collected from duodenal and jejunal segments in situ. Corresponding to in vitro findings iron transfer was close to linear over the experimental period. In iron deficient duodenal segments iron transfer increased approximately 3 times as compared to controls while no adaptational changes were found in the distal jejunum. No significant longitudinal gradient was found in the mucosal content of ferritin and nonheme iron. Both parameters were decreased in iron deficiency by about half. The mucosal transferrin content showed no longitudinal gradient in control animals. In iron deficiency transferrin was significantly increased in the duodenum and in the three most proximal jejunal segments. The results indicate that in rats adaptation of iron absorption to the demand can only be expected in the duodenum and in the proximal 20 cm of the jejunum. Because this process shows a steep gradient in the proximal small intestine, studies on the adaptation of intestinal iron transfer to the demand should use short and well-defined segments in order to provide reproducible results.     

Increased duodenal expression of divalent metal transporter 1 and iron-regulated gene 1 in cirrhosis

Hepatic hemosiderosis and increased iron absorption are common findings in cirrhosis. It has been proposed that a positive relation exists between intestinal iron absorption and the development of hepatic hemosiderosis. The current study investigated the duodenal expression of the iron transport molecules divalent metal transporter 1 (DMT1 [IRE]), iron-regulated gene 1 (Ireg1 [ferroportin]), hephaestin, and duodenal cytochrome b (Dyctb) in 46 patients with cirrhosis and 20 control subjects. Total RNA samples were extracted from duodenal biopsy samples and the expression of the iron transport genes was assessed by ribonuclease protection assays. Expression of DMT1 and Ireg1 was increased 1.5 to 3-fold in subjects with cirrhosis compared with iron-replete control subjects. The presence of cirrhosis per se and serum ferritin (SF) concentration were independent factors that influenced the expression of DMT1. However, only SF concentration was independently associated with Ireg1 expression. In cirrhosis, the expression of DMT1 and Ireg1 was not related to the severity of liver disease or cirrhosis type. There was no correlation between the duodenal expression of DMT1 and Ireg1 and the degree of hepatic siderosis. In conclusion, the presence of cirrhosis is an independent factor associated with increased expression of DMT1 but not Ireg1. The mechanism by which cirrhosis mediates this change in DMT1 expression has yet to be determined. Increased expression of DMT1 may play an important role in the pathogenesis of cirrhosis-associated hepatic iron overload.     

Cellular and subcellular localization of the Nramp2 iron transporter in the intestinal brush border and regulation by dietary iron.

Genetic studies in animal models of microcytic anemia and biochemical studies of transport have implicated the Nramp2 gene in iron transport. Nramp2 generates two alternatively spliced mRNAs that differ at their 3' untranslated region by the presence or absence of an iron-response element (IRE) and that encode two proteins with distinct carboxy termini. Antisera raised against Nramp2 fusion proteins containing either the carboxy or amino termini of Nramp2 and that can help distinguish between the two Nramp2 protein isoforms (IRE: isoform I; non-IRE: isoform II) were generated. These antibodies were used to identify the cellular and subcellular localization of Nramp2 in normal tissues and to study possible regulation by dietary iron deprivation. Immunoblotting experiments with membrane fractions from intact organs show that Nramp2 is expressed at low levels throughout the small intestine and to a higher extent in kidney. Dietary iron starvation results in a dramatic upregulation of the Nramp2 isoform I in the proximal portion of the duodenum only, whereas expression in the rest of the small intestine and in kidney remains largely unchanged in response to the lack of dietary iron. In proximal duodenum, immunostaining studies of tissue sections show that Nramp2 protein expression is abundant under iron deplete condition and limited to the villi and is absent in the crypts. In the villi, staining is limited to the columnar absorptive epithelium of the mucosa (enterocytes), with no expression in mucus-secreting goblet cells or in the lamina propria. Nramp2 expression is strongest in the apical two thirds of the villi and is very intense at the brush border of the apical pole of the enterocytes, whereas the basolateral membrane of these cells is negative for Nramp2. These results strongly suggest that Nramp2 is indeed responsible for transferrin-independent iron uptake in the duodenum. These findings are discussed in the context of overall mechanisms of iron acquisition by the body.     

Bone morphogenetic protein 2 is expressed by, and acts upon, mature epithelial cells in the colon

BACKGROUND AND AIMS: The recent findings of bone morphogenetic protein (BMP) receptor Ia mutations in juvenile polyposis and frequent Smad4 mutations in colon cancer suggest a role for BMPs in the colonic epithelium and colon cancer. We investigated the role of BMP2 in the colon. METHODS: We assessed BMP receptor expression in cell lines using the reverse-transcribed polymerase chain reaction and immunoblotting. We investigated the effect of BMP2 on cell lines using the MTT assay and by immunoblotting for markers of differentiation, proliferation, and apoptosis. We assessed the expression of BMP2, its receptors, and signal transduction elements in mouse and human colon tissue using immunohistochemistry. We also investigated the effect of the BMP antagonist noggin in vivo in mice by assessing colon tissue with immunohistochemistry and immunoblotting. Finally, we investigated the expression of BMP2 in microadenomas from familial adenomatous polyposis patients. RESULTS: BMP receptors (BMPR) Ia, BMPR Ib, and BMPR II are all expressed in colonic epithelial cell lines. BMP2 inhibits colonic epithelial cell growth in vitro, promoting apoptosis and differentiation and inhibiting proliferation. BMP2, BMPRIa, BMPRIb, BMPRII, phosphorylated Smad1, and Smad4 are expressed predominantly in mature colonocytes at the epithelial surface in normal adult human and mouse colon. Noggin inhibits apoptosis and proliferation in mouse colonic epithelium in vivo. BMP2 expression is lost in the microadenomas of familial adenomatous polyposis patients. CONCLUSIONS: These data suggest that BMP2 acts as a tumor suppressor promoting apoptosis in mature colonic epithelial cells.     

Iron Transport by Rat Small Intestine in Vitro: Effect of Body Iron Status

S ummary . Both mucosal uptake and serosal transfer of iron by everted sacs of rat small intestine were maximal in the duodenum and decreased progressively towards the ileum. Transport of iron against a concentration gradient did not occur in any part of the intestine. When the iron concentration in the bathing medium was varied over the range 10–500 μM a saturable transport process was revealed in the proximal small intestine. In iron-deficient rats mucosal uptake of iron was increased but in iron-loaded rats it remained normal. The results suggest that iron transport across the proximal small intestine of the rat in vitro involves a carrier mechanism which responds adaptively to iron deficiency but not to overload.     

Gene expression patterns of human colon tops and basal crypts and BMP antagonists as intestinal stem cell niche factors

Human colonic epithelial cell renewal, proliferation, and differentiation are stringently controlled by numerous regulatory pathways. To identify genetic programs of human colonic epithelial cell differentiation in vivo as well as candidate marker genes that define colonic epithelial stem/progenitor cells and the stem cell niche, we applied gene expression analysis of normal human colon tops and basal crypts by using expression microarrays with 30,000 genes. Nine hundred and sixty-nine cDNA clones were found to be differentially expressed between human colon crypts and tops. Pathway analysis revealed the differential expression of genes involved in cell cycle maintenance and apoptosis, as well as genes in bone morphogenetic protein (BMP), Notch, Wnt, EPH, and MYC signaling pathways. BMP antagonists gremlin 1, gremlin 2, and chordin-like 1 were found to be expressed by colon crypts. In situ hybridization and RT-PCR confirmed that these BMP antagonists are expressed by intestinal cryptal myofibroblasts and smooth muscle cells at the colon crypt. In vitro analysis demonstrated that gremlin 1 partially inhibits Caco-2 cell differentiation upon confluence and activates Wnt signaling in normal rat intestinal epithelial cells. Collectively, the expression data set provides a comprehensive picture of human colonic epithelial cell differentiation. Our study also suggests that BMP antagonists are candidate signaling components that make up the intestinal epithelial stem cell niche.