Abstract

Iron absorption from the small intestine is regulated according to the body's needs, increasing in iron deficiency and decreasing in iron overload. It has been proposed that the efficiency of absorption is determined by the amount of iron acquired by developing enterocytes when they are in the crypts of Lieberkůhn and that this regulates expression of iron transporters such as DMT1 in mature enterocytes of the intestinal villi. In the crypts the cells take up iron from plasma transferrin by receptor-mediated endocytosis, a process that is influenced by the hemochromatosis protein, HFE. Hence, the availability of plasma transferrin-bound iron and the expression and function of transferrin receptors (TfR1), HFE and DMT1 should all contribute to the absorptive capacity of villus enterocytes. These aspects of the regulation and mechanism of iron absorption were investigated in genetically normal rats and mice, and in Belgrade anemic (b/b) rats and HFE knockout mice. In most experiments the function of the TfR1 was assessed by the uptake of radiolabeled transferrin-bound iron given intravenously. Absorption of non-heme iron was measured using closed in situ duodenal loops. The expression and cellular distribution of DMT1 and TfR1 were determined by in situ hybridisation and immunohistochemistry. The uptake of transferrin-bound iron and expression of functional TfR1 was shown to occur mainly in crypt cells and to be proportional to the plasma concentration of iron. It was not impaired by the mutation of DMT1 that occurs in b/b rats but was impaired in HFE knockout mice. Iron absorption was increased in these mice but was still influenced by the level of iron stores, as in normal mice. These results are in accordance with the proposed regulation of iron absorption and suggest that DMT1 is not the only iron transporter operating within endosomes of crypt cells. This view was supported by the failure to detect DMT1 mRNA or protein in crypt cells. Expression of DMT1 mRNA and protein started at the crypt-villus junction and increased to reach highest levels in the mid-villus region. Greater expression was found in iron deficiency and less in iron loaded animals than in controls and in the iron deficient rats most of the protein was present on the brush border membrane. In normal rats the efficiency of iron absorption parallelled the level of DMT1 expression, but in b/b rats absorption was very low and independent of dietary iron content even though DMT1 was present in villus enterocytes. The results confirm the essential role of DMT1 in the uptake phase of non-heme iron absorption. When normal rats previously fed a low iron diet were given a bolus of iron by stomach tube, the subsequent absorption of iron from a test dose placed in the duodenum diminished in parallel with the expression of DMT1 mRNA and protein, commencing within 1hour and reaching low levels by 7 hours. The margination of DMT1 to the brush border membrane disappeared. These results show the level of expression and intracellular distribution and function of DMT1 respond very quickly to the iron content of the diet as well as being affected by storage iron levels.