Recent advances in understanding of the regulation of iron absorption at a molecular level have implicated two gene products, haemochromatosis (Hfe) and hepcidin, as important negative regulators of iron absorption (Fleming & Sly, 2002; Nicolas et al, 2002). Mutations affecting Hfe that lead to loss of cell surface expression of the Hfe protein result in inappropriately enhanced iron absorption (Fleming & Sly, 2002). Iron absorption is regulated by multiple factors, including iron stores, erythroid iron requirements, inflammation and hypoxia, and recent work has suggested that the liver secretes a peptide, hepcidin, that acts as a regulatory ‘hormone’ for iron absorption common to all these factors (Nicolas et al, 2002). Evidence from knock-out mice suggests that this peptide is a negative regulator of intestinal iron absorption (Pigeon et al, 2001) that may be expressed at inappropriately low levels in haemochromatosis (Bridle et al, 2003). Central to understanding such a hypothesis is mapping out the interactions between the various regulators of iron absorption. We therefore set out to measure the hypoxic response of iron absorption in Hfe knock-out mice.
Mice (129/Ola–C57BL/6 mixed background strain) with a 2-kb pgk-neor gene flanked by loxP sites replacing a 2·5-kb BglII fragment (for details, see Bahram et al, 1999) were used as Hfe knock-outs. Heterozygotes were mated, and wild-type (wt) and homozygote Hfe knock-out littermates were identified at 4–5 weeks of age. Mice were fed a standard rodent diet (SDS, Witham, Essex, diet RM1; this diet contains 106 mg/kg Fe) ad libitum and studied at 8–12 weeks of age. Hypoxia was induced by placing mice in a hypobaric chamber set at 0·5 atmospheres for 3 d (Raja et al, 1988). All experiments were carried out under the authority of the UK Home Office. Methods and statistical analysis have been described in previous publications (Simpson, 1996).
Table I shows that Hfe knock-out mice had increased liver iron stores as reported previously (Bahram et al, 1999). No effects of Hfe genotype on body weight, haemoglobin or spleen iron were seen. In addition, hypoxia was found to have no significant effect on any of these parameters.
|Genotype||Treatment||Body weight (g)||Haemoglobin (g/dl)||Liver non-haem iron |
|Spleen non-haem iron (nmol/mg)|
|wt||Normal||26·3 ± 1·5||18·3 ± 0·4||1·4 ± 0·1||5·5 ± 0·4|
|Hypoxia||23·3 ± 1·1||19·3 ± 0·7||1·6 ± 0·2||5·2 ± 1·0|
|Hfe knock-out||Normal||24·6 ± 0·9||16·1 ± 1·1||5·5 ± 0·6||3·3 ± 0·7|
|Hypoxia||24·6 ± 0·4||17·5 ± 1·8||4·9 ± 0·7||4·9 ± 2·3|
|P-value||Hypoxia||> 0·1||> 0·3||> 0·6||> 0·6|
|Genotype||> 0·8||> 0·1||< 0·001||> 0·4|
Table II shows that there was no significant difference in the rate of iron absorption between the genotypes. No difference was observed in either uptake of luminal iron to the duodenal mucosa or transfer of iron to the carcass. This is consistent with the relatively modest increase in iron stores seen in these mice, compared with other strains (Fleming et al, 2001; Dupic et al, 2002). Dupic et al (2002) have shown that adult mice of the C57BL/6 strain show no significant increase in iron absorption gene products such as duodenal cytochrome B (Dcytb), divalent metal transporter 1 (DMT1; SLC11A2) and iron-regulated protein 1 (Ireg1; SLC11A3) in response to deletion of the Hfe gene. Presumably, these knock-outs are able to downregulate these genes in response to their increased iron stores and regain a ‘steady state’ in iron metabolism, albeit with higher iron stores than wt. This finding means that these mice could be useful for the study of the interaction of iron absorption regulators with Hfe, as the absence of the gene leads to relatively modest iron loading (other strains can show 10-fold increases in liver iron). Hence, there is little potential interference from variation in iron stores, and there remains a capacity for increased absorption.
|Genotype||Treatment||Mucosal retention (pmol/mg)||Carcass transfer (pmol/mg)||Total mucosal uptake (pmol/mg)|
|wt||Normal||9·3 ± 1·3||47·6 ± 3·2||56·9 ± 2·1|
|Hypoxia||13·6 ± 1·3||73·7 ± 10·6||87·3 ± 9·7|
|Hfe knock-out||Normal||10·1 ± 1·0||30·2 ± 5·5||40·3 ± 5·|
|Hypoxia||17·4 ± 1·6||74·1 ± 19·3||91·5 ± 19·4|
|P-value||Hypoxia||< 0·001||< 0·05||< 0·005|
|Genotype||> 0·1||> 0·4||> 0·6|
Table II shows that hypoxia causes an increased iron absorption in both wt and Hfe knock-out mice, with a similar increase being seen in both genotypes, suggesting that the Hfe gene is not required for the regulation of iron absorption by hypoxia. Similar increases in both uptake of iron from the lumen and transfer of iron to the body suggest that there is comparable regulation of the putative carriers DMT1 (for uptake) or Ireg1 (for transfer) in hypoxia in the presence or absence of Hfe.
The present data contrast with our earlier report that iron absorption was enhanced by Hfe gene knock-out (Bahram et al, 1999), although a reduction in duodenal iron absorption has been found when hepatic iron overload builds up over longer periods of time (Lebeau et al, 2002). The subsequent breeding of this gene knock-out into a background containing more C57BL/6 seems to have reduced the effect of Hfe gene knock-out on iron absorption. Enhanced iron stores are clearly present (the mice used in the present study were also fed a different diet from those in the previous work). Ajioka et al (2002) have demonstrated that age-dependent downregulation of iron absorption in Hfe knock-out mice leads to a situation in which the enhanced liver iron stores equalize the iron absorption rate of older wt and Hfe knock-out mice. Downregulation of iron absorption in response to increased hepatic iron stores may thus explain why there are no differences in iron absorption in our Hfe knock-out mice. This equality of iron absorption has the advantage of providing a useful model for studies of enhancement of iron absorption in Hfe knock-out mice.
Ajioka et al (2002) demonstrated that 129/SvEvTac strain Hfe knock-out mice retain the capacity to increase iron absorption when erythropoiesis is stimulated by phenylhydrazine treatment. Hypoxia is reported to stimulate iron absorption independently of enhanced erythropoiesis (Raja et al, 1988) or alterations in iron stores (Raja et al, 1990; Simpson, 1996). Our data show that hypoxia can increase iron absorption in the absence of the Hfe gene product. It therefore seems that various stimuli for iron absorption, including hypoxia, enhanced erythropoiesis and altered iron stores, can all function in Hfe knock-out mice.
The finding that hepcidin is downregulated by hypoxia, erythropoiesis and iron stores (Pigeon et al, 2001; Nicolas et al, 2002) suggests that hepcidin may be a common final regulator for iron absorption for these stimuli. Hfe seems to function only to modulate the response to iron stores. Our data support this model for the regulation of iron absorption. In particular, our data show that hypoxia can regulate iron absorption in the absence of Hfe. The molecular basis for the interactions between the various regulators remains to be established.