The renal hepcidin/ferroportin axis controls iron reabsorption and determines renal and hepatic susceptibility to iron overload

The hepcidin/ferroportin axis controls systemic iron homeostasis by regulating iron acquisition from the duodenum and the reticuloendothelial system, respective sites of iron absorption and recycling. Ferroportin is also abundant in the kidney, where it has been implicated in iron reabsorption. However, it remains unknown whether hepcidin regulates ferroportin-mediated iron reabsorption and whether such regulation is important for systemic iron homeostasis. To address these questions, we generated a novel mouse model with an inducible renal-tubule specific knock-in of fpnC326Y, which encodes a hepcidin-resistant FPNC326Y. Under iron-replete conditions, female mice harbouring this allele had lower renal iron content and higher serum and liver iron levels than controls. Under conditions of excess iron availability, male and female mice harbouring this allele had greater liver iron overload, but lower renal iron overload relative to controls. In addition, hemochromatosis mice harbouring a ubiquitous knock-in of fpnC326Y did not develop renal iron overload otherwise seen in the setting of excess iron availability. These findings are the first formal demonstration that hepcidin regulates ferroportin-mediated iron reabsorption. They also show that loss of this regulation contributes to liver iron overload while protecting the kidney in the setting of hemochromatosis. Our findings have important implications. First, they indicate that targeting the hepcidin/ferroportin axis for treating iron overload disorders will inhibit iron reabsorption and increase renal iron content. Second, they suggest that inhibition of iron reabsorption by raised hepcidin in chronic inflammatory conditions contributes to iron deficiency and that parenteral iron supplementation in this setting may cause renal iron overload.


INTRODUCTION
Ferroportin (FPN) is the only known mammalian iron export protein. It mediates iron release into the circulation from duodenal enterocytes and splenic reticuloendothelial macrophages, the respective sites of iron absorption and recycling (1,2). FPN-mediated iron release is antagonized by the hormone hepcidin, also known as hepcidin antimicrobial peptide (HAMP). Produced primarily in the liver, hepcidin binds to and induces internalization of FPN, thereby limiting iron release into the circulation and its availability to peripheral tissues (3,4). Thus, the HAMP/FPN axis operates at the sites of absorption and recycling to control systemic iron homeostasis.
FPN is also abundant in the kidney, the site of iron reabsorption (5)(6)(7)(8). In this setting, it has been reported that FPN contributes to iron reabsorption by exporting filtered iron from renal tubules into the circulation (8). However, it remains unknown if renal FPN is also subject to regulation by HAMP, and if so, whether such regulation is important for systemic iron homeostasis.
To address these questions, we generated a novel mouse model with an inducible renaltubule specific knock-in of fpnC326Y, which encodes a HAMP-resistant FPNC326Y protein.
Additionally, to confirm the previously reported role of FPN in iron reabsorption, we also generated a mouse model with an inducible renal-tubule specific deletion of the fpn gene.
We found that renal iron content was decreased by loss of HAMP responsiveness in renal tubules and increased by loss of FPN in renal tubules. Under iron replete conditions, these effects were confined to female mice and accompanied by transient changes in serum and liver iron levels. Under conditions of excess iron availability, loss of HAMP responsiveness in renal tubules exacerbated liver iron overload and reduced renal iron overload in male and female mice. Additionally, hemochromatosis mice harbouring a ubiquitous knock-in of fpnC326Y did not develop renal iron overload otherwise seen in iron-loaded wild type mice, despite similar degrees of liver iron overload in the two settings. Our findings are the first formal demonstration that HAMP directly controls FPN-dependent iron reabsorption. They show that this renal HAMP/FPN axis determines the degree of renal and hepatic iron overload in the setting of excess iron availability. They also demonstrate that loss of the renal HAMP/FPN axis contributes to liver iron overload while protecting the kidney in the setting of hemochromatosis.
Currently, there is considerable interest in strategies that target the HAMP/FPN axis for the treatment of diseases of iron overload (9). Our findings suggest that these strategies will additionally reduce iron reabsorption and increase renal iron content. Additionally, parenteral iron is increasingly being used to treat iron deficiency in chronic inflammatory conditions where serum HAMP is also raised, e.g. chronic kidney disease (10,11). Our findings suggest that inhibition of iron reabsorption may contribute to iron deficiency in these conditions and that increasing serum iron availability in this setting may cause renal iron overload.

The hepcidin/ferroportin axis in renal tubules controls iron reabsorption
To manipulate the FPN/HAMP axis in renal tubules, we used mice harbouring a Pax8.CreER T2+ knock-in transgene which drives tamoxifen-inducible expression of the Cre recombinase under control of the paired box gene 8 pax8 promoter in proximal and distal tubules and in collecting ducts (12). To confirm further the efficacy and specificity of this Cre transgene, we crossed Pax8.CreER T2+ mice with mice harbouring the Fpn fl/fl allele, and found that the presence of the deletion allele (ΔFpn) was indeed confined to the kidneys of Fpn fl/fl , To confirm the role of the FPN in renal iron homeostasis, we crossed Pax8.CreER T2+ mice with those harbouring a conditional knockout floxed Fpn allele. Furthermore, to understand the role of HAMP in renal iron homeostasis, we crossed Pax8.CreER T2+ mice with those harbouring a conditional knock-in floxed allele FpnC326Y, which encodes a hepcidinresistant FPN. Fpn fl/fl ,Pax8.CreER T2+ and FpnC326Y fl/fl ,Pax8.CreER T2+ mice and their respective Fpn fl/fl and FpnC326Y fl/fl controls were induced with tamoxifen at 4 weeks of age, and their iron status characterised 1 week, 1 month, 3 months and 6 months later.
We found that renal iron content was higher in Fpn fl/fl ,Pax8.CreER T2+ females than in the  Additionally, decreased renal content and increased serum iron levels in FpnC326Y fl/fl ,Pax8.CreER T2+ females demonstrate that FPN-dependent iron reabsorption is subject to regulation by HAMP. These data also indicate that, under normal physiological conditions, the control of iron reabsorption by the renal HAMP/FPN axis is more important in females than in males.

The renal HAMP/FPN axis contributes normal systemic iron homeostasis under ironreplete conditions
Next, we set out to determine the contribution of the renal HAMP/FPN axis to systemic iron homeostasis. We found that

The renal HAMP/FPN axis determines the pattern of tissue iron overload in hemochromatosis
Next, we set out to explore the role of the renal HAMP/FPN axis in the context of hereditary hemochromatosis, a genetic condition of iron overload caused by defects in hepcidin production or hepcidin responsiveness (13). To that effect, we used mice generated inhouse harbouring a heterozygous ubiquitous knock-in of the fpnC326Y allele (Fpn wt/C326Y ).
We had previously demonstrated that these mice develop the iron-overload phenotype characteristic of hereditary hemochromatosis (14,15). We found that while Fpn wt/C326Y mice had higher liver iron content than their Fpn wt/wt controls, their renal iron content was normal Relevant to this, we found that expression of the hamp gene in renal tubules was raised 8 following provision of iron-loaded diet to wild type mice but not in hemochromatosis mice, suggesting that renal HAMP expression may be regulated by intracellular iron levels within tubules (Supplemental Fig2). In the future, it would be interesting to determine the relative contributions of renal and hepatic HAMPs to the control of iron reabsorption.
Another important finding of the present study is that, under normal physiological conditions, the renal HAMP/FPN axis contributes to systemic iron homeostasis but that this contribution is minor compared to that of the HAMP/FPN axes in the duodenum and reticuloendothelial system. Indeed, the observed changes in serum and liver iron indices resulting from loss of FPN or of HAMP-responsiveness in the real tubules were transient, suggesting the involvement compensatory mechanism(s). One possible compensatory mechanism is modulation of hepatic HAMP in response to changes in serum iron availability. Consistent with this notion, hepatic hamp gene expression was decreased in livers of female mice with renal-tubule specific loss of FPN and increased in livers of female mice with renal-tubule specific loss of HAMP-responsiveness (Supplemental Fig3). As well as being transient in nature, the observed changes resulting from loss of FPN or of HAMP-responsiveness in the renal tubules were confined to female mice. This finding could not be attributed to differences between males and females in the activity of the Pax8.CreERT 2+ transgene because the product of fpn knockout (ΔFpn) was detected in kidneys of both male and female fpn fl/fl , Pax8.CreERT 2+ mice (Supplemental Fig1A). Instead, this finding suggests that, at least in C57BL/6 mice, the contribution of renal iron reabsorption to systemic iron levels is more important in females than in males. A previous study using a different Cre recombinase transgene, driven by a constitutively active Nestin promoter to delete fpn in entire nephron also reported an increase in renal iron levels, and decrease in serum iron and liver iron stores, although that study was conducted in a different mouse strain (129/SvEvTac), did not distinguish between males and females and did not report on the timecourse of these changes (8).
The findings of the present study have potentially important clinical implications. First, they suggest that strategies targeting the HAMP/FPN axis for the treatment of iron overload e.g.
HAMP mimetics, FPN inhibitors, may reduce iron reabsorption and increase renal iron content (9). Second, they indicate that inhibition of iron reabsorption by HAMP may contribute to iron deficiency in the setting of chronic conditions, where HAMP is raised by inflammation, e.g. chronic kidney disease. Finally, they suggest that raising systemic iron availability in this setting (e.g. using parenteral iron supplementation) may affect renal function by causing renal iron overload (10,11).

Mice
All animal procedures were compliant with the UK Home Office Animals (Scientific

Procedures) Act 1986 and approved by the University of Oxford Medical Sciences Division
Ethical Review Committee.
The conditional fpn fl , and fpnC326Y fl alleles was generated as described previously (14,18).
Mice harbouring the Pax8.CreERT2+ transgene were a gift from Dr Athena Matakidou, Cancer Research UK Cambridge Institute, University of Cambridge. These mice were generated as described previously (12).

Diets
Unless otherwise stated, animals were provided with a standard rodent chow diet containing 200ppm iron. In iron manipulation experiments, mice were given an iron-loaded diet (5,000ppm iron; Teklad TD.140464) or a matched control diet (200ppm iron; Teklad TD.08713) from weaning for 3 months.

Iron quantitation
Serum iron and ferritin levels were determined using the ABX-Pentra system (Horiba Medical, CA). Determination of total elemental iron in tissues was carried out by inductively coupled plasma mass spectrometry (ICP-MS) as described previously (14,15,18). Calibration was achieved using the process of standard additions, where spikes of 0ng/g, 0,5ng/g, 1ng/g, 10ng/g, 20ng/g and 100ng/g iron were added to replicates of a selected sample. An external iron standard (High Purity Standards ICP-MS-68-A solution) was diluted and measured to confirm the validity of the calibration. Rhodium was also spiked onto each blank, standard and sample as an internal standard at a concentration of 1ng/g. Concentrations from ICP-MS were normalised to starting tissue weight.

DAB-enhanced Perls stain
Formalin-fixed paraffin-embedded tissue sections were deparaffinised using Xylene, then rehydrated in ethanol. Slides were then stained for 1 hour with 1% potassium ferricyanide in 0.1mol/L HCl buffer. Endogenous peroxidase activity was quenched, then slides were stained with DAB chromogen substrate and counterstained with haematoxylin. They were visualised using a standard brightfield microscope.

Quantitative PCR
Gene expression was measured using Applied Biosystems Taqman       B. In mice with intact iron homeostasis, provision of iron-loaded diet increases serum iron levels and iron levels in the glomerular filtrate. However, iron reabsorption is blocked by HAMP which is elevated in this setting. The combination of increased iron in the glomerular filtrate and decreased iron reabsorption leads to renal iron overload. In contract, iron reabsorption from renal tubules is enhanced in hemochromatosis mice (Fpn wt/C326Y ), due to loss of HAMP responsiveness in renal tubules. Enhanced iron reabsorption mitigates against the effects of increased iron levels in the glomerular filtrate, protecting the kidney from iron overload.

Supplemental Figure 1-Confirmation of Pax8.CreER T2+ activity and specificity
Genotyping of Fpn fl/fl and Fpn fl/fl ,Pax8.CreER T2+ mice using genomic DNA extracted from the liver, spleen and kidney at 1 week post tamoxifen induction. Values are shown as mean± standard error of the mean. *p<0.05, **p<0.01.