BMP6 is a key endogenous regulator of hepcidin expression and iron metabolism

Abstract

Juvenile hemochromatosis is an iron-overload disorder caused by mutations in the genes encoding the major iron regulatory hormone hepcidin (HAMP) and hemojuvelin (HFE2). We have previously shown that hemojuvelin is a co-receptor for bone morphogenetic proteins (BMPs) and that BMP signals regulate hepcidin expression and iron metabolism. However, the endogenous BMP regulator(s) of hepcidin in vivo is unknown. Here we show that compared with soluble hemojuvelin (HJV.Fc), the homologous DRAGON.Fc is a more potent inhibitor of BMP2 or BMP4 but a less potent inhibitor of BMP6 in vitro. In vivo, HJV.Fc or a neutralizing antibody to BMP6 inhibits hepcidin expression and increases serum iron, whereas DRAGON.Fc has no effect. Notably, Bmp6-null mice have a phenotype resembling hereditary hemochromatosis, with reduced hepcidin expression and tissue iron overload. Finally, we demonstrate a physical interaction between HJV.Fc and BMP6, and we show that BMP6 increases hepcidin expression and reduces serum iron in mice. These data support a key role for BMP6 as a ligand for hemojuvelin and an endogenous regulator of hepcidin expression and iron metabolism in vivo.

Figure 1. DRAGON.Fc selectively inhibits BMP induction of hepcidin expression in vitro but does not affect hepcidin expression or iron metabolism in vivo

(a–d) Hep3B cells were transfected with a hepcidin promoter luciferase reporter and control Renilla luciferase vector (pRL-TK), and incubated in the absence or presence of BMP ligands, either alone or in combination with 0.2 to 25 μg/mL purified DRAGON.Fc (Dra.Fc) or HJV.Fc as shown. Relative luciferase activity was calculated as the ratio of firefly to Renilla luciferase to control for transfection efficiency. The fold increase over baseline was generally between 40 to 150-fold for all BMP ligands for most experiments, although there was some variability. Results are reported as the mean +/− s.d. of the percent decrease in relative luciferase activity for cells treated with BMP ligands in combination with Dra.Fc or HJV.Fc compared with cells treated with BMP ligands alone, n = 6 to 11 per group. (a) Effects of Dra.Fc on BMP-2, BMP-4, BMP-5, BMP-6, BMP-7, and BMP-9 ligands. * For the indicated cells treated with BMP-2 or BMP-4 in combination with all concentrations of Dra.Fc, P < 0.001 compared with cells treated with BMP-2 or BMP-4 alone, and P ≤0.02 compared with cells stimulated with all other BMPs and treated with identical concentrations of Dra.Fc. (b–d) Head to head comparison of DRAGON.Fc and HJV.Fc for inhibiting BMP-2 (b), BMP-4 (c), and BMP-6 (d). Exact P values are shown. (e–f) Eight week-old male 129S6/SvEvTac mice received an intraperitoneal injection of purified soluble DRAGON.Fc (Dra.Fc) at 5 or10 mg/kg, HJV.Fc at 5 or 7 mg/kg, or an equal volume of isotonic saline (Con) three times weekly for three weeks. Results for both DRAGON.Fc and HJV.Fc doses were similar and were therefore combined into one group. (e) Total liver RNA was isolated and analyzed by quantitative real-time RT-PCR for hepcidin mRNA relative to Rpl19 mRNA as an internal control. (f) Measurement of serum iron. (e–f) Results are reported as the mean +/− s.d, n = 7 per group. Exact P values are shown.

Figure 2. Neutralizing BMP-6 antibody inhibits hepatic hepcidin expression and increases serum iron and transferrin saturation in vivo

(a) Hep3B cells were transfected with a hepcidin promoter luciferase reporter and control pRL-TK as in Figure 1. Forty-eight hours after transfection, cells were incubated in the absence or presence of BMP ligands, either alone or in combination with 0.2 to 25 μg/mL neutralizing BMP-6 antibody as shown (n = 5 per group). Relative luciferase activity was calculated as in Figure 1. (b–d) Eight week-old male 129S6/SvEvTac mice received an intraperitoneal injection of neutralizing BMP-6 antibody at 10 mg/kg (α-BMP6, n = 4) or an equal volume of isotonic saline (Control, n = 4) daily for three days. (b) Total liver RNA was isolated and analyzed by quantitative real-time RT-PCR for hepcidin mRNA relative to Rpl19 mRNA as an internal control. (c and d) Measurement of serum iron (c) and transferrin saturation (d). Results are expressed as the mean +/− s.d. (a) * P = 0.036 for cells treated with 0.2 μg/mL BMP-6 antibody in combination with BMP-6 ligand compared with cells treated with BMP-6 alone. ** P ≤0.01 for all other concentrations of BMP-6 antibody compared with cells treated with BMP-6 alone and compared with cells stimulated with all other BMPs and treated with identical concentrations of BMP-6 antibody. (b–d) Exact P values are shown.

Figure 3. Bmp6 null mice exhibit reduced hepatic hepcidin expression, increased spleen ferroportin expression, increased serum iron and transferrin saturation, increased liver, heart, and pancreas iron content, and reduced spleen iron content

Eight-week-old male Bmp6 null mice (n = 5) and strain matched wildtype control mice (WT, n = 5) were analyzed for (a) hepcidin mRNA expression relative to Rpl19 mRNA expression by quantitative real-time RT-PCR, (b–c) ferroportin expression relative to β-actin expression by Western blot (c) followed by quantitation using IPLab Spectrum software (b), (d) serum iron, (e) serum transferrin saturation, (f) liver iron content, (g) heart, pancreas, and spleen iron content. (H) Perls Prussian blue staining of tissue iron in wildtype (WT) and a Bmp6 null mouse livers. Original magnification x 40. Results are expressed as mean +/− s.d. Exact P values are shown.

Figure 4. BMP-6 interacts with HJV.Fc

BMP-6 alone, HJV.Fc alone, or BMP-6 in combination with HJV.Fc were incubated in solution. (Upper panel) BMP-6 bound to HJV.Fc was precipitated with protein A beads, and the eluted protein complex was analyzed by SDS-PAGE, followed by Western blot with BMP-6 antibody under reducing conditions. The band migrating at ~18 kDa present only in the lane containing BMP-6 in combination with HJV.Fc corresponds to the predicted size of BMP-6 monomer (arrow). (Lower panels) As a control to demonstrate input proteins, solution aliquots prior to Protein-A pull-down were also analyzed by Western blot with anti-Fc antibody and anti-BMP-6 antibody under reducing conditions. Control Westerns revealed the predicted bands migrating at 18 kDa for BMP-6 monomer and 75 and 60 kDa for HJV.Fc. A background band at ~19 kDa was present in the lanes containing HJV.Fc probed with BMP-6 antibody suggesting nonspecific crossreactivity of this antibody with a component of the HJV.Fc solution.

Figure 5. BMP-6 administration in mice increases hepcidin mRNA expression and reduces serum iron

Eight-week-old male 129S6/SvEvTac mice received an intraperitoneal injection of BMP-6 at 250 μg/kg (n = 6) or 1000 μg/kg (n = 7) or an equal volume of vehicle alone (n = 6). Six hours after injection, blood and livers were harvested. (a) Total liver RNA was isolated and analyzed by quantitative real-time RT-PCR for hepcidin mRNA relative to Rpl19 mRNA as an internal control. (b–c) Measurement of serum iron (b) and transferrin saturation (c). Results are reported as the mean +/− s.d. Exact P values are shown where significant.