Modulation of bone morphogenetic protein signaling in vivo regulates systemic iron balance

Abstract

Systemic iron balance is regulated by hepcidin, a peptide hormone secreted by the liver. By decreasing cell surface expression of the iron exporter ferroportin, hepcidin decreases iron absorption from the intestine and iron release from reticuloendothelial stores. Hepcidin excess has been implicated in the pathogenesis of anemia of chronic disease, while hepcidin deficiency has a key role in the pathogenesis of the iron overload disorder hemochromatosis. We have recently shown that hemojuvelin is a coreceptor for bone morphogenetic protein (BMP) signaling and that BMP signaling positively regulates hepcidin expression in liver cells in vitro. Here we show that BMP-2 administration increases hepcidin expression and decreases serum iron levels in vivo. We also show that soluble hemojuvelin (HJV.Fc) selectively inhibits BMP induction of hepcidin expression in vitro and that administration of HJV.Fc decreases hepcidin expression, increases ferroportin expression, mobilizes splenic iron stores, and increases serum iron levels in vivo. These data support a role for modulators of the BMP signaling pathway in treating diseases of iron overload and anemia of chronic disease.

Figures

Figure 1. Induction of hepcidin expression by TGF-β/BMP superfamily ligands.

(A) Hep3B cells were transfected with a hepcidin promoter firefly luciferase reporter and a control pRL-TK. Transfected cells were incubated either alone (control) or with 50 ng/ml BMP or GDF ligands, 5 ng/ml TGF-β ligands, or 30 ng/ml activin A (ActA) as indicated. Cell lysates were analyzed for luciferase activity. To control for transfection efficiency, relative luciferase activity was calculated as the ratio of firefly luciferase values to Renilla luciferase values and is expressed as the fold increase compared with control. Results are reported as the mean ± SD (n = 2–3 per group). (B) Hep3B cells were treated with BMP, GDF, TGF-β, or activin A ligands as in A. Total RNA was analyzed by quantitative real-time RT-PCR for hepcidin mRNA expression and β-actin mRNA expression. Samples were analyzed in triplicate and are reported as the ratio of the mean values of hepcidin to β-actin.

Figure 2. BMP-2 administration in mice increases hepcidin mRNA expression and decreases serum iron.

129S6/SvEvTac mice were injected retroorbitally with 1 mg/kg BMP-2 (n = 8) or an equal volume of vehicle alone (n = 7). Four hours after injection, blood and livers were harvested. (A) Total mRNA was isolated from livers and analyzed by quantitative real-time RT-PCR for hepcidin mRNA expression relative to expression of GAPDH mRNA, which was used as an internal control. (B) Serum iron was measured by colorimetric assay. Results are reported as the mean ± SD. *P = 0.02 for BMP-2–treated mice compared with controls.

Figure 3. Soluble HJV.Fc inhibits basal hepcidin expression and selectively inhibits BMP induction of hepcidin expression.

(A) Western blot of purified soluble HJV.Fc fusion protein with anti-hemojuvelin antibody (α-HJV) and anti-Fc antibody (α-Fc). (B and C) HepG2 cells were incubated alone (control) or with 25 μg/ml HJV.Fc alone, 25 ng/ml BMP-2 alone, or a combination of HJV.Fc and BMP-2 as indicated. Total RNA was isolated and quantitative real-time RT-PCR for hepcidin mRNA relative to β-actin mRNA was performed as in Figure 1. Results are reported as the mean ± SD (n = 3 per group; *P = 0.03 for HJV.Fc compared with control; †P = 0.009 for HJV.Fc plus BMP-2 compared with BMP-2 alone). (D) Hep3B cells were transfected with the hepcidin promoter luciferase construct and pRL-TK. Transfected cells were incubated alone, with 5 ng/ml BMP-9, 50 ng/ml BMP-5, or 25 ng/ml BMP-2, BMP-4, BMP-6, or BMP-7 ligands, or with the BMP ligands plus 0.2 to 25 μg/ml HJV.Fc as indicated, followed by measurement of relative luciferase activity as in Figure 1. Results are reported as the mean ± SD of the percent decrease in relative luciferase activity for cells treated with BMP ligands in combination with HJV.Fc compared with cells treated with BMP ligands alone (n = 2 per group).

Figure 4. siRNA inhibition of endogenous BMP ligands decreases basal hepcidin expression.

(A) Expression of endogenous BMP ligands in HepG2 cells as measured by RT-PCR. Purified plasmid cDNAs expressing BMP ligands were used as positive controls. (B and C) HepG2 cells were transfected with BMP ligand siRNAs or a control scrambled siRNA as indicated. Total RNA was analyzed for BMP ligand expression (B) or hepcidin expression (C) relative to β-actin expression by real-time quantitative RT-PCR. Results are reported as the mean ± SD of the percent decrease in the ratio of hepcidin or BMP ligand to β-actin for cells treated with various BMP siRNAs compared with cells treated with control siRNA; n = 3–6 per group; *P < 0.05.

Figure 5. Soluble HJV.Fc administration in mice decreases hepatic phosphorylated Smad1, Smad5, and Smad8 expression, decreases hepcidin expression, increases serum iron, increases liver iron content, and decreases spleen iron content.

129S6/SvEvTac mice received an i.p. injection of 25 mg/kg HJV.Fc or normal saline (control) 3 times weekly for 3 weeks. (A) Liver lysates were analyzed for phosphorylated Smad1, Smad5, and Smad8 (α–p-Smad1/5/8) expression by western blot. Blots were stripped and reprobed for expression of total Smad1 and β-actin, which were used as loading controls. Chemiluminescence was quantitated by IPLab Spectrum software for phosphorylated Smad1, Smad5, and Smad8 expression relative to total Smad1 expression. (B) Total mRNA was isolated from livers and analyzed by quantitative real-time PCR for hepcidin mRNA expression relative to GAPDH mRNA expression as an internal control. (C) Spleen membrane preparations were analyzed for ferroportin expression by western blot. Blots were stripped and reprobed for expression of β-actin, which was used as a loading control. (D and E) Measurement of serum iron (D) and transferrin saturation (Serum Tf sat; E). (F and G) Quantitation of liver (F) and spleen (G) tissue iron content. Results are expressed as mean ± SD, n = 3 mice per group; *P = 0.0497, †P = 0.003, ‡P = 0.01, ΧP = 0.004, ζP = 0.03, #P = 0.009 for HJV.Fc-treated mice compared with controls.

Figure 6. Soluble HJV.Fc inhibits IL-6 induction of hepcidin expression.

HepG2 cells were incubated for 16 hours alone (control), with 100 ng/ml IL-6, or with 100 ng/ml IL-6 in combination with HJV.Fc after pre-incubation with HJV.Fc for 1 hour. Total RNA was analyzed for hepcidin expression relative to β-actin expression by quantitative real-time RT-PCR. Results are expressed as mean ± SD, n = 3 per group; *P = 0.003 for IL-6–treated cells compared with control cells; †P = 0.0006 for cells treated with HJV.Fc in combination with IL-6 compared with cells treated with IL-6 alone.