Scientific Report 2007

Molecular and Experimental Medicine

Division of Hematology and Genetics

Regulation of Hepcidin

E. Beutler, K. Crain, J. Flanagan, T. Gelbart, P. Lee, H. Peng, J. Truksa, J. Waalen

It has been known for many years that body iron content is normally tightly regulated. Iron deficiency increases iron absorption, and iron loading decreases iron absorption. However, the basis of this "mucosal intelligence" remained entirely obscure. Hereditary hemochromatosis is a disorder in which the regulatory system does not function normally, and large amounts of iron accumulate, causing tissue damage. The discovery of the role of hepcidin, a 25 amino acid peptide, has greatly furthered our understanding of the regulation of iron homeostasis, and we now realize that it is dysregulation of hepcidin transcription that is the cause of most forms of hemochromatosis. But how is this transcription regulated? The short answer is that we do not know.

In intact animals, and to the extent that studies have been performed in humans, hepcidin is upregulated by iron and by inflammatory cytokines induced by endotoxin and is downregulated by anemia and by hypoxia. Although in primary hepatocytes and in many cells lines hepcidin transcription is regulated in a similar manner by the inflammatory cytokines IL-6, IL-1α, and IL-1β and by hypoxia as in intact animals, the effect of iron on hepcidin transcription in cell cultures or primary hepatocytes is the opposite of that in intact animals. This paradoxical effect is unexplained. Apparently an iron-sensing and signaling mechanism exists in vivo that does not operate in the in vitro systems. In an effort to identify the components of a signaling mechanism, we have performed many studies in which we have coincubated primary murine hepatocytes with a variety of tissues, including macrophages, marrow stromal cells, and marrow hematopoietic cells, without being able to recapitulate the course of events observed when iron is administered in vivo.

Because transcription of hepcidin in HepG2 cells responds to IL-6 and to bone morphogenetic proteins (BMPs), it is possible to identify sequences in the promoter that are required for these responses. We have performed such studies by using the conventional approach of attaching promoters of various lengths to a firefly luciferase reporter and comparing the light output with a cotransfected renilla luciferase control. It had been shown by others that the proximal part of the promoter contains elements important for the response of hepcidin transcription to IL-6, but apparently further upstream sequences increase the IL-6 response, and the response to BMP depends strongly on far-upstream elements that we are currently defining (Fig. 1). We were also able to demonstrate that no downstream elements were located in introns or 3′ or 5′ untranslated regions that affected hepcidin expression.

Fig. 1. Induction of a luciferase reporter with mHepc1 promoter fragments of different lengths transfected into HepG2 cells. Even short promoter fragments support the induction of luciferase expression when IL-6 is used as a stimulus (right panel), but only fragments longer than 1.5 kb are effective when BMP-9 is the stimulus.

Nuclear extracts have been prepared from mice fed high- and low-iron diets. Attention has been focused on certain candidate areas that are of interest because of a high level of homology between human and murine hepcidin genes and because of results obtained with footprinting.

To narrow the regions that are critical for iron responsiveness in the hepcidin promoter, we have made probes that corresponded to conserved regions between human and murine hepcidin genes. By gel-shift analyses, we have been able to determine which probes are differentially bound by proteins in iron-deficient and high-iron nuclear extracts. For example, we found differential binding of nuclear factors to probes hybridizing with 2 regions: –1797 to –1754 and –1680 to –1641 (Fig. 2).

Fig. 2 Gel shift analysis of a probe composed of nucleotides –1797 to –1754 with nuclear extracts from a pool of 4–8 mice fed an iron-deficient diet (D) or 4–8 mice fed a high-iron diet (H) for more than 4 weeks. Mice from 4 strains, as indicated at the bottom of each gel, were investigated.

To identify provisionally iron-responsive transcription factors that regulate hepcidin transcription, we determined which liver-specific transcription factors are differentially expressed between iron-deficient and iron-loaded mice (>4 weeks). We used a protein/DNA array method to identify both transcriptional activators and repressors (Fig. 3). This method also permits identification of coactivators or corepressors that may not directly bind to the DNA but may associate with the DNA-binding proteins. Several exposures were obtained in order to identify high- and low-abundance transcription factors within the factors' dynamic range. In this particular experiment, we identified 43 and 39 transcription factors that were upregulated in iron-overload and iron-deficient nuclear extracts, respectively. Of these, 10 and 4 transcription factors from iron-overload and iron-deficient nuclear extracts, respectively, have corresponding consensus motifs in the hepcidin promoter.

Fig. 3. Protein/DNA array of transcription factors. Nuclear extracts (pool of 8 mice) from 129 strain mice fed an iron-deficient (A) or a high-iron (B) diet for 4 weeks were used to select specific biotinylated probes that were then used to bind to the array membrane. Red squares indicate examples of transcription factors that are upregulated in iron-deficient nuclear extracts; black squares, examples of transcription factors upregulated in iron overload. The chemiluminescent exposure was 10 minutes.

Because hepcidin transcription is not stimulated by iron in ex vivo systems, we have devised a method for measuring the response of the hepcidin promoter in vivo. We hydrodynamically transfect mice with constructs containing fragments of the hepcidin promoter fused to a firefly luciferase (luc) reporter. To elicit a response to iron, we maintained mice for at least 2 weeks on an iron-poor diet containing only 2–5 ppm of iron and then fed them a diet containing 2 x 10⁴ ppm iron for 24 hours. Luciferase expression driven by the different hepcidin promoter regions was measured in the intact animals by using a live imaging instrument.

The results of our studies are shown in Figures 4 and 5. It is apparent that a region of the promoter between 1.6 and 1.8 kb upstream from the start of translation is required for the response to iron. This region, together with the first 260 bp of the promoter, is sufficient to provide a near-maximal response to iron stimulation. Interestingly, this region is the same one required for the in vitro response of HepG2 cells to stimulation with BMP-4 or BMP-9 (Fig. 2).

Fig. 4. Location of the iron-responsive element in vivo. Mice fed an iron-poor diet (2–5 ppm) were hydrodynamically transfected with a pGL3 reporter plasmid containing the firefly luciferase gene (luc) under the control of various lengths of the murine hepcidin 1 (mHepc1) promoter. After 3 days, the basal level of bioluminescence was determined, and mice were divided into 2 groups. The first group received a high-iron diet (2 x 104 ppm); the second group remained on the iron-poor diet. After 24 hours, bioluminescence was measured; the results were expressed as a percentage of the basal level bioluminescence. The number of base pairs upstream of the start of translation is given for each promoter construct. The construct designated 260 bp + (1.6 to 1.8 kb) contains the first 260 bp and the part of the promoter between 1.6 and 1.8 kb after the start of translation. Each group consisted of at least 5 animals. The error bars represent 1 SE of the mean.

Fig. 5. Bioluminescence in mice hydrodynamically transfected with a hepcidin promoter fused to firefly luciferase. LPS = lipopolysaccharide.

Our database of 41,000 subjects participating in the hemochromatosis screening study with Kaiser-Permanente has facilitated epidemiologic investigation of fundamental questions in clinical hematology, including one even as basic as what is the definition of anemia. For most epidemiologic studies, the answer has been assumed to be simple: on the basis of recommendations from the World Health Organization, anemia is defined as hemoglobin levels less than 13 g/dL for men and less than 12 g/dL for women. However, analysis of our data has shown that the distribution of hemoglobin values varies significantly, not only between men and women but also between subjects of different ethnic origins; African Americans have hemoglobin values that are, on average, 0.6 g/dL lower than those of whites. In addition, hemoglobin values decline significantly with age among men, but not among women, and in all ethnic groups. We have validated these findings by using data from the National Health and Nutrition Examination Survey, a population-based, representative sample of the U.S. population. The work has established new sex-, age-, and ethnicity-specific reference values for the definition of anemia. The studies have provided the basis for further investigation into the biological mechanisms underlying ethnic differences in hemoglobin levels and the anemia of aging.

PUBLICATIONS

Bartfai, T., Waalen, J., Buxbaum, J.N. Adipose tissue as a modulator of clinical inflammation: does obesity reduce the prevalence of rheumatoid arthritis? J. Rheumatol. 34:488, 2007.

Barton, J.C., Lee, P.L., West, C., Bottomley, S.S. Iron overload and prolonged ingestion of iron supplements: clinical features and mutation analysis of hemochromatosis-associated genes in four cases. Am. J. Hematol. 81:760, 2006.

Beutler, E. Hematopoietic cell transplantation in the future. In: Thomas' Hematopoietic Cell Transplantation, 4th ed. Forman, S.J., Negrin, R.S., Blume, K. (Eds.). Blackwell Science, Boston, in press.

Beutler, E. Hemochromatosis. In: Encyclopedic Reference of Genomics and Proteomics in Molecular Medicine. Ganten, D., Ruckpaul, K. (Eds.). Springer, New York, 2006, p. 758.

Beutler, E. Hepcidin mimetics from microorganisms? A possible explanation for the effect of Helicobacter pylori on iron homeostasis. Blood Cells Mol. Dis. 38:54, 2007.

Beutler, E. PGK deficiency. Br. J. Haematol. 136:3, 2007.

Beutler, E., Gelbart, T. Helicobacter pylori infection and HFE hemochromatosis. Blood Cells Mol. Dis. 37:188, 2006.

Flanagan, J.M., Beutler, E. The genetic basis of human erythrocyte pyridoxal kinase activity variation. Haematologica 91:801, 2006.

Flanagan, J.M., Gerber, A.L., Cadet, J.L., Beutler, E., Sipe, J.C. The fatty acid amide hydrolase 385 A/A (P129T) variant: haplotype analysis of an ancient missense mutation and validation of risk for drug addiction. Hum. Genet. 120:581, 2006.

Flanagan, J.M., Peng, H., Beutler, E. Effects of alcohol consumption on iron metabolism in mice with hemochromatosis mutations. Alcohol. Clin. Exp. Res. 31:138, 2007.

Flanagan, J.M., Peng, H., Wang, L., Gelbart, T., Lee, P., Sasu, B.J., Beutler, E. Soluble transferrin receptor-1 levels in mice do not affect iron absorption. Acta Haematol. 116:249, 2006.

Flanagan, J.M., Rhodes, M., Wilson, M., Beutler, E. The identification of a recurrent phosphoglycerate kinase mutation associated with chronic haemolytic anaemia and neurological dysfunction in a family from USA. Br. J. Haematol. 134:233, 2006.

Flanagan, J.M., Truksa, J., Peng, H., Lee, P., Beutler, E. In vivo imaging of hepcidin promoter stimulation by iron and inflammation. Blood Cells Mol. Dis. 38:253, 2007.

Lichtman, M.A., Beutler, E., Kaushansky, K., Kipps, T.J., Seligsohn, U., Prchal, J. (Eds.). Williams Hematology, 7th ed. McGraw-Hill, New York, 2006.

Rana, B.K., Insel, P.A., Payne, S.H., Abel, K., Beutler, E., Ziegler, M.G., Schork, N.J., O'Connor, D.T. Population-based sample reveals gene-gender interactions in blood pressure in white Americans. Hypertension 49:96, 2007.

Repiso, A., Oliva, B., Vives-Corrons, J.-L., Beutler, E., Carreras, J., Climent, F. Red cell glucose phosphate isomerase (GPI): a molecular study of three novel mutations associated with hereditary nonspherocytic hemolytic anemia. Hum. Mutat. 27:1159, 2006.

Truksa, J., Peng, H., Gelbart, T., Lee, P., Beutler, E. Bone morphogenetic proteins 2, 4, and 9 stimulate murine hepcidin 1 expression independently of Hfe, transferrin receptor 2 (Tfr2), and IL-6. Proc. Natl. Acad. Sci. U. S. A. 103:10289, 2006.

Zimran, A., Elstein, D., Beutler, E. Low-dose therapy trumps high-dose therapy again in the treatment of Gaucher disease. Blood 108:802, 2006.