Matriptase-2 links erythropoietin to iron

In this issue of Blood,Nai et al demonstrate that the serine protease matriptase-2 mediates erythropoietin (EPO)-dependent suppression of hepcidin, a central regulator of iron homeostasis with causal or contributing roles in iron-loading diseases, hemoglobinopathies, and infectious, inflammatory, and neoplastic disorders.¹ 

Hepcidin is a peptide hormone synthesized mainly by hepatocytes.²  It regulates iron homeostasis by posttranslationally downregulating levels of the cellular iron export protein ferroportin. Because ferroportin is expressed in duodenal enterocytes and reticuloendothelial macrophages, hepcidin limits dietary iron absorption and macrophage iron export. Hepcidin itself is regulated by a variety of conditions including iron excess and deficiency, and erythropoiesis. In states of anemia and/or iron deficiency, when the demand for iron by erythroid precursors and other cell types is high, hepcidin expression is limited, assuring continued dietary iron absorption and release of macrophage iron stores. Under conditions of iron excess, hepcidin is abundantly expressed, limiting dietary iron absorption and sequestering iron stores in macrophages. This regulation ensures adequate but not excessive iron levels in line with the needs of the body. Aberrant regulation of hepcidin expression causes or contributes to pathophysiology of a wide variety of diseases, including inherited iron loading diseases, hemoglobinopathies, infectious diseases, and inflammatory and neoplastic disorders.

The means by which erythropoiesis impacts iron metabolism has perplexed researchers for decades.³  The discovery of hepcidin as a central regulator of iron metabolism and the demonstration that EPO inhibits hepcidin expression permits a refocusing of this question at a molecular level: How does EPO regulate hepcidin expression? The study by Nai et al provides one piece of, if not the, answer to this question by exploiting a novel condition, the deficiency of a hepatic membrane-bound serine protease known as matriptase-2, also referred to as Tmprss6. Matriptase-2 deficiency causes iron-refractory iron deficiency anemia, a condition of hepcidin excess characterized by anemia and iron deficiency.⁴  Matriptase-2 deficiency causes hepcidin excess because the function of matriptase-2 is to cleave hemojuvelin (HJV), a bone morphogenetic protein (BMP) co-receptor essential for hepatic hepcidin expression in conditions of iron excess. As such, matriptase-2 serves to inhibit expression, whereas deficiency of this protease leads to unregulated BMP signaling, hepcidin excess, and iron-deficiency anemia. The anemia is refractory to parenteral iron because hepcidin excess impairs dietary iron absorption.

One of the intriguing aspects of matriptase-2 deficiency is that hepcidin excess persists even in the context of excess EPO, a known inhibitor of hepcidin expression. The study at hand explores this phenomenon in mechanistic detail. Nai et al first confirm that EPO administration does not suppress hepcidin expression in matriptase-2–deficient mice as it does in mice rendered iron-deficient by dietary means. The authors then take a series of pharmacologic and genetic approaches that interrogate different steps in the BMP-dependent hepcidin expression pathway in mice (see figure). Two of the most compelling experiments in this data-rich paper are described here. First, to tackle the possibility that hepcidin expression is too high to be suppressed by EPO in matriptase-2 deficiency, they induce excess BMP signaling without inactivating matriptase-2 by injecting wild-type mice with iron dextran. Hepcidin levels were unresponsive to EPO administration in these mice, suggesting that excess BMP signaling can block suppression of hepcidin by EPO. Second, the authors examine mice deficient in both matriptase-2 and Bmp6, the BMP most relevant to hepcidin regulation and iron metabolism. These mice act as a model of matriptase-2 deficiency without excess BMP signaling and intriguingly do not repress hepcidin expression after EPO administration. This solidly demonstrates that matriptase-2 is essential to EPO-mediated hepcidin suppression.

The authors also consider the role of erythroferrone (ERFE) in hepcidin suppression by erythropoiesis. ERFE was recently identified as an erythroid regulator,⁵  defined here as a soluble factor secreted by erythroid precursors that inhibits hepcidin expression, and is abundantly expressed in bone marrow of matriptase-2–deficient mice. Although Nai et al do not directly manipulate ERFE in their mouse models, their analysis of ERFE levels indicates that matriptase-2 is not essential for physiologic regulation of ERFE expression but that matriptase-2 deficiency does impair regulation of hepcidin expression by ERFE. The precise mechanism by which ERFE modulates hepcidin and iron metabolism, and the nature of the functional relationship between matriptase-2 and ERFE, has yet to be established, but will undoubtedly be revealed in the near future by elegant, mechanistic studies similar to the work at hand by Nai et al.