Inactivation of Tmprss6 Hampers Hepcidin Inhibition By Erythroferrone through Upregulation of the BMP-SMAD Pathway

Introduction

Erythroferrone (ERFE), a member of the C1q-TNF protein family produced by erythropoietin (EPO)-stimulated erythroid cells, has been recently identified as the erythroid regulator that suppresses hepcidin (Kautz et al., 2014). In the liver, the main hepcidin inhibitor is TMPRSS6, the only known in vivo inhibitor of the BMP-SMAD pathway since its inactivation causes Iron Refractory Iron Deficiency Anemia (IRIDA) (Finberg et al., 2008), a genetic disorder due to inappropriately high hepcidin. We have previously demonstrated that inactivation of Tmprss6 in thalassemia mice rescues iron overload and ameliorate anemia by increasing hepcidin, notwithstanding high Epo (Nai et al., 2012) and Erfe(Silvestri et al., ASH Meeting 2014) thus suggesting that Tmprss6 is indispensable for hepcidin inhibition by the erythroid regulator(s).

Methods

To analyze BM and spleen erythroid differentiation, Ter119+ cells were separated according to Forward Scatter (FSC) value and Cd44 expression, as previously described (Liu et al., 2013). Expression of Erfe, Epo, hepcidin and BMP-SMAD target genes were evaluated by qRT-PCR and iron parameters by standard methods (Pagani et al., 2011). We have analyzed different animal models after a single (200U) EPO injection: 1) Tmprss6 KO mice and wild type littermates fed an iron balanced (IB) and iron deficient (ID), diet; 2) wild type mice with activated BMP-SMAD pathway by iron dextran (1g/kg) treatment; 3) Tmprss6KO animals with the BMP-SMAD pathway inhibited by 12.5 μg/g dorsomorphin. Both males and females mice were analyzed at 8-10 weeks of age and sacrificed 15 hours after EPO treatment.

Results

EPO administration increases bone marrow (BM) erythroid precursors in all mice models. In the spleen erythroid precursors are not further upregulated by EPO in ID and Tmprss6 KO mice, since already high in basal conditions likely due to stress erythropoiesis. Erfe expression is upregulated in basal conditions in BM and spleen of ID and Tmprss6 KO mice, which have high serum Epo levels, and is further increased by EPO treatment in all models. As expected, hepcidin is strongly reduced in ID animals, whereas it is high in Tmprss6 KO mice. EPO injection strongly inhibits hepcidin both in IB and ID mice but not in Tmprss6 KO animals, notwithstanding the upregulation of Erfe. This suggests that lack of Tmprss6 impairs the Erfe-mediated hepcidin inhibition. To evaluate the contribution of Tmprss6 and the BMP signaling to Erfe function, we modulated the BMP pathway before EPO challenge. We activated the signaling by means of iron dextran injection in wild type mice and we inhibited the hyperactive pathway of Tmprss6 KO animals by using the BMP receptor inhibitor dorsomorphin. Neither iron dextran injection nor dorsomorphin treatment impairs Erfe activation by EPO that equals that observed in saline treated animals. In iron loaded mice with hyperactive BMP signaling, hepcidin was not inhibited by Erfe. Dorsomorphin treatment in Tmprss6KO animals rescues the hepcidin responsiveness to Erfe.

Conclusions

We confirm that the BMP-SMAD pathway is dispensable for Erfe-mediated hepcidin inhibition since Erfe decreases hepcidin in ID, when the pathway is suppressed. However, activation of the BMP-SMAD pathway consequent to iron loading or to the loss of Tmprss6 hampers the hepcidin downregulation by Erfe, suggesting that the Tmprss6 function, and not Tmprss6 per se, is essential to facilitate the Erfe-dependent hepcidininhibition by blunting the BMP-SMAD signaling.