Unraveling the Erythroid Function of Tfr2 in Beta-Thalassemia

Abstract

Transferrin receptor 2 (TFR2), a liver activator of the iron hormone hepcidin mutated in hemochromatosis type 3 (Camaschella C. et al., Nat Genet 2000), is an erythropoietin (EPO) receptor partner in erythroid cells (Forejtnikovà H. et al., Blood 2010). We have recently shown that the loss of bone marrow (BM) Tfr2 increases erythroblast EPO sensitivity and mimics a condition of mild iron-deficiency in mice (Nai A. et al., Blood 2015). β-thalassemias are iron-loading anemias due to recessive mutations in the β-globin gene, characterized by severe ineffective erythropoiesis. Since several studies demonstrated that iron restriction ameliorates the β-thalassemia phenotype, we generated thalassemic mice (Hbbth3/+) with selective BM inactivation of Tfr2 (Tfr2BMKO/Hbbth3/+) through transplantation of BM cells (BMT) from Hbbth3/+ and Tfr2-/-/Hbbth3/+ mice into lethally irradiated wt recipients. Two months after BMT Tfr2BMKO/Hbbth3/+ mice have Hb levels higher than Hbbth3/+ (Hbbth3/+: 9.47±0.59 g/dL; Tfr2BMKO/Hbbth3/+: 11.22±0.47 g/dL; p<0.0001). In contrast, 4 months after BMT Tfr2BMKO/Hbbth3/+ mice have more severe anemia, ineffective and extramedullary erythropoiesis, increased iron in serum (TS) and liver (LIC), reduced spleen iron (SIC), high hepcidin levels, increased Epo-target gene (Epor, Bcl-xL, Erfe) expression in BM and high levels of serum Erfe. These results suggest that deleting Tfr2 in thalassemic erythroblasts increases their Epo sensitivity as occurs in normal ones. This effect is positive during BM repopulation, but transient: on a long-term a severe phenotype develops, similar to the one observed in thalassemia major models (Rivella S. et al., Blood 2003; Huo Y. et al., Blood 2009), likely induced by excessive EPO stimulation of ineffective immature erythropoiesis. We speculate thatin Tfr2BMKO/Hbbth3/+ mice the continuous production of Erfe, the hepcidin inhibitor released by the expanded marrow (Kautz L. et al., Nat Genet 2014), worsens iron-overload, further damaging the thalassemic erythropoiesis. In order to exclude any interference of BMT, we analyzed also Hbbth3/+, Tfr2+/-/Hbbth3/+ and Tfr2-/-/Hbbth3/+ mice, lacking germ-line Tfr2 at the heterozygous and homozygous state respectively. Preliminary data indicate that young Tfr2-/-/Hbbth3/+ mice have less severe anemia than Hbbth3/+, while Tfr2+/-/Hbbth3/+ animals have an intermediate phenotype, suggesting that Tfr2 haploinsufficiency may be sufficient to improve the ineffective erythropoiesis at short time point. Follow up analysis will test whether the phenotype of Tfr2-/-/Hbbth3/+ mice worsens with aging as occurs in transplanted animals. Interestingly we noticed that 2-month-old Tfr2-/-/Hbbth3/+ mice, beside higher Hb levels, have higher LIC than thalassemic littermates, with comparable SIC. This observation lead us to hypothesize that low SIC, rather than high LIC, contributes to the worsening of the phenotype in Tfr2BMKO/Hbbth3/+ mice 4 months after BMT, likely impairing the compensatory spleen erythropoiesis. To dissect causality between high LIC and low SIC we are challenging transplanted mice with iron-deficient diet to decrease LIC or with iron-dextran injection to increase SIC. Moreover, a group of animals sacrificed 2 months after BMT will help to elucidate the mechanism(s) of the early amelioration of anemia. Preliminary results confirm the Hb and red cells count increase in Tfr2BMKO/Hbbth3/+ as compared to Hbbth3/+ mice 2 months after BMT without increased spleen size, thus suggesting a real erythropoiesis improvement at this time point. Erythroid cells at the different stages of maturation sorted from these animals will be subjected to RNAseq analysis in comparison to cells isolated from animals at the time of phenotype worsening. The results are expected to define the pathway/s altered by the loss of Tfr2 in thalassemic erythroid cells and those responsible for the amelioration of the phenotype with the aim of identifying potential novel target/s for therapy of β-thalassemia. DisclosuresNo relevant conflicts of interest to declare.