GATA/Heme Multi-Omics Reveals a Trace Metal-Dependent Erythrocyte Developmental Mechanism

By functioning as an enzyme cofactor, hemoglobin component and gene regulator, heme is vital for life. One mode of heme-regulated transcription involves amplifying the activity of GATA-1, a key determinant of erythrocyte differentiation. To discover biological consequences of the metal cofactor-transcription factor mechanism, we used CRISPR/Cas9 to delete intron 1 and 8 GATA motifs in Alas2, encoding the rate-limiting enzyme in heme biosynthesis, in erythroid precursor G1E-ER-GATA-1 cells (Tanimura et al., 2016, EMBO Rep.). These cells stably express a b-estradiol-activated allele encoding the estrogen receptor hormone binding domain fused to GATA-1. The GATA motif mutations abrogated GATA-1-mediated activation of Alas2 and decreased heme ~30 fold. We merged GATA-1/heme-regulated sectors of the proteome and transcriptome. This analysis revealed a GATA-1/heme circuit involving hemoglobin subunits, ubiquitination components, and proteins not implicated in erythrocyte biology, including the zinc exporter Slc30a1.

Slc30a1 and the zinc importer Slc39a8 were the most highly expressed zinc transporters in G1E-ER-GATA-1 cells. Both were GATA-1-activated, and Slc30a1, but not Slc39a8, expression declined in heme-deficient Alas2-intronic mutant cells. We tested whether the zinc transporter genes Slc30a1 and Slc39a8 that share GATA-1 regulation and differ in heme regulation are controlled similarly in primary cells. Lineage-depleted hematopoietic precursors from murine fetal livers were cultured for 72 hours, and flow cytometry was used to isolate cells based on erythroid surface markers. Whereas Slc30a1 expression increased 5.7-fold upon differentiation, Slc39a8 increased 4.6-fold during the initial phase, but was downregulated 8.7-fold thereafter. To quantify intracellular zinc levels, we used the fluorescent zinc indicator FluoZin-3. The FluoZin-3 signal increased during early maturation and decreased thereafter, as predicted by the zinc transporter switch, in which expression of the zinc exporter and importer was sustained and decreased, respectively, during terminal differentiation. We demonstrated that the zinc transporter switch occurs in a primary human erythroblast culture system. Intracellular zinc rises during initial erythroid maturation, followed by a steep decline during terminal differentiation. The heme biosynthesis inhibitor succinylacetone reduced SLC30A1 expression 2.8-fold (p < 0.01), as well as other heme-dependent globin genes HBB and HBA1. These primary mouse and human cell studies revealed an evolutionarily conserved mechanism to differentially regulate zinc transporter genes during erythroid differentiation, culminating in a switch from importer and exporter expression to solely exporter expression.

To test whether reconfiguring the mechanism governing intracellular zinc impacts differentiation and to elucidate Slc30a1 and Slc39a8 function, we developed multiple shRNAs and used them in loss-of-function studies with lineage-negative hematopoietic precursors from murine fetal livers (E14.5). Slc39a8 downregulation decreased intracellular zinc levels in immature erythroblasts. We also used the zinc chelator TPEN to reduce zinc in immature erythroblasts. TPEN reduced intracellular zinc and considerably decreased the live cell population (p < 0.001). Downregulating Slc30a1 increased intracellular zinc 2.2-fold and, strikingly, accelerated differentiation (p < 0.001). This analysis established a conserved paradigm in which a GATA-1/heme circuit controls trace metal transport machinery and trace metal levels as a mechanism governing cellular differentiation. Ongoing studies involve the development of innovative strategies to leverage this mechanism to enhance or suppress erythrocyte development with primary mouse and human systems. We are conducting multi-disciplinary metallomic analyses to test the hypothesis that the dynamic changes in zinc selectively commission and decommission zinc-binding proteins in erythroblasts. We anticipate that this approach can be readily extended to other sectors of the hematopoietic system and more broadly.