Three-dimensional organization of a gene locus is important for its regulation, as recently demonstrated for the ␤-globin locus. When actively expressed, the cis-regulatory elements of the ␤-globin locus are in proximity in the nuclear space, forming a compartment termed the Active Chromatin Hub (ACH). However, it is unknown which proteins are involved in ACH formation. Here, we show that EKLF, an erythroid transcription factor required for adult ␤-globin gene transcription, is also required for ACH formation. We conclude that transcription factors can play an essential role in the three-dimensional organization of gene loci. The mouse ␤-globin locus contains multiple ␤-like globin genes, arranged from 5Ј to 3Ј in order of their developmental expression (Fig. 1A). The adult-type ␤ maj -gene is transcribed at a very low level during primitive erythropoiesis in the embryonic yolk sac, but becomes expressed at high levels around day 11 of gestation (E11) when definitive erythropoiesis commences in the fetal liver (Trimborn et al. 1999). The ␤-globin locus control region (LCR) is essential for efficient globin transcription (Grosveld et al. 1987;Bender et al. 2000). It consists of a series of DNaseI hypersensitive sites (HS) located ∼50 kb upstream of the ␤ maj promoter (Fig. 1A). We have shown that the ␤-globin locus forms an Active Chromatin Hub (ACH) in erythroid cells (Tolhuis et al. 2002). The ACH is a nuclear compartment dedicated to RNA polymerase II transcription, formed by the cis-regulatory elements of the ␤-globin locus with the intervening DNA looping out. The ACH consists of the HS of the LCR, two HS located ∼60 kb upstream of the embryonic y-globin gene (5ЈHS-62/-60) and 3ЈHS1 downstream of the genes. In addition, the actively expressed globin genes are part of the ACH (Carter et al. 2002;Tolhuis et al. 2002). In erythroid precursors that do not express the globin genes yet, a substructure of the ACH, called a chromatin hub (CH) (Patrinos et al. 2004) is found, which excludes the genes and the HS at the 3Ј site of the LCR (Palstra et al. 2003).Expression of the ␤ maj -gene requires the presence of the erythroid Krüppel-like transcription factor EKLF, the erythroid-specific member of the Sp/XKLF-family (Miller and Bieker 1993). EKLF −/− mice die of anemia around E14, because of a deficit in ␤-globin expression (Nuez et al. 1995;Perkins et al. 1995). The ␤-globin locus contains a number of EKLF-binding sites, in particular in the LCR and the ␤ maj -globin promoter (Perkins 1999;Bieker 2001). Because ␤ maj -globin expression depends on the presence of EKLF, we were interested in determining whether EKLF is involved in the formation of the ACH. Results and DiscussionWe used chromatin conformation capture (3C) technology (Dekker et al. 2002) to investigate the three-dimensional conformation of the mouse ␤-globin locus in the absence of EKLF. Cells from E12.5 EKLF −/− and wild-type fetal livers were cross-linked with formaldehyde, followed by restriction enzyme digestion of the DNA. The samples were ligated ...

Development of red blood cells requires the correct regulation of cellular processes including changes in cell morphology, globin expression and heme synthesis. Transcription factors such as erythroid Krüppel-like factor EKLF (Klf1) play a critical role in erythropoiesis. Mice lacking EKLF die around embryonic day 14 because of defective definitive erythropoiesis, partly caused by a deficit in ␤-globin expression. To identify additional target genes, we analyzed the phenotype and gene expression profiles of wild-type and EKLF null primary erythroid progenitors that were differentiated synchronously in vitro. We show that EKLF is dispensable for expansion of erythroid progenitors, but required for the last steps of erythroid differentiation. We identify EKLF-dependent genes involved in hemoglobin metabolism and membrane stability. Strikingly, expression of these genes is also EKLF-dependent in primitive, yolk sac-derived, blood cells. Consistent with lack of upregulation of these genes we find previously undetected morphological abnormalities in EKLF-null primitive cells. Our data provide an explanation for the hitherto unexplained severity of the EKLF null phenotype in erythropoiesis.

According to current models for hematopoiesis, lymphoid-primed multi-potent progenitors (LMPPs; Lin−Sca-1+c-Kit+CD34+Flt3hi) and common myeloid progenitors (CMPs; Lin−Sca-1+c-Kit+CD34+CD41hi) establish an early branch point for separate lineage commitment pathways from hematopoietic stem cells, with the notable exception that both pathways are proposed to generate all myeloid innate immune cell types through the same myeloid-restricted pre-granulocyte-macrophage progenitor (pre-GM; Lin−Sca-1−c-Kit+CD41−FcγRII/III−CD150−CD105−). By single cell transcriptome profiling of pre-GMs we identify distinct myeloid differentiation pathways: a Gata1-expressing pathway generates mast cells, eosinophils, megakaryocytes and erythroid cells, and a Gata1-negative pathway that generates monocytes, neutrophils and lymphocytes. These results identify an early hematopoietic lineage bifurcation, separating the myeloid lineages prior to their segregation from other hematopoietic lineage potentials.