We have previously described a SWI/SNF-related protein complex (PYR complex) that is restricted to definitive (adult-type) hematopoietic cells and that specifically binds DNA sequences containing long stretches of pyrimidines. Deletion of an intergenic DNA-binding site for this complex from a human ␤-globin locus construct results in delayed human ␥-to ␤-globin switching in transgenic mice, suggesting that the PYR complex acts to facilitate the switch. We now show that PYR complex DNA-binding activity also copurifies with subunits of a second type of chromatin-remodeling complex, nucleosome-remodeling deacetylase (NuRD), that has been shown to have both nucleosome-remodeling and histone deacetylase activities. Gel supershift assays using antibodies to the ATPase-helicase subunit of the NuRD complex, Mi-2 (CHD4), confirm that Mi-2 is a component of the PYR complex. In addition, we show that the hematopoietic cell-restricted zinc finger protein Ikaros copurifies with PYR complex DNA-binding activity and that antibodies to Ikaros also supershift the complex. We also show that NuRD and SWI/SNF components coimmunopurify with each other as well as with Ikaros. Competition gel shift experiments using partially purified PYR complex and recombinant Ikaros protein indicate that Ikaros functions as a DNA-binding subunit of the PYR complex. Our results suggest that Ikaros targets two types of chromatin-remodeling factors-activators (SWI/SNF) and repressors (NuRD)-in a single complex (PYR complex) to the ␤-globin locus in adult erythroid cells. At the time of the switch from fetal to adult globin production, the PYR complex is assembled and may function to repress ␥-globin gene expression and facilitate ␥-to ␤-globin switching.

We have previously reported the structure of a chromatin remodeling complex (PYR complex) with Ikaros as its DNA binding subunit that is specifically present in adult murine and human hematopoietic cells. We now show that homozygous Ikaros ''knockout'' (null) mice lack the PYR complex, demonstrating the requirement for Ikaros in the formation of the complex on DNA. Heterozygous Ikaros null mice have about half as much PYR complex, indicating a dosage effect for both Ikaros and PYR complex. We also show that Ikaros null mice have multiple hematopoietic cell defects including anemia and megakaryocytic abnormalities, in addition to previously reported lymphoid and stem cell defects. The null mice also have a delay in murine embryonic to adult ␤-globin switching and a delay in human ␥ to ␤ switching, consistent with a previously suggested role for PYR complex in this process. Lastly, cDNA array analyses indicate that several hematopoietic cell-specific genes in all blood lineages are either up-or down-regulated in 14-day embryos from Ikaros null as compared with wild-type mice. These results indicate that Ikaros and PYR complex function together in vivo at many adult hematopoietic cell-specific genes and at intergenic sites, affecting their expression and leading to pleiotropic hematopoietic defects. C hromatin remodeling complexes are associated with both activation and repression of expression of specific eukaryotic genes. SWI͞SNF containing ATPase͞helicase subunits (e.g., BRG1) have been shown to remodel chromatin and activate gene transcription at the ␤-globin locus as well as elsewhere (1, 2). Similarly, NuRD complexes using other ATPase͞helicase subunits (e.g.Mi-2␤) together with histone deacetylases have been associated with chromatin deacetylation and repression of specific gene expression (3). We have recently described a SWI͞SNF and NuRD-containing complex associated with the DNA binding transcription factor Ikaros that is present only in adult hematopoietic cells (4,5). This complex binds to Ikaros-like DNA binding sites including a long polypyrimidine-rich sequence upstream of the human ␦-globin gene and was thus called PYR complex. Deletion of this sequence in a human ␤-locus containing cosmid (carrying sequences from the human A␥ through the adult ␤-globin gene) in transgenic mice results in delayed human ␥-to ␤-globin switching (4).We now have studied Ikaros null mice whose Ikaros gene has been truncated by deletion of the two C-terminal zinc fingers required for Ikaros protein dimerization and function (6, 7). We show that Ikaros null mice completely lack PYR complex, indicating that Ikaros is required for PYR complex formation. In addition, heterozygous null mice have about half as much PYR complex as wild-type mice, indicating a dosage effect for both Ikaros and PYR complex. In this model for evaluating Ikaros and PYR complex function, we find multiple hematopoietic defects including anemia and thrombocytosis, along with previously reported lymphoid and stem cell abnormalities (8). We also show tha...

No abstract

Retroviral-mediated gene transfer into hematopoietic stem cells provides the only means of stable transduction of these cells and their progeny for use with a variety of potentially therapeutic genes. Expression of the Moloney amphotropic retroviral receptor-pit-2 or GLVR-2-is critical to the recognition and entry of Moloney leukemia virus-derived viruses into human target cells such as CD34+ hematopoietic cells. GLVR-2 functions as a sodium-dependent phosphate transporter as well as a receptor. We have previously shown that the expression of the murine homologue of the amphotropic receptor Ram 1, also a phosphate transporter, is developmentally regulated in murine hematopoietic fetal liver cells. We also demonstrated that culture of murine fetal liver cells in phosphate-free (PO(4)-free) medium increases levels of receptor mRNA and makes murine fetal liver cells susceptible to Moloney amphotropic viral gene transfer. We now examine the effect of culture conditions on the expression of GLVR-2 in human CD34+ cells. In this report, we demonstrate that there is a 2-3 fold increase in GLVR-2 mRNA levels in CD34+ cells after 3 days in culture with interleukin 3, interleukin 6, and stem-cell factor. In addition, the use of PO(4)-free medium increases expression of GLVR-2 an additional 2-fold in these cells during this time. These results indicate that GLVR-2 expression can be up-regulated on these cells, and may permit improved retroviral gene transfer efficiencies.