Transcription Factor PU.1
PU.1 is a DNA-binding nuclear protein (transcription factor) that is a member of the E26 transformation-specific (ETS) family. PU.1 was named because it can interact with a “PU box,” a stretch of purines in the SV40 viral enhancer, that was identified as being functionally active in lymphoid cells (Petterson and Schaffner 1987). SPI1 was independently identified as the protein product of a gene dysregulated in murine erythroleukemia, as a consequence of proviral integration of the spleen focus-forming virus (SFFV) (Moreau-Gachelin et al. 1988). This gene/protein was named SPI1 for “spleen focus-forming virus proviral integration site 1.” PU.1 and SPI1 were subsequently recognized to be the same protein (Goebl 1990; Paul et al. 1991). PU.1 is encoded by Spi1 on mouse chromosome 2 and by SPI1 on human chromosome 11 (Nguyen et al. 1990). The presence of functional PU boxes led to the identification of PU.1 as a transcription factor that could interact with the immunoglobulin (Ig) kappa 3′ enhancer (Pongubala and Atchison 1991; Pongubala et al. 1992), the Ig heavy chain intronic enhancer (Nelsen et al. 1993), and the CD11b promoter (Hickstein et al. 1992). PU.1 was thereafter described to interact with a number of lymphoid- and myeloid-specific promoters and enhancers (Turkistany and DeKoter 2011).
PU.1 Structure and Function
PU.1 interacts as a monomer with an AGAA or GGAA purine-rich core motif (Fig. 1b). However, PU.1 can also bind DNA cooperatively with interferon regulatory factor (IRF4) or IRF8 transcription factors to recognize an ETS-IRF composite element (EICE) that has the consensus sequence GGAAgtGAAA (Brass et al. 1999; Escalante et al. 2002). Cooperative binding of DNA with IRF4/IRF8 requires interaction with serine 148 located in the PEST domain of PU.1 (Eisenbeis et al. 1995). Another well-studied interaction is between PU.1 and the regulator of erythroid cell fate GATA-1 (Rekhtman et al. 1999; Nerlov et al. 2000; Zhang et al. 2000). PU.1 protein interacts with GATA-1 protein to impair the enforcement of erythroid cell fate by GATA-1, whereas GATA-1 interaction with PU.1 protein impairs enforcement of myeloid cell fate by PU.1 (Stopka et al. 2005; Hoppe et al. 2016).
Studies utilizing chromatin immunoprecipitation coupled with next-generation sequencing (ChIP-seq) permitted genome-wide analysis of PU.1 interaction in four hematopoietic lineages: erythroid (Wontakal et al. 2011), B lymphocyte (Heinz et al. 2010; Schwickert et al. 2014; Solomon et al. 2015), T lymphocyte (Zhang et al. 2012), and macrophage (Ghisletti et al. 2010; Heinz et al. 2010). These studies revealed that PU.1 interacts with transcription start sites, within gene bodies, and within intergenic regions. PU.1-binding locations highly correlate with locations of predicted enhancers (Ghisletti et al. 2010; Whyte et al. 2013). PU.1-binding locations highly correlate with locations of activating histone modifications, including in T cells and erythroid cells where PU.1 functions as an antagonist of differentiation (Ghisletti et al. 2010; Wontakal et al. 2011; Zhang et al. 2012). In macrophages, PU.1 binding is associated with nearly all enhancers marked by the activating histone mark H3K4me1 (Ghisletti et al. 2010; Lara-Astiaso et al. 2014). PU.1 interaction with target sites in the genome may be an important first step for nucleosome remodeling, facilitating binding of other transcription factors and cofactors (Ghisletti and Natoli 2013). In summary, PU.1 interaction with the consensus motif GGAA or AGAA at transcription start sites and enhancers is associated with recruitment of transcriptional cofactors and induction of activating histone modifications.
PU.1 Expression and Regulation
PU.1 expression is regulated by both transcriptional and posttranscriptional mechanisms. Spi1 transcription is regulated by interaction of the promoter region with an upstream regulatory element located at − 14 kb relative to the transcription start site (Rosenbauer et al. 2004; Leddin et al. 2011). Mutation of the − 14 kb URE sequence led to reduced Spi1 transcription in myeloid and B cells and increased PU.1 expression in T cells (Rosenbauer et al. 2006). The half-life of PU.1 protein allows for accumulation during increased cell cycle length (Kueh et al. 2013). Finally, PU.1 expression is modulated by microRNA-155 in the B cell lineage. Mutation of the gene-encoding miR-155, or mutation of the miR-155 target site in the 3′ UTR of the Spi1 gene, caused PU.1 expression to be upregulated in B cells (Vigorito et al. 2007; Lu et al. 2014). In summary, PU.1 levels are regulated by transcription, by protein stability, and by microRNAs.
Biological Function of PU.1
The biological function of PU.1 has been studied using genetically modified mouse models. Descriptions of mice homozygous for germline null alleles of Spi1 were published by Harinder Singh’s laboratory in 1994 (Scott et al. 1994) and Richard Maki’s laboratory in 1996 (McKercher et al. 1996). The DeKoter laboratory described mice homozygous for a hypomorphic allele of Spi1 in 2007 (Houston et al. 2007). These studies showed that PU.1 is required to generate all white blood cell lineages including B cells, T cells, granulocytes, and macrophages. Erythroid and megakaryocyte development was largely normal in these mice. In order to determine at what stage PU.1 is required for development, conditional knockout alleles of Spi1 were described by three groups in 2005 (Iwasaki et al. 2005; Polli et al. 2005; Ye et al. 2005). These studies revealed that PU.1 is required to generate myeloid and lymphoid progenitors. Spi1 deletion at any stage resulted in blocked myeloid development (Anderson et al. 2000; Iwasaki et al. 2005; Chopin et al. 2013). After commitment to the B cell lineage, PU.1 is not required for B cell development, because its function is complemented by the highly related transcription factor Spi-B. In the absence of both PU.1 and Spi-B, B cell development is arrested at an early stage (Sokalski et al. 2011).
PU.1 concentration plays key roles in regulation of immune function. Upregulation of PU.1 expression in B cells, caused by mutation of the gene-encoding miR-155 or by mutation of the miR-155 target site in the 3′ UTR of the Spi1 gene, led to defects in B-T interaction, impaired antibody-forming responses, and impaired plasma cell differentiation (Vigorito et al. 2007; Lu et al. 2014). In the T cell lineage, PU.1 expression is downregulated at the beta-selection checkpoint during T cell development in the thymus (Anderson et al. 1999). However, PU.1 is upregulated in a peripheral T cell population capable of expressing interleukin-9 (IL-9). This population, known as T helper 9 (Th9) cells, requires PU.1 for their proper development and function (Chang et al. 2010). This group showed that T helper cells are exquisitely sensitive to PU.1 concentration, as even low levels of expression have major functional consequences for T helper cell development (Awe et al. 2015).
PU.1 concentration plays an important role in regulating self-renewal of hematopoietic stem cells (HSCs). Deletion of Spi1 in HSCs promoted premature cell cycle progression and consequently stem cell exhaustion (Staber et al. 2013; Will et al. 2015). In contrast, increased expression of PU.1 in HSCs, induced by macrophage colony-stimulating factor (M-CSF) signaling, opposed self-renewal and increased the frequency of “choice” toward the myeloid cell fate (Mossadegh-Keller et al. 2013). Therefore, HSC and myeloid progenitor cell self-renewal is sensitive to PU.1 concentration. In general, high PU.1 levels promote increased differentiation and reduced self-renewal, while reduced PU.1 levels lead to reduced differentiation and increased self-renewal. Taken together, these studies showed that PU.1 levels are biologically important for development, differentiation, and self-renewal of hematopoietic lineages.
Role of PU.1 in Disease
PU.1 has been associated with leukemia since its discovery. In 1988, researchers first identified Spi1 as a gene recurrently disrupted in murine erythroleukemia by proviral insertion of murine spleen focus-forming virus (SFFV) (Moreau-Gachelin et al. 1988; Goebl 1990). SFFV proviral integration resulted in overexpression of PU.1; and overexpression of PU.1 was sufficient to induce erythroleukemia in mice (Moreau-Gachelin et al. 1996). In contrast, reduced levels of PU.1 in mice can cause B cell acute lymphoblastic leukemia (B-ALL) and acute myeloid leukemia (AML). Mice that developed AML as a consequence of gamma-irradiation had recurrent deletions or point mutations of a 2Mbp region of chromosome 2 that included Spi1 (Cook et al. 2004). Mice expressing 20% of normal levels of PU.1, as a consequence of deletion of an upstream regulatory element (URE) located at −14 kb relative to the transcription start site, developed myeloid leukemia at high incidence (Rosenbauer et al. 2004, 2006). Interestingly, these mice also developed T cell lymphoma associated with increased levels of PU.1 in the thymus (Rosenbauer et al. 2006). Reduced PU.1 levels led to a preleukemic condition by increasing rates of proliferation and reducing differentiation of hematopoietic stem cells and myeloid progenitor cells (Staber et al. 2013; Ziliotto et al. 2014; Will et al. 2015). Even a modest 35% reduction in PU.1 expression strongly cooperated with Msh2 homozygous mutation to promote abnormal self-renewal of myeloid cells leading to leukemia (Will et al. 2015). Restoration of PU.1 levels in AML cells restored the monocyte-macrophage differentiation program and opposed continued proliferation (Sive et al. 2016). In the B cell lineage, mutation of Spi1 led to B-ALL, but with low incidence and with long latency (Pang et al. 2016). However, mice with deletions of genes encoding both PU.1 and Spi-B had impaired B cell development and developed B cell acute lymphoblastic leukemia at 100% incidence by 21 weeks of age (Sokalski et al. 2011). Thus PU.1 and Spi-B function as complementary tumor suppressors in the B cell lineage. In summary, PU.1 is a tumor suppressor in the B cell and myeloid lineages but may function as an oncogene in the erythroid and T cell lineages. These studies illustrate the importance of maintaining appropriate PU.1 levels throughout hematopoietic development.
Mutation of human SPI1 is not associated with erythroleukemia or T cell leukemia, but is associated with human acute myeloid leukemia (AML). PU.1 expression is repressed by the protein product of the RUNX1-ETO translocation, one of the most frequent chromosomal abnormalities in AML (Vangala et al. 2003). Reduced PU.1 levels are associated with MLL mutations in mixed lineage leukemia (Lavallee et al. 2015). In one cohort of patients, inactivating mutations of SPI1 were associated with AML (Mueller et al. 2003). These studies suggest that altered levels of PU.1 protein in human cells contribute to leukemogenesis.
Genome-wide analyses of enhancers associated with disease have implicated PU.1 as potentially having important roles. PU.1 marks super-enhancers in B cells that are associated with a wide variety of diseases (Hnisz et al. 2013; Whyte et al. 2013). Alzheimer’s disease was recently shown to be associated with upregulation of immune response genes that are targeted by PU.1 (Gjoneska et al. 2015). In summary, these studies indicate that perturbations in PU.1 levels are associated with cancer, B cell dysfunction, and inflammation.
PU.1 is an ETS transcription factor that is highly expressed in hematopoietic cell types, with expression highest in myeloid cells, lower in lymphocytes, and lowest in red blood cells and megakaryocytes. PU.1 interacts with “PU boxes” in the genome to activate transcription of target genes. Deletion of the gene encoding PU.1 results in multiple hematopoietic defects including loss of development of myeloid cells and lymphoid cells. Reduced PU.1 expression leads to impaired differentiation and increased proliferation of HSCs, myeloid progenitor cells, and progenitor B cells. Even modestly reduced or increased PU.1 expression can lead to leukemia and impaired immune function. Rapid advances in next-generation sequencing approaches will allow a better understanding of the mechanisms by which deregulated levels of PU.1 lead to diseases such as leukemia. The global identification of PU.1 target genes in myeloid and B cell progenitors, as well as the determination of genomic alterations found in leukemic cells, will enhance knowledge of the molecular mechanisms of human disease.
- Hagemeier C, Bannister AJ, Cook A, Kouzarides T. The activation domain of transcription factor PU.1 binds the retinoblastoma (RB) protein and the transcription factor TFIID in vitro: RB shows sequence similarity to TFIID and TFIIB. Proc Natl Acad Sci U S A. 1993;90(4):1580–4.Google Scholar
- Mueller BU, Pabst T, Osato M, Asou N, Johansen LM, Minden MD, et al. Heterozygous PU.1 mutations are associated with acute myeloid leukemia. Blood. United States2003;101:2074.Google Scholar
- Nelsen B, Tian G, Erman B, Gregoire J, Maki R, Graves B, et al. Regulation of lymphoid-specific immunoglobulin mu heavy chain gene enhancer by ETS-domain proteins. Science (80-). 1993;261(5117):82–6.Google Scholar
- Scott EW, Simon MC, Anastasi J, Singh H. Requirement of transcription factor PU.1 in the development of multiple hematopoietic lineages. Science (80-). 1994;265(5178):1573–7.Google Scholar
- Solomon LA, Li SK, Piskorz J, LS X, DeKoter RP. Genome-wide comparison of PU.1 and Spi-B binding sites in a mouse B lymphoma cell line. BMC Genomics. 2015;16.Google Scholar
- Yamamoto H, Kihara-Negishi F, Yamada T, Hashimoto Y, Oikawa T. Physical and functional interactions between the transcription factor PU.1 and the coactivator CBP. Oncogene. 1999;18(7):1495–501.Google Scholar