The c-Myb DNA-binding domain is highly conserved and defines a family of related transcription factors. In addition to c-Myb, vertebrates also express A-Myb (MYBL1) and B-Myb (MYBL2) proteins, which have highly related structures including nearly identical DNA-binding domains near their N-termini and large, unique C-terminal domains (Fig. 1). The three Myb proteins are often co-expressed, can bind the same sets of DNA sequences, and can activate the same reporter genes in transfection assays, but they activate different sets of target genes and have nonoverlapping functions in differentiation and development (Lipsick et al. 2001; Ness 2003; Sala 2005; Gonda et al. 2008; Ramsay and Gonda 2008; Zhou and Ness 2011), suggesting that the conserved DNA-binding domains are not sufficient to determine the target gene specificities or activities of the Myb proteins (Ness 2003; Zhou and Ness 2011). All three Myb proteins are probably derived from a primordial precursor similar to the one D-Myb protein found in Drosophila, which appears to be most closely related in function to the vertebrate B-Myb protein (Lipsick et al. 2001). Interestingly, plants have a much larger family of Myb-related transcription factors, involved in controlling the expression of differentiation-specific genes (Roy 2016).
The c-Myb DNA-binding domain is composed of three “Myb repeats,” only two of which are retained in the v-Myb proteins, which have structures that resemble homeodomains. The structures of the Myb repeats have been solved using both solution and crystallographic methods, providing important information about the locations of the v-Myb mutations that affect transcriptional activity (Ness 1996). One of the most interesting features of the Myb proteins is that the Myb DNA-binding domain has been so highly conserved – even the surface-exposed residues that do not contact DNA are conserved. This conservation suggests that the surface-exposed residues of c-Myb are important for protein-protein interactions. Indeed, the DNA-binding domain of c-Myb has been found to interact with a large number of protein partners (discussed below), suggesting that the so-called DNA-binding domain is actually an important DNA- and protein-binding domain (Ness 1996, 1999; Zhou and Ness 2011). DNA-binding proteins with Myb repeats are found in microorganisms, insects, plants, and vertebrates (Lipsick et al. 2001). Recently, some laboratories have begun referring to the Myb repeats as “SANT” domains, to reflect their relatedness to several other regulatory proteins that also contain complete or partial Myb domains (Boyer et al. 2002).
c-Myb in Stem Cells and Non-hematopoietic Tissues
The v-Myb oncoproteins transform hematopoietic cells, and the c-Myb protein has been linked to the regulation of hematopoietic cell differentiation and proliferation, both through antisense knockdown experiments demonstrating that c-Myb is required for hematopoietic progenitor cell differentiation in vitro and through mouse knockout experiments showing that c-Myb-deficient mice die in utero due to a defect in definitive erythropoiesis. More recent experiments using conditional knockout strategies demonstrated that c-Myb function is required for the development of most, if not all, myeloid and lymphoid cells (Thomas et al. 2005) and that c-Myb plays an important role in many other tissues, including breast and colon epithelia and some neural tissues (Ramsay 2005; Ramsay and Gonda 2008; Zhou and Ness 2011). Thus, although it is often thought to be a hematopoietic cell-specific transcription factor, c-Myb is actually expressed in a wide variety of tissues, especially in cells that are proliferating and/or differentiating.
With the understanding that c-Myb is expressed in and plays an important role in many other cell types has come the realization that mutated, rearranged, and/or overexpressed versions of c-myb play an important role in the development of many types of human tumors. Once thought of as an oncogene that was specific for leukemias and lymphomas, c-Myb is now recognized as an important oncogene in a variety of human malignancies, including colon and breast tumors and some neural and head and neck tumors (Ramsay and Gonda 2008; Zhou and Ness 2011). For example, recurrent fusions of the MYB gene in adenoid cystic carcinomas suggest that mutated c-Myb is a driver oncogene in that epithelial, head and neck malignancy (Moskaluk 2013; Stenman 2013). Although the perceived importance of c-Myb in human tumorigenesis has grown, the role that c-Myb plays in cell transformation and tumor formation has yet to be delineated. For example, it is not known what genes c-Myb regulates in different types of tumors or even if c-Myb regulates the same genes in hematopoietic, epithelial, or other tumor types. The activities of c-Myb appear to be complex and context specific, and it is not yet clear how those activities play a role in human malignancies (Ramsay and Gonda 2008; Zhou and Ness 2011).
Multiple Pathways Regulate c-Myb Expression
The MYB gene is subject to dramatic and highly regulated alternative RNA splicing, which primarily involves the region of the gene (Fig. 2) that encodes the large C-terminal regulatory domain of c-Myb protein (Fig. 1). Recent studies have shown that the MYB gene can generate as many as 60 distinct splice variants in some cell types. The variants are formed through the use of alternative exons (e.g., 8A, 9A, 9B), by skipping exons, by using alternative splice donor sites that effectively create “short” versions of some exons (e.g., 9 S), and through the combinatorial mixing of these mechanisms in single mRNA molecules. The levels of alternative splicing can be quite variable, but in some leukemia samples, the alternatively spliced variants represent the majority of the MYB transcripts. Interestingly, the alternative splicing does not affect the part of the MYB gene that encodes the highly conserved DNA-binding domain, but leads to changes in the large C-terminal domains of c-Myb protein that affect regulation and activity. Thus, the alternative splicing results in the synthesis of a family of c-Myb proteins with the same DNA-binding domains but different C-terminal domains and distinct activities. Alternative RNA splicing allows the MYB gene to generate numerous, slightly different c-Myb transcription factors with altered specificities. Context-specific regulation of alternative splicing could lead to the production of specialized versions of c-Myb proteins capable of regulating different target genes, for example, during T-cell or B-cell differentiation (Ness 2003; Zhou and Ness 2011).
Mechanisms Controlling the Specificity and Activities of c-Myb
The c-Myb protein has also been shown to interact with numerous proteins that are likely to modify or regulate its activity or help control which target genes are regulated and in which situations (Fig. 3). Some of these interacting proteins include Pim-1, an oncogenic protein kinase that binds and phosphorylates the c-Myb DNA-binding domain and that is regulated in turn by upstream JAK-STAT signaling pathways, cyclin D1 and the cyclin-dependent kinases CDK4 and CDK6, which could regulate c-Myb during the cell cycle (Quintana et al. 2011), and the proline isomerases Cyp40 and Pin1, which could induce changes in conformation of the c-Myb protein (Ness 1999, 2003; Zhou and Ness 2011). Several studies have identified regulatory complexes or mechanisms that involve homeodomain-interacting protein kinase 2 (HIPK2) and simultaneous interactions with both the DNA-binding and the C-terminal domains, suggesting that c-Myb is likely folded into one or more different conformational states that influence which partner proteins it is able to interact with and consequently which genes it can regulate (Zhou and Ness 2011).
Myb Functions at the Decision Point Between Proliferation and Differentiation
A number of studies have linked the activity of c-Myb to the proliferation and differentiation of hematopoietic, epithelial, and neural stem cells (Ramsay 2005; Ramsay and Gonda 2008; Zhou and Ness 2011). The results suggest that c-Myb plays a critical role in the initial differentiation of stem cells to the most immature committed progenitor cells, where c-Myb probably plays an important role in regulating proliferation and the cell cycle. This highlights one of the most important and most difficult to understand properties of c-Myb: It is involved in the regulation of both proliferation and differentiation, two processes that are often thought of as opposite and contradictory. The oncogenic v-Myb proteins have been described as transforming cells by “blocking” differentiation. But experiments using temperature-sensitive variants of v-Myb showed that reactivating or reintroducing the v-myb oncogene into differentiated cells induced the cells to “dedifferentiate.” The v-Myb proteins actually induced a transformed phenotype, probably by inducing the expression of target genes that are specific for immature, proliferating cells, which resulted in increased proliferation.
The Importance of c-Myb as a Human Oncogene
As an oncogene, MYB was first identified in the context of avian leukemia virus-induced leukemias, but mapping the integration sites in retrovirus-induced leukemias in mice showed early on that the MYB gene was a preferred target for activation and that it was an important mammalian oncogene (Ness 1996). Nevertheless, despite high levels of expression in most leukemias and lymphomas and many other tumors, convincing evidence that MYB plays a causal role in human tumorigenesis was only discovered relatively recently. The MYB gene was found to be involved in recurrent chromosomal translocations in some types of human T-cell acute leukemia (T-ALL) and in carcinomas of the breast and head and neck (Ramsay and Gonda 2008; Zhou and Ness 2011; Stenman 2013). These results showed conclusively that activation of the c-myb gene by amplification and/or rearrangement is associated with specific types of human cancers, demonstrating definitively that MYB is an important oncogene in human hematopoietic and solid tumors. Recently, small molecule inhibitors of c-Myb transcriptional activity have been identified and have been shown to block the growth of leukemia cells (Uttarkar et al. 2016), reinforcing the importance of c-Myb activity in human leukemia.
Although there are no longer any arguments about whether MYB is a human oncogene, there is still relatively little information about how MYB transforms cells, either in human or animal systems. The transforming activities of v-Myb require both the DNA-binding and transcriptional activation domains, suggesting that v-Myb transforms cells by activating specific target genes that lead to transformation and oncogenesis (Lipsick and Wang 1999; Ness 2003). However, the identities of those genes have not been elucidated. One of the most intriguing results came from the use of microarrays to compare the patterns of gene expression induced by the normal c-Myb and oncogenic v-Myb proteins. Surprisingly, expression of the c-Myb and v-Myb proteins in human cells led to the activation of totally different sets of target genes. Thus, v-Myb is not merely a more active or deregulated version of c-Myb, but instead activates an entirely different set of target genes (Ness 2003; Zhou and Ness 2011). Furthermore, the two proteins activated almost totally different sets of genes when expressed in different cell types, showing that their activities were extremely context specific. Thus, it seems likely that mutated or alternatively spliced variants of c-Myb expressed in tumor cells probably activate totally different sets of target genes than the normal “wild-type” c-Myb protein. In addition, “normal” c-Myb expressed in cells that have constitutively activated upstream signaling pathways, as often happens in tumors, is likely to regulate different target genes than it would in normal cells.
Summary and Future Directions
The c-Myb protein is a transcription factor with many activities that change in response to numerous upstream signals. It is subject to complex regulation at the level of expression, alternative splicing, stability, and through posttranslational modifications. The oncogenic activity of c-Myb may be due to its role as an integrator of many different upstream signals. This flexibility, even “plasticity,” also makes it vulnerable to activation through mutations or when the upstream signaling pathways that regulate it become corrupted. However, understanding how the activities of c-Myb change and respond to various signals may also offer a potential means of regulating it. Small molecules targeted to c-Myb might be able to shift its activity to induce tumor cells to stop proliferating and instead to differentiate. Clearly, the most important things to do are to identify and characterize the target genes that c-Myb regulates in different situations, in different cell types, in response to different upstream signals and in normal vs. transformed cells. This information will unleash new types of assays for determining how c-Myb is regulated and how its activity might be influenced as a novel therapeutic strategy.
- Quintana AM, Zhou YE, Pena JJ, O’Rourke JP, Ness SA. Dramatic repositioning of c-Myb to different promoters during the cell cycle observed by combining cell sorting with chromatin immunoprecipitation. PLoS One. 2011;6:e17362. doi:10.1371/journal.pone.0017362.CrossRefPubMedPubMedCentralGoogle Scholar