Neurogenin3 Activates the Islet Differentiation Program while Repressing Its Own Expression

0888-8809/04/$15.00/0 Printed in U.S.A. Molecular Endocrinology 18(1):142–149 Copyright © 2004 by The Endocrine Society doi: 10.1210/me.2003-0037 Neurogenin3 Activates the Islet Differentiation Program while Repressing Its Own Expression STUART B. SMITH, HIROTAKA WATADA, AND MICHAEL S. GERMAN Diabetes Center (S.B.S., H.W., M.S.G.), Hormone Research Institute and Department of Medicine (M.S.G.), University of California San Francisco, San Francisco, California 94143-0534 Expression of the proendocrine factor Neurogenin3 determines which progenitor cells in the developing pancreas will differentiate into the endocrine cells of the islets of Langerhans. To better understand how Neurogenin3 directs endocrine differentiation, we examined the mechanisms by which Neurogenin3 regulates the promoters of three transcription factor genes expressed in endocrine precursor cells: the nkx2.2 gene, the PAX4 gene, and the NEUROG3 gene, the human gene encoding Neurogenin3 itself. The function of all three promoters depends on at least one critical E box, a common DNA sequence that forms a binding site for basic helix-loop-helix proteins like Neurogenin3. Neurogenin3 bound to and effectively activated transcription through the nkx2.2 and PAX4 E boxes. In contrast, Neurogenin3 strongly repressed the NEUROG3 promoter, although a proximal E box was required for activity in the absence of Neurogenin3, suggesting that a ubiquitous transcriptional activator may bind to this site, and that Neurogenin3 could act as a competitive inhibitor of this activator. This hypothesis was supported by the lack of evidence for significant intrinsic transcriptional repression capacity in the Neurogenin3 protein, and by the ability of isolated DNAbinding basic helix-loop-helix domains to repress the NEUROG3 promoter. Neurogenin3 produced additional repression, however, when the protein included an intact transcriptional activation domain, suggesting that it may also induce the expression of a downstream transcriptional repressor. In summary, while Neurogenin3 orchestrates islet cell differentiation by activating islet cell transcription factor genes, it simultaneously represses its own promoter. (Molecular Endocrinology 18: 142–149, 2004) A moter down to the proximal few hundred base pairs are highly active in transformed cell lines irrespective of their tissue of origin, demonstrating the presence of universal activators working through the proximal promoter. In contrast, longer promoter constructs can direct expression of a transgene specifically to progenitor cells in the pancreas, gut, and neural tissues (16). Pancreatic transcriptional activators of the HNF6 (17), HNF1 and HNF3/ FoxA (16) families all bind to this longer NEUROG3 promoter. In pancreatic cells not fated to become islet cells, signals through the notch receptor activate the expression of the transcriptional repressor HES-1, which binds to and prevents the activation of the NEUROG3 promoter (4, 16, 18). The mechanisms by which the expression of Neurogenin3 is extinguished in progenitor cells before final differentiation are unknown. In addition, the downstream genes through which Neurogenin3 activates the islet cell differentiation program are largely unknown. Proposed targets for Neurogenin3 include the genes encoding the islet transcription factors NeuroD1 (19), Pax4 (20, 21), and Nkx2.2 (22). Neurogenin3 has been shown to cooperate with HNF1 homeodomain factors in activating both the PAX4 promoter and the chromosomal pax4 gene (21), and with FoxA winged-helix factors in activating the nkx2.2 1A promoter (22), but otherwise little is known about how Neurogenin3 may activate these genes. In the current study, we set out to understand more clearly how Neurogenin3 expression is controlled, and S THE PANCREAS develops from a cluster of multipotent epithelial progenitor cells into the distinct populations of mature cells that form the adult pancreas, a diverse group of transcription factors orchestrate the determination of cell-type fates and the progression of these cells through specific lineages (for review see Ref. 1). Key among these factors is the basic helix-loop-helix (bHLH) factor Neurogenin3. The expression of Neurogenin3 is both necessary and sufficient (2–4) to drive undifferentiated progenitor cells to an endocrine fate, but only initiates the islet differentiation program because it is extinguished before final differentiation of the cells (3, 5). Additional factors, such as the bHLH factor neuroD1 (6) and the homeodomain factors Pax4 (7), Pax6 (8, 9), nkx2.2 (10), nkx6.1 (11), isl1 (12), and PDX1 (13–15) are also necessary for this process of differentiation and for maintenance of the mature, differentiated cells. Subsequent to the identification of Neurogenin3 as a pancreatic proendocrine gene, establishing the factors that lie directly upstream and downstream has stimulated keen interest. Studies of the human NEUROG3 promoter have demonstrated that fragments of the proAbbreviations: bHLH, Basic helix-loop-helix; HSV, herpes simplex virus; TK, thymidine kinase; UAS, upstream activating sequence. Molecular Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community. 142 Smith et al. • Neurogenin3 Activates the Islet Differentiation Program Fig. 1. NEUROG3 Promoter Activity Is Dependent upon an E Box Located at ⫺149 bp The three cell lines shown were transfected with reporter plasmids containing the firefly luciferase gene under the control of the either the wild-type ⫺207-bp NEUROG3 promoter, or the ⫺207-bp NEUROG3 promoter containing a 2-bp mutation in the E box (⫺207-bp ME). Luciferase activities of all samples were determined 48 h after transfection and are expressed relative to the activity of the promoterless backbone vector (pFOXluc1). Results are expressed as the mean ⫾ SEM of data from experiments performed in triplicate on at least three separate occasions. to understand the function of Neurogenin3 on a molecular level. We found that Neurogenin3 can function as a potent transcriptional activator when bound to E elements from the PAX4 and nkx2.2 gene promoters; and one hybrid analysis mapped this activation capacity to the carboxyl terminus of the Neurogenin3 protein but did not detect significant intrinsic transcriptional repression capacity. The proximal NEUROG3 promoter itself contains a critical E box; but surprisingly, exogenously expressed bHLH activating factors including Neurogenin3 itself repress the activity of the NEUROG3 promoter. We propose that Neurogenin3 can act as a transcriptional repressor of its own promoter by competing with a ubiquitous transcriptional activator, possibly in combination with the induction of a transcriptional repressor. RESULTS A previous study demonstrated by transient transfection assay that NEUROG3 promoter constructs as short as 520 bp are active in a variety of cell lin (...truncated)