Research People Photos Publications Links
Home




The Sewer lab focuses on the following three areas of research:

    1.   CYP17 Transcription
                   required for glucocorticoid and sex steroid synthesis

· Lipid Signaling and Transcription (a, b, c)
· Chromatin Modification Oscillators (d)

    2.   The Role of Steroids in Cancer

    3.   Signal-Dependent Modification of Transcription Regulators

1)                      

We continue to establish mechanisms of transcriptional regulation of steroid-synthesizing enzymes, particularly CYP17 (encoding cytochrome P450 c17 with independently regulated cholesterol 17a-hydroxylase and 17,20 lyase activities required for cortisol and androgen formation) in human adrenal cortex fasciculata and reticularis zones. These cells respond to pituitary ACTH by upregulating transcription, and consequently, activity, of CYP17 and other steroid-synthesizing genes via a GPCR and cAMP dependent pathway that results in rapid DNA binding of the nuclear receptor steroidogenic factor-1 (SF-1) to the CYP17 promoter, and the promoters of virtually all genes which encode enzymes involved in acute and chronic steroidogenesis. However, in our experience PKA does not directly modify SF-1. Transcription of CYP17 and synthesis of cortisol (glucocorticoid) or DHEA (androgen) are well correlated.


a) Specifically, our laboratory has uncovered a novel pathway in which cAMP activates sphingolipid metabolism which results in autocrine/paracrine stimulation of a GPCR via secreted sphingosine-1-phosphate, in turn enabling SREBP cleavage, and activation of CYP17 transcription via an upstream sterol regulatory element (Ozbay et al, 2006). SREBPs are proto-transcription factors which reside in the ER and are typically cleaved in response to changes in cellular cholesterol levels. Thus, cholesterol metabolism, sphingolipid metabolism, and ACTH endocrine signaling converge in their regulation of a gene required for steroidogenesis of sex hormones and glucocorticoids.

b) In addition, we have identified sphingosine as a bona fide antagonist of SF-1, which competes for access to the ligand binding pocket of SF-1 and exchanges with other lipids, particularly in response to cAMP (Urs et al, 2006). We have recently identified that one mechanism by which sphingosine antagonizes SF-1 nuclear receptor upon binding is through the disruption of cooperative interaction of SF-1 with coactivators, particularly the acetyltransferase GCN5 and the coactivator scaffold SRC-1 (Dammer et al, 2007). Thus, levels of nuclear lipids, in addition to cholesterol in the ER, and sphingosine-1-phosphate outside the cell, appear to be actively monitored by the endogenous factors which determine expression of CYP17.

c) We are actively characterizing endogenous ligands which are agonists for SF-1 transcription factor activity in steroid-synthesizing tissues. We have identified that ACTH/cAMP increases SF-1 interaction with diacylglycerol kinases, which are also stimulated by PI3K signaling. The effect of this activation is to increase binding of phosphatidic acid with 12 or 14 carbon fatty acids to SF-1, thereby agonising trans activation activity of the receptor (Li et al, 2007). These findings provide the first functional characterization of an endogenous agonist of SF-1 mediated transcription downstream of ACTH/cAMP signaling in adrenal cortex.

d) Professor Sewer has previously shown cooperativity of two RNA helicases (PSF and p54) that have the ability to bind RNA polymerase II, to modulate splicing site choices, and to modulate transcription, in DNA binding with SF-1 to the SF-1 recognized CYP17 promoter. Here, they form a quaternary protein/DNA complex with SF-1 in response to ACTH/cAMP (Sewer et al, 2002). Recently, our lab has verified and expanded the network of cAMP-dependent protein-protein interactions which occur on this region of the CYP17 promoter, focusing on coregulators which mediate chromatin modifications. We find that cAMP induction of CYP17 causes a sequence of recruitment and dismissal of coactivators which include histone N-acetyltransferases, particularly GCN5 and SRC-1, as well as histone methyltransferases (See below figure). Chromatin modification by these factors is closely correlated with ATP-dependent chromatin remodeling which disrupts core histones, particularly by disturbing histone H2 dimers on the CYP17 promoter and transcription start site, enabling read-through and transcription by RNA polymerase II (See second figure below). Our data indicate that inducible transcription activation is a highly dynamic process which involves cycles of chromatin modifications made by transiently recruited coregulators which act to reversibly modify chromatin, then later return to repeat the process. Coregulators involved in repression bind the promoter between transcription cycles, reversing chromatin modifications, pausing transcription, and reducing transcription rate. Corepressors may also prolong or limit transient interaction of coactivators on promoters during induced transcription. Thus, we have established the temporal dynamics of transcription coactivators and corepressors, cooperating with each other and regulating each other’s recruitment and dismissal from the CYP17 promoter in response to cAMP, but also NADH and intrinsic acetyltransferase activity of GCN5 (Dammer et al, 2007, and Dammer et al, in review).

These findings imply that transcription of select genes is a rheostat, integrating metabolism and acute signaling to determine interactions of the components of a transcriptional clock (term coined in Métivier et al, 2003). This clock relies on oscillatory assembly and disassembly of select nucleosomes associated with target genes and preinitiation complexes with chromatin modifying capability. A gene encodes usage of this oscillator in cis but the enzymatic machinery in trans processes the promoter-specific histone code of the CYP17 gene via poorly characterized oscillator(s) which dictate the duration of cooperative coregulator interactions on target promoters. These oscillator(s) are adjusted by metabolic, energetic and other parameters and their output is the delay or continuation of subsequent transcription cycles. In the below model, ATP-dependent chromatin remodeling of nucleosome structure is one key checkpoint or gateway to transcription [before/after] which transcription of a single gene [occurs/does not occur], because activating signal, e.g. cAMP, [remains above/has decreased below] a threshold. The involvement of both corepressors and coactivators in complete cycles of transcription underscores how both groups of factors are likely required for reversible chromatin stucture changes which occur during induced, as opposed to basal, gene transcription. Our data and model sugget that structural coupling of chromatin to transcription-inducing signals is key in a cell's monitoring of transcription-inducing signals, and provides a means for integrating signals into a specific transcription rate, the output of an evolving oscillator centered on transient protein/DNA interactions.


**Denotes cAMP-inducible (cooperative) interaction on CYP17 promoter.

Gross changes in chromatin structure of the nucleosome at the CYP17 promoter where SF-1 binds (gray oval above) are represented specifically in the following figure:




2) We are researching the role of steroid synthesis and bioactive lipids in the initiation and progression of hormone responsive cancer, particularly in breast, endometrial, ovarian, and prostate cancers, where existing evidence shows that steroidogenesis can occur ectopically via transcription of steroidogenic genes in the absence of ACTH/cAMP (Kim et al, 2004; Wang et al, 2001), or other physiological signals usually required to produce steroids. For example, pituitary hormones which induce steroid production are not monitored in breast and prostate, because these tissues normally do not synthesize steroids, though they do respond to them, increasing growth. If these tissues begin to synthesize steroids ectopically, then abnormal, unregulated growth ensues.

We are interested in the roles of nonsteroidogenic cytochromes P450 in the induction of cancer, which can result from inappropriate oxidation of steroids by these enzymes. It is thought that these active steroid metabolites in turn damage DNA. Many known or suspected carcinogens can upregulate P450 genes which inappropriately target steroids. They do this via transcription (co)factors which include nuclear receptors and related transcription factors and coregulators. In cancers which have high levels of these P450 enzymes, hormone metabolism by these enzymes could aid tumor progression by enabling a hormone-dependent accumulation of DNA damage, accelerating cancer progression toward a phenotype capable of metastasis.


3) Post-translational modification of coregulators may occur in response to diverse upstream signals. Such modification is capable of modulating protein-protein and protein-DNA interactions, and thereby, diverse signaling pathways are likely to converge in the modification state of factors which participate in transcription preinitiation complexes, including coregulators described above, and the chromatin modifications that they affect. Because SF-1 is not directly targeted by PKA, though transcription that requires SF-1 is dependent on ACTH/cAMP/PKA pathways, we are actively searching for PKA phosphorylation of coregulators which corresponds with induction of interactions on the CYP17 promoter and, in turn, induction of CYP17 by factors that are induced to cooperate with SF-1. Some of these factors have enzymatic activities that may also introduce other post-translational modifications to preinitiation complexes, and therefore, additional modifications of SF-1 and cooperative coregulators which occur in response to ACTH/cAMP/PKA are also being characterized with regard to their functional and physiological significance. In addition, Dr. Sewer has previously shown that MAPK phosphatase-1, a dual-specificity Tyr/Ser/Thr phosphatase, induces CYP17 transcription during cAMP induction. Because cAMP-induced transcription via SF-1 occurs in successive iterations, or cycles (Dammer et al, 2007, cf. 1d), it is very possible that both kinase activity and reversal of phosphorylation by MKP-1 are required for succession of iterative transcription cycles (Winnay and Hammer, 2006), and we seek the phosphorylation target(s) which may bear out this hypothesis.

Recent work has identified constitutive and clock-regulated kinases casein kinase 2 and glycogen synthase kinase 3ß as SF-1 targeting allosteric modifiers of SF-1 phospholipid ligand binding, thereby providing a mechanism for SF-1 procession through transcription cycles.