EFFECTS OF INHIBITING CDK9 ON THE EXPRESSION OF PRIMARY RESPONSE GENES

Abstract

Flavopridol (FVP) is a well known pharmacological inhibitor of Cyclin Dependent Kinases (CDKs), with significant selectivity for Cyclin Dependent Kinase 9 (CDK9). Treatment of cells with FVP results in inhibition of transcription elongation. CDK9 is a serine/threonine kinase that associates with T-type cyclins. These complexes are designated transcription elongation factors (P-TEFb). P-TEFb controls transcription elongation by phosphorylating the carboxyl terminal domain (CTD) of RNA polymerase II (RNAPII) and negative elongation factors. Whether P-TEFb is required for the elongation of most genes transcribed by RNAPII or fraction of them is still debatable. The aim of my Thesis is to understand the early and late effects of FVP on primary response gene expression. Two different microarray analyses with RNA probes obtained from T98G and BJ-TERT cells were performed by Drs. Graña and Garriga to determine the effect of inhibiting CDK9 on global mRNA expression using a dominant negative mutant of CDK9 (dnCDK9) and FVP. These gene profiling experiments showed that FVP and dnCDK9 downregulate the expression of several genes. However, these studies also showed upregulation of a group of primary response genes (PRGs). The goal of this thesis was to bring some light into this unexpected phenomenon. I have found that several PRGs including FOS, JUNB, EGR1 and GADD45B, are rapidly and potently downregulated before they are upregulated upon FVP treatment in exponentially growing cells. In serum starved cells restimulated with serum, FVP also inhibits the expression of these genes, but subsequently, JUNB, GADD45B and EGR1 are upregulated in the presence of FVP. Chromatin Immunoprecipitation of RNAPII revealed that EGR1 and GADD45B are apparently transcribed at the FVP-treatment time points where their corresponding mRNAs accumulate. These results suggest a possible stress response triggered by CDK9 inhibition. I also show that serum starvation does not affect the localization of RNAPII immediately downstream of the promoter of a PRG where RNAPII remains paused in the absence of mitogenic stimulation, suggesting that initiation is not rate limiting for transcription of at least certain PRGs in the absence of mitogens and remains dependent on transcription elongation. In sum, I have shown that certain PRG/IRGs are transcribed in the presence of FVP and their transcription might be independent of CDK9 suggesting a possible alternative mechanism of their transcription. I also determined that transcription initiation is not affected by serum starvation, as paused RNAPII appears to remain bound downstream of a PRG promoter in quiescent cells independently of the length of mitogenic starvation.