model of transcription
he prevailing view of the RNA polymerase II (RNAP II) transcription cycle is that RNAP II is recruited to a promoter, transcribes a linear DNA template, terminates transcription and dissociates from the template. Subsequent rounds of transcription are thought to require de novo recruitment of RNAP II to the promoter. However, several recent findings, including physical interaction of transcription initiation factors and termination factors with both promoter and terminator regions, challenge this concept. Using a novel technique called “chromosome conformation capture” (3C), designed to probe the three-dimensional architecture of the genome, we have detected gene loops that juxtapose the promoter and terminator regions of yeast genes. Our model is one where gene loops enhance transcription by facilitating translocation of RNAP II from the terminator to the promoter of the same gene. Conclusion from our studies with yeast can be extrapolated to better understand mammalian gene expression as several recent studies have identified comparable gene loops in human cells, including gene loops at the BRCA1 tumor suppressor gene and between the 5’LTR promoter and 3’LTR poly(A) signal of the HIV genome, the causative agent of AIDS.
Extraordinary progress has been made in the past 25 years toward understanding the structure of the eukaryotic genome and the mechanisms that underlie gene expression. Missing from the picture, though, is a fundamental understanding of the 3-dimensional organization of the genome and how higher order chromatin structure affects gene expression. Recently, a novel technique called “chromosome conformation capture” (3C) was developed that enables detection and characterization of long range chromatin interaction. In effect, 3C converts physical chromatin interactions into specific ligation products that can readily be detected by the polymerase chain reaction (PCR) such that the abundance of PCR products represents the frequency of physical interactions (Fig.1).
Using 3C, we have detected gene loops that juxtapose the promoter and terminator regions of genes transcribed by RNA polymerase II in the yeast Saccharomyces cerevisiae. Our initial studies focused on exceptionally long genes, encoding very large proteins. These genes were chosen simply because of the technical advantages offered by long genes for 3C analysis. However, as we refined our 3C technique, we have been able to detect loops at genes that are only a few hundred base pairs in length. Moreover, we have detected gene loops at every gene analyzed in our studies, suggesting that gene looping is a general characteristic of RNAP II transcription.
Yeast offers many advantages over more complex organisms for studies of gene expression and other fundamental cellular processes. One advantage is the wealth of readily available, well-defined mutants that can be used to tease out the mechanism of loop formation and its physiological relevance. Here, we can make an analogy to discerning the relevance of each component of an automobile. If the driver’s left pedal is removed, we would quickly come to the conclusion that the pedal is involved in stopping the car. We would just as quickly conclude that the pedal is critical to the operation of the car as soon as the car sailed through the first red light.
Our approach to understanding gene loops and their significance involves essentially the same logic and approach. For example, using a yeast mutant known to be defective for transcription, we were no longer able to detect gene loops, allowing us to conclude that loop formation requires a pioneer round of transcription. Similarly, we found that specific proteins that either associate with the promoter element (Fig. 1), or are involved in 3’-end processing of the newly minted mRNA, also failed to form loops. These results were especially interesting and informative because the genes encoding these factors had been found 10 years ago to interact with each other, yet we could not explain why proteins at the front end of the gene were functionally interacting with proteins at the back end of the gene. Gene looping nicely accounts for these interactions – and in each case defines a protein involved in loop formation.
So, what proteins are responsible for bridging the promoter-terminator regions to form a gene loop? A clue came once again from our genetics. A gene called sua7 codes for the transcription initiation factor TFIIB. As far as we knew, TFIIB was involved only in initiation. Yet for reasons that we did not understand, sua7 mutants were either dead (or so sick that they wished they were) when combined with mutations in the components of the transcription termination machinery. Using antibody to TFIIB and a technique called “chromatin immunoprecipitation,” we discovered that TFIIB occupies not only the promoter, as expected, but also the terminator. Moreover, specific TFIIB defects (sua7 mutants) that blocked looping also rendered TFIIB unable to occupy the terminator region, with no effect on binding to the promoter.
We also found that TFIIB amino acid replacements encoded by the looping-defective sua7 mutations occur in a region of TFIIB that plugs into RNA polymerase. These results led to our model (Fig. 2) proposing that TFIIB dissociates from RNAP II once mRNA synthesis begins, then reassociates with RNAP II at the terminator when the newly synthesized mRNA is released from RNAP II. This terminator-RNAP II-TFIIB complex then reassociates with the transcription initiation complex forming a gene loop. The beauty of this model is that it nicely accounts for the observed high rates of transcription reinitiation: RNAP II can be handed off from the terminator to promoter for another round of transcription without passing through the rate-limiting step in transcription, namely, recruitment of RNAP II to the promoter. These discoveries and my model were published last year in Molecular Cell and recognized by the 2007 Dean’s Research Award, and highlighted in the press by the “Faculty of 1000.”
The discovery of gene loops adds a new dimension to the mechanism and regulation of gene expression. We are especially encouraged that similar loops have been identified at mammalian genes, most notably the BRCA1 tumor suppressor gene that is involved in the majority of breast cancers, and between the 5’- and 3’-LTRs of HIV, the virus that causes AIDS. Perhaps expression of these and other genes is regulated by looping, thereby offering a novel approach for drug design and therapeutic intervention in cancers and other human afflictions that are a consequence of transcription run amok.
Badri Nath Singh received his BS degree in physics and chemistry from the University of Gorakhpur and MS degree from Goa University, India, in marine biotechnology. He continued his graduate education at the International Centre for Genetic Engineering and Biotechnology, Delhi, in the laboratory of Dr. M.K. Reddy and Professor S.K. Sopory, studying topoisomerase II, an ubiquitous enzyme required for DNA metabolism and the target of many anti-cancer drugs. He was a postdoctoral fellow in the laboratory of Dr. K. V. S. Rao before accepting his current position in 2005. He is the recipient of the 2007 Dean’s Research Award for the best publication by a postdoctoral fellow.