Highlight Publication January 2008

Jennifer Mitchell and Peter Fraser
Transcription factories are nuclear subcompartments that remain in the absence of transcription
Genes & Development 22: 20 - 25, 2008
http://dx.doi.org/10.1101/gad.454008

Lay Description

All of the cells in our body contain the same genetic information encoded in our DNA.  This DNA is tightly packed into the microscopic cell nucleus: imagine a 6-mile long thread (the DNA of the genome) stuffed into a football (the nucleus).  Although each cell contains the same genetic information different cell types use different parts of this information to carry out all the functions required in our bodies.

Located on that DNA thread are genes; short pieces of genetic information, each one about two centimetres in length. Each gene is separated by about half a metre of thread whose function is almost completely unknown. Each cell uses only about half of the genes in the genome to survive and carry out their normal functions, and each of the different cell-types (i.e., skin cells, liver cells, brain cells, etc.) uses a unique combination of the available genes. In order for a gene to be used or “expressed”, hundreds of different protein molecules have to cooperate simultaneously to read and copy the DNA code through a process called transcription. The logistical nightmare of how those hundreds of different protein molecules simultaneously zero in on the correct set of genes in the tangled thread of the genome has been a mystery that has puzzled researchers for decades.

This paper turns that problem on its head. Rather than hundreds of protein molecules simultaneously descending on just the correct set of target genes, we provide evidence that the protein molecule machinery is grouped together at way-stations in the nucleus called transcription factories. These factories contain the necessary protein molecules to express or transcribe the correct set of genes.  The existence and nature of these factories has been a much disputed point since they were first described in 1993.  Our data show that these factory proteins remain in the nucleus in the absence of the process of transcription, and even when the genes that were previously engaged there have moved away.  This indicates that transcription factories are distinct structures in the nucleus and suggests they may be preassembled before genes associate with them.

This idea changes the way that we think about how genes are controlled, suggesting that they must move through the nuclear space before they can be transcribed and expressed.  It also opens up a new way for gene expression to be controlled as now it appears that relocation within the nucleus is a required step in regulating gene expression.

About the Lead Author

Jennifer Mitchell joined the Laboratory of Chromatin and Gene Expression headed by Peter Fraser in 2003 after completing her PhD in Medical Science at the University of Toronto.  Her doctoral research, carried out at the Samuel Lunenfeld Research Institute investigated the expression of transcription factor proteins throughout pregnancy and their effect on regulating the expression genes encoding contractile proteins.  At the Babraham Institute she is now investigating the process of trancriptional regulation at a more fundamental level using erythroid cells as a model to investigate the spatial organisation of transcription in the nucleus.