NEW YORK (GenomeWeb News) – Members of the FANTOM Consortium reported in Cell online today that they have developed a map of transcription factor interactions in mice and humans.
The researchers integrated physical interaction data and information on transcription factor expression patterns to create the atlas, which represents more than 750 human transcription factor interactions and nearly 900 transcription factor interactions in mice. They then started applying this network to better understand the processes behind tissue differentiation and development.
"The availability of this large combinatorial network of transcription factors will provide scientists with many opportunities to study gene regulation, tissue differentiation and evolution in mammals," co-corresponding author Trey Ideker, genetics chief for the University of California at San Diego's department of medicine and a researcher with UCSD's Institute for Genomic Medicine, said in a statement.
Transcription factors contribute to the development and maintenance of various tissue types by interacting with gene promoters and influencing which genes are expressed or repressed — as well as contributing to the timing of this gene regulation.
But transcription factors don't act alone, Ideker and his co-authors explained. Instead, they typically team up with one another and/or with chromatin modifiers or co-factors. Being able to chart such transcription factor interactions "would represent a significant leap forward in our understanding of how tissue specificity is determined," they added.
In an effort to develop such a map, the researchers sifted through data from public databases on 1,988 human transcription factors and 1,727 mouse transcription factors, focusing in on 1,222 human and 1,112 mouse transcription factors for which cDNA clones were available.
They then used the mammalian two-hybrid approach in CHO-K1 cells to screen for protein-protein interactions between these transcription factors, turning up 762 human and 877 mouse transcription factor interactions.
Next, the researchers began applying this interaction data, exploring everything from the relationship between transcription factor interactions and expression to transcription factor conservation and tissue specificity.
When they used reverse transcriptase quantitative PCR to measure transcription factor expression levels in 34 human tissues and 20 mouse tissues, the team found that transcription factors with extensive interactions also tend to be expressed in a wide range of tissue types. On the other hand, transcription factors with more limited interactions usually showed tissue-specific expression.
"[T]issue identity is not determined by tissue-restricted [transcription factors], but relies on tissue-restricted interaction among [transcription factors]," the authors explained. "The majority of [transcription factors] in these networks are broadly expressed, and it is the minority of [transcription factors] that confer tissue specificity."
Between around 34 and 64 percent of the transcription factor interactions seem to be conserved between mice and humans, the researchers noted, with most conserved transcription factors falling into similar sub-groups within the larger interaction network.
By looking at how transcription factor expression patterns clustered in 34 human tissue types, the researchers identified a sub-network of transcription factors containing so-called homeobox transcription factors that can be used to classify tissues based on the embryonic germ layer from which they're derived.
They also delved into more detailed studies of specific groups of transcription factors, using pull-down assays and other approaches to hone in on transcription factors involved in the differentiation of macrophages, a type of white blood cell involved in immune system function.
Down the road, the new transcription factor interaction atlas is expected to serve as a resource for more extensive studies of gene regulation and development.
"It has long been appreciated that gene regulation involves combinatorial interactions among transcription factors. The contribution of the present work is to map, on a global scale, precisely what many of these connections are," the team concluded. "Future work will dissect more precisely how each of these combinations contributes to developmental programs and to an individual's relative state of health or disease."