In the murky darkness, blue and green blobs are dancing. Sometimes they keep decorous distances from each other, but other times they go cheek to cheek—and when that happens, other colors flare.
The video, reported last year, is fuzzy and a few seconds long, but it wowed the scientists who saw it. For the first time, they were witnessing details of an early step—long unseen, just cleverly inferred—in a central event in biology: the act of turning on a gene. Those blue and green blobs were two key bits of DNA called an enhancer and a promoter (labeled to fluoresce). When they touched, a gene powered up, as revealed by bursts of red.
The event is all-important. All the cells in our body contain by and large the same set of around 20,000 distinct genes, encoded in several billion building blocks (nucleotides) that string together in long strands of DNA. By awakening subsets of genes in different combinations and at different times, cells take on specialized identities and build startlingly different tissues: heart, kidney, bone, brain. Yet until recently, researchers had no way of directly seeing just what happens during gene activation.
They’ve long known the broad outlines of the process, called transcription. Proteins aptly called transcription factors bind to a place in the gene—a promoter—as well as to a more distant DNA spot, an enhancer. Those two bindings allow an enzyme called RNA polymerase to glom onto the gene and make a copy of it.
That copy is processed a bit and then makes its way to the cytoplasm as messenger RNA (mRNA). There, the cellular machinery uses the mRNA instructions to create proteins with specific jobs: catalyzing metabolic reactions, say, or sensing chemical signals from outside the cell.
This textbook take is true as far as it goes, but it raises many questions: What tells a given gene to turn on or off? How do transcription factors find the right sites to bind to? How does a gene know how much mRNA to make? How do enhancers influence gene activity when they can be a million DNA building blocks away from the gene itself?