Engineers have invented a way to store a single rewriteable bit of data within the chromosome of a living cell—a kind of cellular switch that offers precise control over how and when genes are expressed.

For three years, Jerome Bonnet, Pakpoom Subsoontorn, and Drew Endy of Stanford University tinkered with the switch in Escherichia coli to get it just right. The team engineered the bacteria to contain the genes for both red and green fluorescent proteins, as well as the genes for two cut-and-paste enzymes adapted from a bacteriophage—a virus that infects bacteria. By rewriting a specific segment of DNA in the E. coli's chromosome with the bacteriophage enzymes, the researchers determined which color the bacteria glowed under ultraviolet light, flipping between a red or green aura for as many as 100 cell divisions. Endy and his colleagues call their system a recombinase addressable data (RAD) module.

After injecting their DNA into a bacterium, some bacteriophages immediately begin to make copies of themselves with the cell's native machinery. Other times, however, the bacteriophage DNA lies dormant in the bacteria's chromosome, only to jump into action later when triggered by environmental factors. Two bacteriophage enzymes in particular coordinate such changes: integrase—which can weave the viral DNA into the bacteria's chromosome—and excisionase, which cuts the viral DNA out again.

In earlier work, scientists discovered that by tweaking the sites on a bacterial chromosome where a bacteriophage attaches, they could get integrase to invert the DNA segment it inserts into a host chromosome, as if it were cutting out a word in a sentence and pasting it back in backwards and upside down. Endy and his colleagues wondered if they could coax integrase and excisionase to continually flip a segment of DNA between a standard and inverted position inside a living cell's chromosome—somewhat like the way an electronic or binary switch can be off or on, 0 or 1.

In addition to genes that code for these bacteriophage enzymes, and genes that code for red and green fluorescent proteins, Endy and his team introduced into the E. coli genome a specific promoter—a sequence of DNA that begins transcription, the process by which various enzymes and cellular machines translate DNA into RNA, which is eventually translated into working proteins. The promoter that Endy and his colleagues used only initiates transcription in one direction along the E. coli chromosome. In one position, the promoter sends enzymes zipping along the chromosome toward the section that includes the gene for the green fluorescent protein; when inverted, the promoter initiates transcription in the opposite direction, where the red fluorescent gene waits.

Endy, Bonnet and Subsoontorn continually flipped the promoter between the standard and inverted position—thereby determining which color the bacteria glowed—by flooding the bacterial cells with sequential pulses of antibiotics or sugar molecules that activated transcription factors, which are proteins that bind to DNA to turn certain genes on or off. One type of pulse amplified the expression of integrase alone; another pulse amplified the expression of both integrase and excisionase, inverting the promoter. The research is published online May 21 in the Proceedings of the National Academy of Sciences.

"Thus far people have not been able to control flipping back and forth—they flip once and then they're done, or they flip randomly. The real technical advance here is to flip reliably back and forth as many times as we want. It's the rewriteability. As an analogy, writing info onto a blank CD once is not as useful as a rewritable CD."

By replacing the genes for red and green fluorescent proteins with whatever genes they want to study—and subsequently flipping the RAD module promoter back and forth—other researchers can precisely control genes of interest, Endy says. Recently, Endy spoke to some MIT undergraduate students who are trying to create a fail-safe for modified microorganisms that escape from the lab. Ideally, they would engineer the microorganisms to express a fatal gene only if they escaped—exactly the kind of problem that Endy thinks the RAD module can help solve.