Synthetic memory in eukaryotes?

Scientists have engineered a DNA-based device that enables eukaryotic cells to "remember" past exposure to a specific stimulus, according to a new report in Genes and Development.

The researchers used a mathematical model to design a genetic switch, triggered by the small sugar molecule, galactose, that maintains steady production of proteins even after the galactose is removed. Similar past efforts in eukaryotes have instead relied on trial and error. The study shows the possibility of building not just fully functional but also completely predictable synthetic components inside the nuclei of single cells, the authors say.

"It's a really nice study. It will help set the stage for additional rational approaches in eukaryotes," James Collins of Boston University, who was not involved in the work, told The Scientist.

He explained that other researchers have used models to describe similar systems in eukaryotes but only after they actually built their systems. "Here," Collins said, "they integrated a mathematical model with their experimental work and actually used the model to help guide the rational design of their network."

Led by Pamela Silver of Harvard Medical School, the team started by defining the quantitative properties of the memory device that they wanted to build, such as rate at which it would produce protein. Then they fed those properties into a mathematical model to describe the desired behavior of the device. Finally, they built the device based on the model's output, and tested and tinkered until it worked.

"We found that under certain conditions the system behaved as it should, but under other conditions it didn't," Silver explained. "Because we had this model, we could ask, what do we have to change it make it work? We figured out what that was, we did it, and the system worked."

The device operated as a closed loop inside a yeast cell, desgined to start running only in the presence of a particular stimulus, in this case galactose. Its components were simple: two synthetic transcription factor-coding genes, which the researchers synthesized using their model and random bits of DNA.

The galactose induced the first gene to switch on production of its transcription factor which, in turn, triggered the second gene to produce its protein; the second transcription factor re-activated itself, creating a self-perpetuating feedback loop. The scientists used fluorescently labeled transcription factors to show that even after the sugar was removed, the loop continued to run. In other words, the cell remembered that it had once been exposed to a particular environmental signal.

"It is a very robust device," coauthor David Drubin told The Scientist. He explained that the researchers grew multiple cultures of the memory device-containing yeast cells and observed that about 90 percent of the cells in each culture retained their memory for at least eight generations of cell division. Other memory modules in eukaryotic cells have "petered out," he said.

The study sets the standard for work to come, according to Ron Weiss of Princeton University, who was not a participant. "This is a really nice demonstration that you can get a functional module by using computational design. As our capabilities improve, we are going to start building more and more complex networks, with many more interacting pieces, and it is going to be essentially impossible to design a system without computation."

For that to occur, said Silver, researchers will need access to affordable and efficient DNA synthesis. Silver noted that when researchers can cheaply order custom-designed DNA from a commercial provider, synthetic biology will advance "very quickly." Meanwhile, Drubin is working on building a similar device in a more complicated, mammalian cell.

Leslie A. Pray
mail@the-scientist.com

Links within this article:

C. Ajo-Franklin et al., "Rational design of memory in eukaryotic cells," Genes and Development, September 15, 2007.
http://www.genesdev.org

A. Becskei et al., "Positive feedback in eukaryotic gene networks: cell differentiation by graded to binary response conversion," EMBO J, May 15, 2001.
http://www.the-scientist.com/pubmed/11350942

J. Lucentini, "Is this life?" The Scientist, January, 2006.
http://www.the-scientist.com/2006/1/1/30/1/

James Collins
http://www.bu.edu/abl/

Pamela Silver
http://silver.med.harvard.edu/

David Drubin
http://openwetware.org/wiki/User:DavidDrubin

N. Vilaboa et al., "Novel gene switches for targeted and timed expression of proteins of interest," Molecular Therapy, August, 2005.
http://www.the-scientist.com/pubmed/15925546

Ron Weiss
http://www.ee.princeton.edu/people/Weiss.php

P. Silver and J. Way, "Cells by design," The Scientist, September 27, 2004.
http://www.the-scientist.com/2004/9/27/30/1/

C.Q. Choi, "Bacteria genome switch-a-roo," The Scientist, June 28, 2007.
http://www.the-scientist.com/news/display/53341/