Stanford Report Online

Stanford Report, December 6, 2000
New protein gives researchers colorful method of tracking gene expression


One of the most exciting recent molecular tools available to researchers studying the transformation of genes into proteins has been the green fluorescent protein, or GFP. Tacking the nucleotide sequence of GFP onto a gene of interest allows researchers to monitor that gene's expression, noting when it is first turned on and where the subsequent green-tagged protein goes within the cell or embryo.

But because GFP marks the finished product -- and doesn't change over time -- there is no way of knowing if the cell is still churning out the protein or if the signal is tracking the last members of a dying population.

This distinction can be important in cells where conditions change rapidly, such as those in a developing embryo. But a new protein developed at Stanford that changes from green to red as it ages can help researchers monitor the starts and stops of gene expression. The protein provides an easy, reliable way to monitor and analyze the "history" of gene expression, the researchers say.

"Lots of people are fascinated by this discovery because now you can essentially follow both parts of the regulatory process -- gene activation and deactivation -- by monitoring these colors," said Alexey Terskikh, PhD, a postdoctoral fellow in the laboratory of Irving Weissman, MD.

Terskikh developed the protein, called E5, while trying to enhance the properties of another tracking protein, drFP583. His finding was published in the Nov. 24 issue of Science magazine.

E5 appears as a kaleidoscope of colors, allowing researchers to see not only when a gene is activated, but also whether it is expressed continuously or as a discrete pulse. Newly synthesized E5 is bright green. In contrast, E5 that's been around for several hours is bright red. Constant synthesis gives rise to a population of green and red molecules, which appears yellow or orange to the human eye.

The discovery of E5 was a surprise, Terskikh said. He was trying to improve on the drFP583 protein, which glows red after several hours. He started by randomly mutating the protein to create versions that would turn red more quickly or would express a more narrow range of color. He found the desired mutations but also noticed that other mutants didn't behave properly.

"We started to select these mutants and very quickly noted some strange things," said Terskikh. Mutants that had been bright green one day were bright red the next.

"I thought maybe I made a mistake. So I repeated the experiment several times and found that I must be persistently making the same the mistake." Eventually he was convinced that E5 was actually changing color over time.

Terskikh realized that if E5 displayed the same unique properties in real systems, it could be useful to researchers. Together with collaborators in the laboratory of Stuart Kim, PhD, in the Department of Developmental Biology, and Andrey Zaraisky, PhD, in the Institute of Bio-organic Chemistry at the Russian Academy of Science in Moscow, Terskikh tested E5's activity in the worm Caenorhabditis elegans and the frog Xenopus laevis.

Terskikh and his collaborators inserted the E5 nucleotide sequence next to a heat-inducible promoter -- a special piece of DNA that determines when and where a gene is expressed. At normal temperatures of 20 C, worm embryos with the modified sequence were non-fluorescent. But they glowed green after being at 33 degrees C for one hour. Four hours after this "heat shock," when the promoter had turned off and E5 was no longer expressed, the embryos began turning red. The intensity of the color increased for up to 50 hours.

The shift from green to red suggested E5 might be useful as a molecular timer in living systems. But, as anyone who's burnt a pizza knows, it's not enough that the timer count up or down. It has to be consistent to be useful; each second has to be the same as the second before and the second after. And it has to be the same for every pizza.

In the worm embryos, E5 appeared to pass the test. The green-to-red shift occurred in an almost linear fashion over the first 14 hours. Individual cellular environments didn't seem to affect the speed of the color change: the ratio of green to red occurred uniformly throughout the embryo's different cell types. Also, even though some embryos glowed more than others, the change in the ratio remained constant within the population. The promising result caused Terskikh and his collaborators to refer to E5 as a "fluorescent timer."

When Terskikh tested E5 in frogs, he confirmed patterns of expression that had previously been visible only by immunostaining thin slices of tissue. He attached E5 to a promoter of Otx-2, a gene involved in brain development in frog embryos. By tracking the appearance of green, red and orange fluorescence he showed that Otx-2 is expressed in different areas of the brain at specific times during development. In contrast, when E5 was coupled to Xanf-1, a gene whose expression doesn't vary as much with time, the brain appeared uniformly orange.

Terskikh acknowledged his discovery might be most exciting to developmental biologists studying genes like Otx-2 and Xanf-1. But he pointed out that E5 could be useful for researchers trying to determine what stimulates a gene to turn on or off in response to combinations of environmental cues. A quick, visible color change should allow researchers to mix and match more sets of conditions than previously thought possible and to view the results in real time, for example.