updated 10/17/2008 8:44:50 PM ET 2008-10-18T00:44:50

Tiny biology-based computers could eventually check up on the health of individual cells, and, based on what they find, could then treat those cells. That's according to new research from scientists at the California Institute of Technology published in the current issue of Science.

"This is the first time that someone has showed a molecular computer that can respond to stimuli inside a living cell," said Ehud Shapiro, a professor of computer science and biology at the Weizmann Institute of Science.

Other scientists have created bio-based computers. Years ago Shapiro demonstrated a DNA-based computer in a test tube. The idea was that since computer code is based on 0s and 1s, DNA's four bases would significantly improve the amount of information a computer could process.

Most scientists have largely given up on challenging electronic computers in processing power, says Shapiro. Instead research has moved to producing a bio-computer that operates in areas where an electrical computer would find difficult.

"DNA computing is all done outside of cells," said Christina Smolke, a coauthor of of the paper. "To make a computer work inside cells you have to use RNA."

As organic molecules go, RNA is pretty simple. Think of it as a piece of string dyed in four different colors. Each color corresponds to one of four nucleotides. Other molecules inside a cell read the colors on the RNA and use that information to build other molecules called proteins.

Proteins have a wide variety of functions. Some, like hemoglobin, deliver oxygen from the lungs to the far reaches of the body. Other proteins, like the one Smolke and coauthor Maung Win used for their RNA device, glow green when exposed to a laser.

The Caltech RNA device operates much like a normal computer, in theory. Information goes in one end and, based on the data, the computer makes a decision. One big difference between normal computer and an RNA device is that a normal computer uses electrical information. A RNA device uses chemical information.

The scientists built their machine, injected it into yeast cells and then exposed the cells to different concentrations of theophylline, which treats certain respiratory diseases, and tetracycline, an antibiotic.

The presence or absence of those chemicals was the input. That information determined whether or not a string of RNA was snipped. Intact RNA produces the glowing green protein. RNA that has been cut does not produce green light.

If the device detected a chemical or a certain combination of the two chemicals, then the strand of RNA was left alone, which produced the green glowing protein.

The beauty of their design, says Smolke is that many of the parts are essentially interchangeable. Right now the more tetracycline, for instance, the more light the yeast cells emit. The scientists say they could change their device so that more tetracycline would produce less light, for instance.

Producing light is proof of principle. What really excites scientists are the potential medical applications, particularly in fighting cancer and infectious diseases.

In an ideal scenario, still years away, RNA devices like this could be injected and essentially pay house calls on each cell in the human body. Once inside the cell, if the RNA computer finds tracers of a virus or cancer it could release the appropriate drugs to kill just that one cell.

"Today you typically ingest a drug and it operates everywhere and at the same time," said Shapiro.

"Ideally you want it to operate at just the right place and at the right time. [An RNA device] would be ideal for selectively killing cancer cells," said Shapiro.

Killing cancer cells inside a human would likely require more than three parts. The next big step, says Smolke, is creating more parts that perform more functions and then cataloging them, much the same way an engineer can flip through a catalog and find exactly what kind of steel or concrete they needs for a building. Right now biologists essentially custom design every part, which can be time consuming.

Smolke and her colleagues aren't waiting around for that catalog to be created. They have already moved from yeast cells and are extracting T cells, which are part of the immune system, from mice. They are re-engineering those T cells and placing them back into the mice to hunt down and kill diseased cells.

"What we want to do is take a patient's own immune cells, re-engineer them outside the body, and put them back into patients," said Smolke.

© 2012 Discovery Channel


Discussion comments


Most active discussions

  1. votes comments
  2. votes comments
  3. votes comments
  4. votes comments