However, Venter said that his approach would eliminate inherent biological traits that make extensive biological engineering tricky. For example, a nonessential metabolic process could suck away a molecular precursor needed for fuel production. In addition, said Venter, "existing organisms have a rapid ability to evolve. In synthetic biology, you don't want a system that will self-evolve into something else. We want to eliminate those elements from the cell from the beginning."

The genomic transfer technique is similar to nuclear transfer--used to clone Dolly the sheep--in which the nucleus of an adult cell is transferred into an egg. But getting the process to work in bacteria has been trickier. Scientists speculate that nuclear proteins transferred along with the DNA during nuclear transfer may help the process. Bacteria do not have a nucleus, and in this experiment, only DNA was transferred into the host cell. Researchers did this on purpose to show that only DNA was needed to successfully reprogram the host bacterium--a property that will be necessary when scientists are ready to transplant entirely synthetic genomes.

Scientists don't fully understand how the genome transfer worked, particularly how the host genome disappeared. And it's not yet clear how well the technique will work in other bacteria or in more-complex organisms. Most bacteria have a defense mechanism that chops up any foreign DNA that enters the cell, so scientists would need to find a way to block the DNA-degrading enzymes in each different species before they could successfully transplant foreign genomes into them.