Thomson's lab was the first in the world to produce human embryonic stem cells. They are taken from very early embryos left over from couples' attempts to have test-tube babies at fertility clinics.
Extracted when the fertilised egg has divided just a few times, each cell still "remembers" how to become any kind of cell in the body. Once they get older, cells are programmed and cannot easily change direction in development.
The hope is that these cells can be used to replace the brain cells destroyed in Parkinson's disease, the cells that die in type-I diabetes or damaged spinal cords.
But so far it has been hard to program these cells. Zwaka, whose work is published in the journal Nature Biotechnology, said this method will now help scientists do that.
"You can purify tissues," he said. One problem with human embryonic stem cells is they tend to remember a little too much, and they can form a random mass known as a teratoma instead of the desired tissue.
With this method, the genes can be manipulated so as to control the kind of tissue the cells form.
His team is already trying this with the dopamine-producing brain cells that die off on Parkinson's, an incurable and fatal brain disease that eventually paralyses victims.
The method could also be used, Zwaka said, to create "universal" donor batches, or cell lines, of cells. The genes that cause the body's immune system to reject foreign tissue could be removed. "You could transplant this line into any patient," Zwaka said.
BYPASSING CLONING
This could bypass the need for therapeutic cloning -- another promising but unproved method that involves taking a cell from a patient using cloning technology to make a very early embryo, and then extracting the cells from it for a personalised transplant.
"It is an important alternative," Zwaka said.
Many people oppose therapeutic cloning on ethical grounds, saying the human embryo created in the process is a life and that destroying it is immoral. A bill pending in the U.S. Congress and backed by President George W. Bush would outlaw all forms of cloning, including therapeutic cloning.
Bush has also severely restricted the use of the embryonic stem cells used by Zwaka and Thomson. Their new method, for which they are seeking a patent, would offer a way to use already-existing batches of the cells.
A more immediate use will be in basic science. The method used by Zwaka's team is called homologous recombination, and it is used to create so-called knockout mice.
Scientists can remove genes one by one from mice and see what happens to them. It is often useful for studying human disease -- but not always.
One example is Lesch-Nyhan disease, a rare type of mental retardation caused by a single genetic defect in boys.
"People wanted to study this gene in mice," Zwaka said. But when mice were engineered to lack the same gene that causes the disease in humans, they seemed perfectly fine.
Now the basic molecular mechanisms behind the disease can be studied in human cells, he said.
One big question is whether the method could be used to make designer babies. If genes can be deleted from or added to mice, why not humans?
Zwaka said this is impossible to do with existing technology and also undesirable.
"You'd have to recreate the entire baby from embryonic stem cells," he said. "No one knows how to generate a human out of embryonic stem cells and who wants to do that?"
