Brain cells that play a critical role in Alzheimer's disease have been created in the laboratory by scientists.
The stem cell breakthrough has been hailed as an important advance in Alzheimer's research, allowing a limitless supply of neurons on which to test new drugs.
Potentially the work could also pave the way to cell treatments - transplanting healthy lab-grown neurons into the brains of patients.
British charity the Alzheimer's Society called the research "a major step forward in developing treatments for Alzheimer's".
Scientists at Northwestern University in Chicago produced the neurons from stem cells extracted from early stage human embryos. Embryonic stem cells have the potential to develop into virtually any kind of tissue in the body.
The "cholinergic" neurons are a special type that help the hippocampus region of the brain to retrieve memories.
It is these brain cells that die off early in Alzheimer's disease. The brain has a relatively small population of the neurons, and their loss has a swift and devastating impact on a patient's recall ability.
Lead researcher Christopher Bissonnette spent six years cracking the genetic code of the stem cells to create the neurons. He was motivated by his grandfather's death from Alzheimer's.
"This technique to produce neurons allows for an almost infinite number of these cells to be grown in labs, allowing other scientists the ability to study why this one population of cells selectively dies in Alzheimer's disease," he said.
The scientists had to switch on an exact sequence of genes to coax the stem cells into making the transformation into cholinergic neurons.
When the artificially produced neurons were transplanted into the brains of mice, they functioned normally. The cells sprouted connecting fibres to the hippocampus and pumped out acetylcholine, a chemical needed for retrieving memories from other parts of the brain.
The research, published in the journal Stem Cells, will make the rapid screening of thousands of potential Alzheimer's drugs possible for the first time.
Promising compounds can be tested over and over again on the laboratory-grown neurons, side-stepping the laborious process of observing their effects on laboratory mice.