The malaria parasite is spread by mosquitoes
Scientists are perfecting a cheap way to synthesise a highly effective malaria drug.
Artemisinin is currently expensive to manufacture, and so is denied to many in the developing world.
US researchers have created a yeast which can churn out large quantities of a related chemical, which can be easily converted into the drug.
Writing in Nature, they say their work may eventually help slash the cost of artemisinin, and improve access.
It would be very valuable if the cost of manufacturing the drug could be cut
Professor David Warhurst
Malaria infects as many as 500 million people a year, and kills more than 1.5 million, mainly in Africa and Asia.
Artemisinin is the drug of choice for treating those infected with multi-drug-resistant strains of the parasite. In combination with other drugs it has proved to be nearly 100% effective.
It is currently extracted from a plant called Artemisia annua (commonly known as sweet wormwood), grown by farmers in Asia.
But natural supplies are limited, and synthesizing the drug is very expensive.
A team from the University of California, Berkeley, succeeded two years ago in engineering bacteria to make a chemical precursor of artemisinin.
They have now gone one step further by developing a strain of yeast that can churn out large quantities of artemisinic acid - a chemical just one tiny change away from the drug itself.
They did this by adding two genes from A. annua to the yeast, Saccharomyces cerevisiae.
Other researchers have shown that this precursor can be converted into artemisinin in a handful of chemical steps.
So in theory, it should be now be possible to manufacture the drug much more cheaply.
The Berkeley team now hope eventually to drive down costs still further by using a similar process to stimulate bacteria to produce artemisinic acid.
Bacteria grow much more quickly than yeast, and so could potentially offer a much more productive source of the chemical.
The researchers say it could still be several years before a microbe-produced version of artemisinin will be widely available.
Lead researcher Professor Jay Keasling said: "While we have made a lot of progress in the past two years, there still are a lot of unknowns."
But he added: "Now that we've got all the parts, I feel it's just a matter of time before we have a microbe ready for scale-up to production."
Professor David Warhurst, an expert in malaria at the Health Protection Agency, told the BBC News website the research was "very interesting and significant".
He said: "Artemisinin and its derivatives are the best available drugs we have got for treating malaria, particularly resistant strains.
"However, the cost is too high for many African governments to cope with, and to get it down to grassroots the price is going to have to be reduced.
"It would be very valuable if the cost of manufacturing the drug could be cut."