There is a potential breakthrough in the war against malaria. Researchers have genetically engineered mosquitoes that are much less effective in transmitting the deadly parasite. The initial research involves mice, but scientists are hoping it will work in humans.

Malaria kills between one and three million people around the world each year, more than AIDS and tuberculosis combined.

Despite years of intensive efforts to eradicate malaria, the number of malaria cases continues to climb. But one day there may be another weapon in the arsenal against malaria - a genetically-altered mosquito that is incapable of spreading malaria to animals or humans.

Marcelo Jacobs-Lorena at Case Western Reserve University in Cleveland, Ohio headed an international team of researchers that genetically engineered a version of malaria mosquitoes that infects mice.

Professor Jacobs-Lorena says the scientists inserted a protein known as SM-1 which makes it hard for the malaria parasite to gain a foothold within the mosquito. "Essentially, what the protein is doing is just coating the surface of the mid-gut, and when the parasite attaches itself to the surface of the mid-gut to invade... it gets stuck there," he explains. "It is like a lock and key. If it cannot find the lock, then nothing works."

In order to infect animals or humans and continue the disease cycle, the malaria parasite must leave the mosquitoes' mid-gut, or stomach, and travel to its salivary glands. Once inside the salivary glands, the parasite can be transmitted to animals or humans through a mosquito's bite. Uninfected mosquitoes become infected with the malaria parasite from humans or animals that carry the parasite.

In an effort to break the cycle, researchers did two types of experiments to see whether SM-1 kept the parasite at bay in the mosquitoes' gut cavity. One looked to see what happened to the genetically engineered mosquitoes when they bit malaria-infected mice. Researchers found that only half of them became infected with malaria parasites.

According to Professor Jacobs-Lorena, the results of the second experiment were far more dramatic. In it, researchers tested whether genetically altered mosquitoes transmitted malaria to uninfected mice.

"The answer was we did three of these experiments, and [in] two of them, the blockage was one hundred percent," he said. "None of the mice bitten by the transgenic mosquitoes got infected. And in one experiment, the number of mice that got infected is much lower."

In theory, genetically altered mosquitoes could be bred with wild ones to produce offspring that could not spread the malaria parasite.

The results of the study were published in the journal Nature.

In a Nature commentary, Fotis Kafatos of the European Molecular Biology Laboratory in Germany calls the results a "milestone" in malaria research. "It is the first time that humans have taken mosquitoes which are good carriers of malaria and made them less able to carry malaria," he said. "And that is an important landmark."

Professor Kafatos cautions the work is at a very early stage, and there is no guarantee that it will work with human malaria.

Mr. Kafatos says efforts are underway in two other areas to make the mosquito inhospitable to the malaria parasite. They include continued attempts to develop a malaria vaccine against the human disease, Plasmodium falciparum. And researchers are trying to find drugs that would essentially do the same thing as the SM-1 protein keep the parasite from thriving within the mosquito.