How an experiment to change the colour of a petunia led to a breakthrough in the treatment of cancer and Aids

It started with the flowers of a purple petunia plant turning white, and ended with a human cell becoming resistant to the deadly embrace of the Aids virus. The intervening decade took in experiments with yeasts, microscopic worms, mice and flies. And they all pointed to one thing: a potential revolution in medical science.

It started with the flowers of a purple petunia plant turning white, and ended with a human cell becoming resistant to the deadly embrace of the Aids virus. The intervening decade took in experiments with yeasts, microscopic worms, mice and flies. And they all pointed to one thing: a potential revolution in medical science.

When Professor Richard Jorgensen, a plant scientist at the University of Arizona in Tucson, tried to make his petunia flowers a deeper shade of purple, he had little idea that he was about to find the key to one of the hidden mysteries of life.

Nor could he have anticipated that his petunia observation would become the basis of one of the most promising weapons in the war against viruses and cancer.

Professor Jorgensen was interested in the esoteric mechanism that controls the regulation of genes in plants. In pursuit of this interest in basic biology, he decided he would try to make purple petunia flowers even more purple by injecting them with the gene for pigment coloration.

To his surprise, the flowers bloomed white. Instead of the two sets of pigment-producing genes complementing each other, they seemed to interact by turning themselves off.

Some flowers bloomed totally white, while others were variegated – including one that his wife and collaborator, Carolyn Napoli, named the "Cossack dancer" because it looked like a man in a voluminous costume with his arms and legs outstretched.

In 1990, when the research was published, Professor Jorgensen called the phenomenon "cosuppression" because it seemed that both sets of pigment genes were preventing each other from working properly. This was one of those unexpected and counter-intuitive findings that sometimes make scientific research so interesting, yet so frustrating.

At first it was thought that this observation was unique to petunias, but other scientists repeated the work and found it was also true for other species of plants. Then some yeast scientists got in on the act as well and found that the process worked particularly well in their beloved laboratory creature, a micro-organism called Neurospora crassa.

Just to show that they had done something different, they called it by another name – gene "quelling".

But no one had any clear idea as to how this quelling or cosuppression worked, or even whether it had any particular importance outside the world of plants and yeasts.

At the same time as these experiments were taking place, molecular biologists had been working on something called "antisense" technology. This was a way of turning genes off using a close cousin of DNA (deoxyribonucleic acid) called RNA (ribonucleic acid).

Antisense worked by injecting into a cell a molecule of RNA that was complementary in its genetic sequence of chemical bases to the RNA of the cell involved in the synthesis of proteins from genes. It was hoped that an antisense strand of RNA would block the manufacture of a particular protein, thereby shutting off the gene.

But yet again something happened that was unexpected. Scientists working on the tiny nematode worm, the species Caenorhabditis elegans, found that the antisense technique worked best when the RNA was injected in the form of a double-stranded molecule, instead of its usual single-stranded form.

The process remained something of a puzzle until a few years later, when Andrew Fire, a researcher at the Carnegie Institution at the Johns Hopkins University in Baltimore, Maryland, looked into the problem. He too had been interested in injecting antisense material into the nematode worm to study the switching on and off of genes.

As is normal in science, Dr Fire set up a "control" experiment which was not supposed to produce any results of interest and was merely there to compare against the actual experiment. In fact it was the control experiment that proved to be seminal.

"What we found was that our control experiments never worked properly," Dr Fire said yesterday. "Not only was the normal gene shut off but the gene we were putting in was shut off as well."

Dr Fire couldn't help but notice that the phenomenon was very similar to the original petunia experiment he had heard about. He called it gene "silencing" – the nematode version of cosupression in petunias and quelling in yeast.

"My lab worked pretty hard to sort out what the actual structure of the RNA that was causing the silencing was. We were surprised because it turned out not to be the major component we were injecting, but a contaminant that is known to be present when you make RNA in a test tube," Dr Fire said.

That contaminant was double-stranded RNA – when a "sense" and an "antisense" strand wrap around each other to form a single molecule, rather like the double helix of DNA. The discovery revealed that, when RNA came in its double-stranded form, it was capable of silencing genes.

Dr Fire and his colleagues, including Craig Mello of the University of Massachusetts, published their results in the scientific journal Nature on 19 February, 1998. They called their discovery "RNA interference", or simply RNAi, and said how surprised they were about the power of double-stranded RNA to silence genes.

Scientists in other disciplines began to take notice. It seemed that Dr Fire and Dr Mello had discovered a way of turning off genes using RNA, and this was not what every biology student had been told to expect. In fact, the "central dogma" of biology was that RNA was the necessary intermediary in the process of turning genes on and making the proteins that are essential for life.

RNA was supposed to be the lubricant that allowed genes to make proteins, not the spanner in the works.

Professor Phillip Sharp, a Nobel laureate at the Massachusetts Institute of Technology, was intrigued by the findings. "It was almost a retro-process," he said. "I was just dumb-founded that it hadn't been described before."

Now, the giants of genetics – scientists working on the Drosophila fruit fly – went buzzing into action. Everybody wanted to know what, precisely, was going on. How could RNA be acting as a genetic switch?

Meanwhile, on the other side of the Atlantic, plant scientists at the Sainsbury Laboratory in Norwich were also working on a double-stranded problem. One scientist there, David Baulcombe, was interested in how plants defended themselves against attack by viruses, notorious for using double-stranded RNA as their primary genetic material.

Professor Baulcombe's findings interested Dr Fire, who was still trying to solve his own problem. Dr Fire said: "We were still thinking about it as something that was confouding our experiments and it was a cool, odd result. Baulcombe's group had been studying viral resistance in plants and they had come to the conclusion that something about foreign RNA mimicked the virus.

"Putting those two things together led to a fundamental understanding of the issue."

Professor Baulcombe had reanalysed Dr Jorgensen's petunias and other plants and found very small stretches of double-stranded RNA floating around in the cells. The teams of fruit fly scientists raced to find the same stretches of double-stranded fragments in their lab animals. They, too, eventually found small RNAs that were interfering with the action of Drosophila genes.

What seemed to be happening was that the large double-stranded RNA molecules were being chopped down into smaller units of a precise length. These units formed a deadly complex with enzymes, which would identify and chop up the critical "messenger" RNA that acted as the lubricant of protein synthesis by transferring genetic information from the genes inside the cell nucleus to the protein synthetic machinery of the cytoplasm – the region outside the nucleus.

Suddenly everybody realised that it would be possible to make these short, double-stranded RNAs in a test tube and tailor them to target a specific messenger RNA from a particular gene. This would mean that scientists could turn off any gene at will.

A critical question was whether this would also work in the cells of mammals, including those of man. If it did work, the medical potential could be enormous. It would mean that we could turn off genes involved in cancer, genes that allowed viruses to infect cells, genes that were involved in tissue rejection after transplant operations and, of course, genes of viruses that had already managed to infect a healthy cell.

Last month came the first hard evidence that RNA interference affected mammalian cells. Scientists have managed, in the test tube, to make human cells resistant to attack by the polio virus as well as the Aids virus, HIV.

Now the focus is on trying to find ways of introducing these short strands of RNA directly into cells. Biotechnology companies are ploughing millions of dollars into different ways to adopt the technology for medical use. Scientists are already working on ways of treating liver disease by silencing the genes of the hepatitis viruses in mice and of cancer by switching off tumour-inducing genes.

"The whole thing's been exciting really from the word go. It's clear from Jorgensen's earlier work that something really interesting and strange was going on here," Dr Fire said. "I think it's early days but it's really exciting early days."

Richard Jorgensen:

Now an associate professor in the department of plants sciences at the University of Arizona in Tucson. Professor Jorgensen, left, carried out the seminal work on petunias that led to the idea of the "cosuppression" of genes. One variety he produced was named "Cossack dancer", above, by his wife and collaborator. His work was published in 1990 and it made the cover of the journal The Plant Cell but it took eight years for other scientists to realise the significance of his work.

THE MEN BEHIND RNA

Andrew Fire: The scientist at the Carnegie Institution, an affiliate of the Johns Hopkins University in Baltimore, who first coined the phrase RNA interference in a seminal paper published in the journal Nature in 1998. Dr Fire's findings set the scene for a race to develop the technique.

Gordon Carmichael: A researcher in the department of microbiology at the University of Connecticut health centre in Farmington who has followed the work on RNA interference closely. He believes the work is leading to a revolution in the understanding of genes and possible treatments for cancer and viruses.

Phillip Sharp: A Nobel prize-winner at the Massachusetts Institute of Technology in Cambridge who has put his intellectual resources behind RNA interference. He heads a $350m (£250m) brain research institute and has formed a company to develop the technique using venture capital.