Scientists have identified a protein that plays a key role in generating the sweet scent of some roses.

The researchers in France have found that a protein named Nudix hydrolase serves as an enzyme in the biochemical pathway for the synthesis of monoterpenes, compounds that contribute to about 70 per cent of the fragrance content of some roses.

"Roses in which the gene for Nudix hydrolase is active smell nice, and the flowers in which this gene is inactive have lost their fragrance," Philippe Hugueney, a plant biochemist at the University of Strasbourg and a member of the research team, told The Telegraph.

Their findings, described in the US journal Science yesterday, help explain why most rose varieties used across the world in the ornamental flower industry have lost the unique fragrance associated with these iconic flowers.

As flower growers developed varieties of roses that were visually attractive through their colours and shapes, Hugueney said, many lost the activity of the Nudix hydrolase gene.

Scientists say the fragrance emerges from hundreds of compounds whose levels vary across different varieties of roses.

Although earlier studies have identified several genes that contribute to the production of rose fragrance, the Nudix hydrolase pathway to monoterpenes was unknown until now.

Dozens of beautiful rose varieties that serve as ornamental flowers lack the characteristic smell of roses .

The scientists in France who studied a rose variety named Papa Meilland have found that the Nudix hydrolase enzyme through a cascade of biochemical events helps produce a fragrant component of rose oil called monoterpene geraniol.

While Nudix has previously been described in other plants, this is the first time it has been identified and linked to the scent of roses, Sylvie Baudino, a plant molecular biologist at the University of Lyon Saint-Etienne, said.

In a commentary in the same issue of the journal, two plant biologists not associated with the study have called the French findings "an unexpected enzymatic pathway to these scent compounds".

Dorothea Tholl at the Virginia Polytechnic Institute in the US and Jonathan Gershenzon at the Max Planck Institute for Chemical Ecology in Germany said the discovery establishes an alternative route that rose flowers use to produce monoterpene geraniol.

The discovery, they wrote, provides a "reliable molecular marker for a major group of floral scent compounds in roses that could be exploited to enhance fragrance in these iconic flowers".

Hugueney said flower breeders seeking ornamental roses would need to ensure that the gene for Nudix hydrolase remains active throughout the breeding process.

"We can now look forward to a new varieties of roses that look good, have nice shapes, and smell good too," Hugueney said.