They seek it here, they seek it there, but the Scarlet Pimpernel molecule that spreads RNA-induced gene-silencing through plants, and confers systemic protection against virus infections, remains elusive.

In a series of elegant experiments with grafted Arabidopsis seedlings, CSIRO Plant Industry molecular geneticist Peter Waterhouse and Dr Bernie Carroll of the Department of Biochemistry at the University of Queensland have identified how the message is transmitted, and determined that it is probably not a small RNA molecule.

The message may yet turn out to be a larger RNA molecule, but it appears to work at the level of the gene, via epigenetic silencing rather than the now-familiar machinery of RNA-induced silencing complexes (RISCs).

A decade ago, Waterhouse and CSIRO colleague Dr Ming-Bo Wang performed a seminal experiment in tobacco which established that plants have a cell-based defence against viral infections.

Carroll developed a system for micrografting Arabidopsis, and then began a series of collaborative experiments with various rootstock-graft combinations.

Their observations not only illuminate the 'spreading silence' phenomenon, but may have explained how the practice of tissue-culturing meristem tissues from perennial plants like grapevines and citrus can eliminate persistent viral infections, restoring them to full genetic health and productivity.

Waterhouse suspects some perennial plants have exploited a quirk of RNAi, exclusive to plants, to achieve extraordinary longevity. These near-immortal plants have survived for thousands, even tens of thousands of years, untroubled by virus infections or epigenetic reprogramming that could disrupt their fitness (see page 4).

Viral protection

RNAi continues to transform the biological sciences - it is undoubtedly the most powerful new tool for exploring and manipulating plant genomes in three decades.

But it seems there is nothing new under the sun - researchers first observed and harnessed its potential to protect crop plants against viral infections more than 75 years ago.

In 1929, pioneering plant pathologist Harold McKinney, working on a US Department of Agriculture farm that later became the site of the Pentagon, began exploring the enigmatic phenomenon of cross-protection.

McKinney infected tobacco plants with a mild strain of tobacco mosaic virus (TMV), and found that, by some act of green magic, they became fully resistant to pathogenic TMV strains. Agronomists took up the technique, using it to protect crops like tobacco, citrus, cucurbits, grapevines and pawpaws against viral infections.

Local inoculation with a virus somehow protected the entire plant. Many hypotheses were advanced, but the mechanism remained an enigma until Waterhouse and Wang's seminal experiment in tobacco in Canberra in 1997.

By 1993, the CSIRO researchers were convinced that the release of the double-stranded RNA (dsRNA) genome of an invading virus triggers a cell-based defensive mechanism that disrupts the virus' replication.

In 1986, US plant virologist Dr Roger Beachy had made tobacco plants resistant to tobacco mosaic virus (TMV) with a transgene coding for the TMV capsid protein. The prevailing idea was that over-expressing coat proteins blocked infection by disrupting the assembly of new virions.

Waterhouse and Wang experimented with Beachy's technique, but like many others, obtained inconsistent results.

Sometimes, transgenic plants exhibited resistance even though they expressed little or no coat protein from the target virus. A vital clue emerged in 1992, when US plant virologists Bill Dougherty and John Lindo showed that the viral RNA was being degraded before it could be translated into protein.

The anti-viral defence was triggered, not by viral proteins, but by the virus's double-stranded RNA genome.

In 1997, Waterhouse and Wang developed two transgenic tobacco lines, each containing a transgene coding for one strand of an RNA sequence from Potato Virus Y (PVY). The parents were susceptible to TRV, but when they were crossed, 25 per cent of their F1 progeny were fully resistant.

The resistant plants had inherited both transgenes. The messenger RNAs had base-paired, forming a double-stranded RNA (dsRNA) molecule. Somehow, the dsRNA molecule directed the degradation of the corresponding sequence in the virus itself, when the plants were inoculated.