Silkworm genome gets solid coverage
Hapless insect regains place among model organisms.
The most thorough genome sequence data so far for the silkworm, Bombyx mori, have been released by Chinese researchers.
Taken together with sequence data previously released by a Japanese group, the sequence will allow researchers the first comprehensive look into the genetics of a lepidopteran a 160,000 species-strong order including butterflies and moths, which accounts for some 10% of the world's animal biodiversity.
Many of the world's most ravenous plant-eating pests are in this order. Besides possible applications to help the millions of people toiling in the silk industry, agricultural researchers will be scouring B. mori's genome data, published in this week's Science, for clues on how to protect their plants.
The silkworm has been domesticated over the past 5,000 years, during which time it diverged from its wild and rarely seen ancestor, B. mandarina. Baby worms feed on mulberry leaves until they have spun their silken cocoons. Silk-industry workers then bake or steam them to death, after which the silk is removed.
silkworm geneticist at the University of Rhode Island, Kingston
As a model organism, B. mori has been critical to the development of genetics. Big, flightless and easy to handle, the moths were used by researchers in Japan in the early twentieth century to match or jump ahead of developments in the genetics of the fruitfly, Drosophila melanogaster. But since then, despite some 400 mutant lines in research laboratories, B. mori has largely receded as a model organism. The new sequences could restore it.
In January, a group of Japanese researchers released silkworm sequence data in which the genome had been covered 3 times (3X coverage)1. The publication followed a rift with many Chinese researchers who were trained by the Japanese, says Wong. On Friday, these researchers published their 6X data2. The higher the coverage, the greater the accuracy of the sequence.
Early reports indicate that the two sequences, both of which are now available in the online sequence database Genbank, are consistent and will form a 9X sequence when they are combined, says Marian Goldsmith, a silkworm geneticist at the University of Rhode Island, Kingston. "We expect to get a pretty good assembly, and now we have at least one model moth so we can start doing comparisons," she says. "Lepidopterists are thrilled."
Wong and his collaborators say that their draft, which lists 18,510 genes on the worm's 28 chromosomes, accounts for 90.9% of the genome. At 428.7 million base pairs long, the genome is 3.6 times larger than that of the fruitfly and 1.54 times larger than that of the mosquito.
Basic research into representative systems such as olfactory function will proceed quickly, says Goldsmith. In Drosophila, genes for around 150 olfactory receptors, essential for food gathering and other behaviour, are known. Around the same number would be expected in the silkworm, but none are currently known. And though only 45 genes were previously known that are related to the silk gland, the new data shows up 1,874.
Valuable comparisons can also be made across species. Of 323 genes known for wing development in Drosophila, 300 are present in the silkworm, according to the recent data. Despite losing the ability to fly, "these genes are present and very highly conserved", says Wong.
But some of the greatest potential will be in agricultural applications. Knowledge of genes will allow researchers to use biomarkers to select for certain desirable traits, such as fibre quality or disease resistance. But more important might be insights that it provides into how to deal with the gluttonous, destructive larvae of related species, such as the fall armyworm and the tobacco budworm.
Many of these species have grown resistant to pesticides. The genomic data will help to pull out sets of genes involved in the development of resistance and help to find pesticides that can target them, says Goldsmith. "We want to find agents that selectively target that group. Now we have a model with which we can do this," she says.
- Mita K., et al. DNA Res, 11. 27 - 35 (2004).
- Xia Q., et al. Science, 306. 1937 (2004).