Lorne 2009: Silk structure from a social insect

Huge Nephila orb-weaving spiders from Australia’s tropics spin webs strong enough to occasionally trap small birds: milligram for millligram, their golden silk is stronger than steel or Kevlar.

The super-strong Nephila dragline silk that forms the guy ropes for golden orb spiders’ elegant webs owes its strength and resilience to a super-stretchy protein called spidroin.

Dr Tara Sutherland of CSIRO Entomology in Canberra says science has been trying for nearly two decades to synthesise spider silk or synthetic analogues on an industrial scale, with dreams of creating super-lightweight bullet-proof vests and gossamer, pocket-portable parachutes.

In 2006, Chinese molecular geneticists reported that they had re-engineered mulberry silkmoth larvae (Bombyx mori) to produce spidroin instead of classical silk, but silk production remains a laborious manual process.

Researchers have even tried immobilising orb-weaving spiders and drawing silk manually from their abdominal spinnerets. In addition to the challenge of domesticating large arachnids, finding spider wranglers prepared to work in an industrial version of Tolkien’s Mirkwood could be problematic.

Sutherland will tell this week’s Lorne Protein Structure and Function conference that spider silk and silkworm silk proteins are very long macromolecules consisting of highly repeated amino acid sequences.

Attempts to express genes for very large, repetitive proteins in recombinant E. coli bacteria or yeast have yielded defective, truncated proteins.

“An alternative is the materials science approach, which involves creating small, synthetic molecules and trying to spin them in a way that will reproduce the mechanical properties of silk,” she says. “But it has proven difficult to establish the environmental conditions needed to spin them into synthetic silk.”

Sutherland’s CSIRO Entomology group took a completely different approach: it went hunting for insects with less complex silks.

They investigated several other moth and butterfly species before conducting a literature search to identify other silk-producing insects. “We found 24 groups that had independently evolved silks,” she says.

The CSIRO group cloned the genes for these proteins, and began analysing the range of silk protein structures across these taxa. They identified five basic structures, two of which turned out to be made from small, non-repetitive proteins – promising candidates for industrial-scale synthesis in E. coli or yeast.

One came from the Neuroptera – lacewings and damsel flies (the latter’s larvae are known as ant lions). The other was from a very familiar insect family that few would associate with silk production: the Hymenoptera (wasps, bees and ants).

---PB--- Independent evolution

The Hymenoptera are closest to Sutherland’s entomologist’s heart – even though she is allergic to bee stings. “If you look at the nest of a paper wasp (Polistes spp), each cell is lined with silk. Our commercial manager has a paper wasp nest right outside his window, which I would personally find a bit daunting.”

Sutherland says this silk evolved in the common ancestor of bees, wasps and ants, which probably lived around 155 million years ago.

Several Asian relatives of the European honeybee, Apis mellifera, and other Apis species elsewhere in the world also make silk, as do many species of wasps and hornets.

Bumblebees (Bombus spp), which diverged from honeybees 80-odd million years ago, also use silk to line their larval cells – Sutherland had to make a trip to New Zealand to collect bumblebees, which are an introduced species in that country, and in Tasmania.

Larvae of Australia’s primitive bulldog and jumper ants (Myrmecia spp), spin larval cocoons from silk, as does the so-called “dinosaur ant”, Nothomyrmeia macrops.

Spiders secrete silk from spinnerets at the tip of the abdomen. Some groups, like the orb spiders, actually produce several different forms of silk from specialised spinnerets. Hymenopteran larvae emulate the larvae of silk moths, secreting silk from specialised labial glands.

Sutherland says lepidopteran and hymenopteran silks differ markedly in their primary, secondary and tertiary structures – hymenopteran silk has a helical structure consisting of tightly coiled coils.

The divergent nature of the proteins, and the distinctive genomic arrangement of the genes that encode them, indicate that silk production evolved independently in the Hymenoptera.

Her team has identified four honeybee fibre genes, (Amelfibroin 1-4), and two silk-associated genes (AmelSA1 and AmelSA2); the latter code for proteins that appear to serve as glue.

The Amelfibroin genes cluster on a short genomic region. They comprise a single exon each, and appear to have arisen through duplication of an ancestral gene.

The other silk-producing Hymenoptera have homologues of the fibre genes, but they have diverged significantly and have quite low primary sequence identity – around 30 per cent.

Only the larvae of social species of wasps and bees produce silk. Ants are universally social, but not all species produce silk.

Weaver ants, represented in Australia by a solitary genus, the green tree ants (Oecophylla spp) of the tropics, are renowned silk producers – green tree ants use their silk to bind leaves together to form arboreal nests.

Sutherland says a single sawfly genus has also independently evolved silk, apparently quite recently. In common with other hymenopterans, the sawfly larvae secrete the silk from labial glands.

“It’s exciting, because it’s a secreted form of collagen, very different from the classical beta-sheet silks of most other hymenopterans, and the coiled coil silks of the social species,” she says.

“The core of the protein has three-residue periodicity, with a glycine at every third position. The original hymenopteran beta-sheet structure has a two-residue periodicity, with alternating side chains that cross-link the parallel protein fibres into beta sheets.

---PB--- Applications

Isolating labial-gland cells to make cDNA libraries for identifying the genes involved in silk production was not difficult. One bull-ant larvae sufficed, and 1/10th of a bumblebee larva was enough. But the Apis bee larvae “just fell apart” when dissected, she says, and it took 40 larvae to produce a cDNA library.

When silk forms in the labial glands, its constituent proteins initially associate in the form of liquid crystals – a state intermediate between a liquid and a solid.

“The labial gland has to produce a very high concentration of the proteins, so the thread doesn’t undergo capillary breakup and form droplets as it is extruded,” she says. “The crystals slide over each other, greatly reducing viscosity, and allowing the silk to be extruded as very fine threads.

“The insect larva then performs a draw step, dabbing the silk onto a surface and then moving away to get additional draw. The proteins are already aligned, and undergo hydrogen bonding as the tension draws them closer together. As they are spun, they become crystalline and water-insoluble.

“With these low-molecular weight silk proteins, we have found something that we can make transgenically, so we’re now trying to replicate the proteins and make threads.”

But many environmental and other variables are involved in the synthesis and drawing processes. Sutherland says the next challenge is to manipulate the variables to optimise the mechanical properties of the silk for different applications.

She says bee silks have less tensile strength than classical spider and Bombyx silks, but greater extensibility. “If you were developing a synthetic bee silk to make body armour, a bullet might penetrate some distance without breaching the silk armour.”

Other potential applications are high-strength, flexible lines to arrest aircraft landing on aircraft carriers, lightweight textiles and synthetic fibres for medical applications, like synthetic blood vessels.

As for sailing vessels, bee silk’s stretchiness might put it at a disadvantage to Kevlar as a material for mainsails, but it might have some advantages as a material for spinnakers – the “balloon” sails that power yachts in downwind runs.

Sutherland says medical uses are an attractive prospect, because bee silk is non-immunogenic and amenable to creating amino-acid substituted variants with novel properties – for example, to make prostheses with surfaces resistant to colonisation by bacteria.

Bee silk could be modified for sun-resistant high-fashion silks, by removing the tyrosine residues that cause classical Bombyx silk to degrade with extended exposure to ultraviolet light in sunlight.

“We’ve now finished the discovery phase, so we’re now moving into research aimed at producing synthetic bee silks.”