Spider Webs and Silks: Tracing Evolution from Molecules to Genes to Phenotypes

  • Catherine L. Craig
Oxford University Press: 2003. 256 pp. $59.95

Spiders are masters of the extended phenotype. The archetypal orb web spider, for example, manages to expand its effective body size by at least an order of magnitude. This remarkable extension is superbly matched to the local conditions, and total daily restoration is often thrown in for free. How does the spider perform this clever trick of morphological inflation? By building its trademark web; by matching and redesigning the structure to suit the changing circumstances of location, weather and prey; by rebuilding it daily; by recycling the silky hardware, so it can devote most of its resources to the software and the 'running costs' of building. Running, or rather walking, is the operative word here, as the web is a frozen record of the spider's path and its manual labour, written in silk.

Silk is the key to the spider's success, with behaviour playing merely a supporting role. Without their silk, spiders would be weaklings among their arthropod peers — if they had survived at all. They are soft-bodied and so are prone to physical damage; they breathe through lungs, and are in constant need of high humidity; and they are wingless, and their hydraulically driven legs buckle when overused. As a result, spiders would have been no match for the tough and virtually indestructible insects. But silk gives spiders a distinct edge, and has become the main weapon in an arms race with their insect relatives, which are both their main prey and major predators.

Mention silk to a polymer chemist and they will get dreamy eyes. It is a natural fibre to match the best man-made ones, but its production is as eco-friendly as it gets, occurring at ambient temperature and near-ambient pressure, with water as the solvent. How is it done? We don't know. Why is it so tough? We don't know. Actually, what is silk, exactly? We don't know. So there's lots of scope for research, then.

This is especially gratifying because state-of-the-art analytical techniques such as X-ray diffraction, Raman spectroscopy and mechanical testing have recently been refined to allow measurements on single silk filaments , which are often less than 2 µm in diameter. Today, single-filament analysis can be done online as the fibres are being spun by the animal; the speed at which the spider reels the silk can be altered over five orders of magnitude by heating and cooling the spider.

Although we have no real answers to even general questions, we have come a long way in our understanding of silks, and Catherine L. Craig's Spider Webs and Silk is an extremely useful staging post. I don't agree with everything that Craig has written, but even so I can highly recommend the book as a comprehensive and up-to-date account of silk.

The four chapters on silk form the core of the book. The first outlines the history of silk evolution, and the second explores the genetic code behind the proteins that make up some silks. Next comes an investigation of how the mechanical properties of bulk silk depend on the protein skeleton of the filament. And it is all wrapped up by an explanation of the economics of silk synthesis and its effect on the evolution of the wide range of silk types that some spiders can produce.

There are also three intriguing chapters on spider webs and their role in feeding the beast, as well as in constraining the more social aspects of its evolution. These chapters are not so much a review as a personal voyage, relying heavily on the author's own work, which focuses on the interaction of insect vision and a web's architecture as well as the silk's structural properties. There is also a chapter on the absence of higher social development in spiders, which, in a twist, is linked to development and related silk-production costs. The brief final chapter summarizes the author's view on the forces that drive silk evolution.

The writing throughout is clear and well presented, and even though there is no glossary, an effort has been made to avoid jargon. There is a good index and the references are, on the whole, comprehensive, although with some curious gaps and biases. Overall, the book provides excellent value for money on a number of levels.

Silk research has undergone something of a renaissance, with the focus of study shifting from the classic (and highly commercial) silkworm silk to the more esoteric spider silk. Silkworms spin through their mouth between their nippers, which means that they can cut the fibre at will. Most silkworm silk is collected from preformed cocoons that the worm has spun at its leisure, and it invariably contains numerous weak points resulting from the worm's typical spinning action. By contrast, spiders spin from their bottoms and cannot interfere with forced silking as long as their legs are kept clear, so spider silk is typically collected as it is spun under highly controlled conditions.

It may be that spinning conditions, rather than silk genes, are a major factor for fibre quality, as silkworm silk that is spun spider-fashion can be almost as tough as spider silk (Z. Shao & F. Vollrath Nature 418, 741; 2002). Using spider silks produced under controlled conditions allows the construction and testing of hypothetical links between protein form and folding. In this way, spider silks provide a new way of studying fibrous proteins. For example, recent studies have shown that spider silk shares important protein-folding configurations with amyloids and prions (J. M. Kenney et al. Eur. J. Biochem. 269, 4159–4163; 2002), which suggests that unravelling the spider's way of making a tough fibre might have surprising results.

Craig's book has the somewhat puzzling subtitle “Tracing evolution from molecules to genes to phenotypes”. Most authors would have gone from genes to phenotypes without bothering with the molecular level. But not Craig, who states that one of her goals is “to illustrate the ease with which evolutionary studies of spiders and silk proteins can cross traditional molecular and organismal borders”. Thus silk proteins are “the perfect system through which to study evolutionary conflicts among molecular genetic constraint, protein architectural constraint, protein diversity and selection”. Indeed, silk has evolved to function as a dead material outside the animal's body. And silk is only one step (spinning) removed from the protein in its native form; the chain from gene to raw protein to dehydrated protein fibre is brief.

But more importantly, the fibre, once drawn — whether as a simple drag safety line or as a structural member in a web — is the phenotype that is selected more or less on its own, distinct from its creator. This is unusual for a protein and suggests that silk might allow for shortcuts in the study of protein evolution — yet another reason why studying silk might provide interesting insights into the more general aspects of protein folding.

Be that as it may, Craig's Spider Webs and Silks brings a fascinating and important subject to a potentially broad audience. And it might even turn some arachnophobes into arachnophiles.