What Skill is in the frame of Insects shown?
How fine the Threds, in their small Textures spun?
RICHARD LEIGH, ‘Greatness in Little ’
The astounding properties of spider silk have been recognized for decades. In the force needed to break it when pulled, spider silk is about half as strong as mild steel, so the oft-quoted ‘spider silk is stronger than steel’ is not strictly true. Steel, however, is nearly eight times denser than spider silk so weight-for-weight spider silk is about six times as strong as steel. Spider silk is much more stretchy than steel, extending by 30–40% before it breaks; it is about twice as stretchy as nylon and eight times more stretchy than Kevlar®. What is special about spider silk is that it is both stretchy and tough: a rubber band will stretch more than spider silk but its breaking strength is very low. Spider silk is the only material with exceptional stretchiness and good breaking strength.
Spider silk has been brought to a pitch of perfection by millions of years of evolution. And this optimization means that there isn’t just one generic spider silk: a single spider can make up to seven different kinds of silk, each tailored towards a specific task: the dragline from which the web is hung is the strongest, the capture threads have the greatest extensibility, and so on. Spider silk’s great resilience has long suggested human applications. The web has to catch a heavy insect at speed, and bring it to a standstill without snapping and without flinging it back out again in recoil, a process reminiscent of the arrester wires used to bring jets landing on aircraft carriers to a halt.
Spiders have been working their magic for over 400 million years – that’s pre-dinosaur time. The oldest existing strand of spider silk was reported in 2003, preserved in Lebanese amber. It dates from the Early Cretaceous Period, more than 120 million years ago and what is fascinating about this specimen is that the small globules of ‘glue’ that are strung along the capture threads are still clearly visible, as they are on spider webs today.
We think of spider webs as delicate filigree structures, best seen with dew or frost accentuating their patterns. The garden spider (Araneus diadematus) is one of the best web spinners (fig. 3.1). But tropical spider webs can be very large: the queen of spinners is the golden orb-weaving spider (Nephila claviceps), which can be 5–8 cm long and 20 cm in total span: her webs are up to 2 m in diameter – big enough in fact to be useful economically. In Papua New Guinea, they have been draped across bamboo poles and looped at the end to make fishing nets. Early Western explorers also encountered such webs: in 1725, Sir Hans Sloane reported the nets were “so strong as to give a man inveigled in them trouble for some time with their viscid, sticking quality”.
For those who fear spiders, a web large enough to enmesh a person is the stuff of nightmares. The fear of spiders is a widespread cultural phenomenon, and my interest in the silk made me question my own attitude to spiders. Primo Levi, who is always a good guide to our reactions to the natural world, summed up the symbolism of spiders like this:
The old cobwebs in cellars and attics are heavy with symbolic significance: they are the banners of desertion, absence, decay and oblivion. They veil human works, envelop them as though in a shroud, dead as the hands which through years and centuries built them.
Other People’s Trades
There is a constellation of factors that creates a general sense of unease. There are very few large, hairy poisonous spiders (the tarantula, despite the legend, is not much more poisonous to a human being than a wasp), but all spiders, by association, share in a little of the horror that these monsters can conjure up. The spider’s snaring and ambushing techniques worry some people. Something as deep-seated as spider phobia most likely has sexual connotations: the fact that the female sometimes consumes the male after mating suggests that men can associate them with women who symbolically castrate, if not devour. But women in particular are sufferers from arachnophobia. One attribute of spiders that would cause unease if it was generally known would be the fact that they have eight eyes. But most people have never seen them because they are only visible through a microscope.
Before I became seriously interested in spider silk, I had realized that not only is the web of the garden spider very beautiful but the creatures themselves, with their light speckled colouring, are much the most comely spiders you are likely to come across unless you become a dedicated arachnologist. Once I learned about the silk, my conversion to spider-worship was complete and I became ashamed of my earlier hostility.
The first documented attempt to exploit spider silk was by a Frenchman: in 1709, Xavier Saint-Hilaire Bon made gloves and stockings from the silk and presented them to Louis XIV. He wrote ‘A Dissertation on the usefulness of spider silk’. The scientist René Réaumur investigated these claims in 1710 and concluded that only egg cocoon silk is good enough for spinning but it lacks lustre (a surprising finding given that all modern research focuses on the dragline silk from which the web is hung). Réaumur estimated that it would need 27,468 female garden spiders to make 1 lb of silk. Despite this discouraging report, the Chinese Emperor requested a copy of his paper and Chinese silk experts attempted to exploit spider silk. In 1876, the Chinese Emperor gave Queen Victoria a spider-silk gown. Whether this had been a century and a half in the making, since their initial interest, we do not know. Despite the impression we have of the evanescence of spider webs, the silk is durable. In Austria in the late 18th century, there was a tradition of painting on spider webs and some of these pictures exist to this day; one of them hangs in Chester Cathedral.
There is no great problem in spinning silk from a single spider. You can just about do it yourself with some improvised kit and patience. First, wait for the webs made by the garden spider, which appear in late July, then catch a spider. The spider has to be restrained, obviously, and a Styrofoam block makes a handy mount. The spider must gently be turned onto its back and a couple of very light rubber bands used to pin four legs on each side close to the body. Once the spider is stable you can start a thread by lightly touching the spinneret under the belly with a glass rod and pulling gently.
If you want to create a reel of silk, the Styrofoam block can be mounted on a piece of wood, with an improvised reel at the other end. This could be a cotton reel on a spindle with a handle. If you happen to know anything about hand-spinning, you could collect a few bobbins of spider silk and try to braid them into a thicker thread that would bring it closer to the dimensions of usable textiles. Whatever you do with the silk, the last stage is to let the spider go, when it will instantly return to work, repairing its web.
Although you cannot see any of this without a microscope, it is worth knowing that a garden spider has three pairs of spinnerets, each with multiple spinning tubes – more than 600 in all (fig. 3.2). Without an explanation, a picture of this apparatus might appear to be some kind of technical glue nozzle system.
The industry of garden spiders is prodigious. When I brought one in to be ‘silked’, its web was damaged in the process, but a few hours after releasing the spider the web had been rebuilt. Spiders keep their webs in good repair. After a few days, a web will become tatty from insect collisions, wind, dust and the spider’s own movements across her domain. Every two or three days, the spider will consume the old web and build a new one, usually in exactly the same place, since they are highly territorial. Around 80–90% of a new web is protein recycled from the old one. This means that a spider catches food mainly to get the energy to build the web; it doesn’t need food to supply much of the material – an example of the amazing efficiency of living processes.
Attempts have been made to silk spiders on an industrial scale. Properly set up, a single golden orb-weaver can produce 300 metres of silk in one session. The problem is that spiders cannot be farmed intensively. They are aggressive, solitary creatures who, if confined in one space, eat each other.
This naturally turns the mind towards the idea of making a synthetic silk. That this might be possible was suggested as far back as 1665 by Robert Hooke:
Probably there might be a way found out, to make an artificial glutinous composition, much resembling, if not full as good, nay better, than that excrement, or whatever other substance it be, out of which the silkworm wire-draws his clew. If such a composition were found, it were certainly an easy matter to find very quick ways of drawing it out into small wires for use. I need not mention the use of such an invention.
For centuries the only way of making silk was with the silkworm. Archaeological evidence has shown this to be an ancient craft, going back to around 2600 BC in China. The silk moth is the only domesticated insect, having lost the power of flight, all pigmentation, and just about any desire to move or to do anything. The silk is produced by the caterpillar to cocoon the chrysalis and for this reason is not as strong as spider dragline silk. But it is a natural product that no synthetic has ever been able to match, although Japanese textile technologists have now come very close.
The basic process is as follows. The eggs are hatched and the caterpillars fed on mulberry leaves. They moult four times before they are ready to spin a cocoon in which the chrysalis will develop. The chrysalises within the cocoons are then killed by steam or fumigation. The cocoon silk consists of two filaments of the silk protein fibroin stuck together by another protein, sericin. To process the silk, the sericin is removed with hot water and the filaments drawn from water and combined to make yarn. The yarn undergoes stretching and is wound onto reels as raw silk.
Because of the finicky nature of the silkworms and the demanding cultivation regime, increasing the production of natural silk is not easy, and silk production has often been threatened by disease. In 1855, silkworms, particularly those in Europe, were afflicted by a parasitic disease called pébrine. This episode is the centrepiece of Alessandro Baricco’s novel Silk, which captures in delicate prose the aura we associate with the fabric:
He felt the lightness of a silken veil dropping onto him. And the hands of a woman – of a woman – drying him all over, caressing his skin; those hands and that material spun out of nothing. He never stirred, not even when he felt the hands move from his shoulders to his neck and the fingers –the silk and the fingers – climb to his lips and brush them once, slowly, then vanish.
Pasteur was called in to solve the pébrine crisis but progress was slow and this seriously focused minds on the possibility of imitating the natural process. At the time, knowledge of the chemistry of silk and all such natural substances was non-existent. Because the caterpillars grew on a diet of the leaves of the white mulberry, Count Hilaire de Chardonnet, who had worked with Pasteur on pébrine, tried ways of by-passing the silkworm by digesting mulberry leaves and creating a solution that could be squeezed through a nozzle similar to the silkworm’s spinnerets. In fact, the main component of leaves is cellulose, a material very different to silk proteins but also a long-chain molecule. Amazingly, it did prove possible to create silk-like substances from cellulose by several processes, the best-known being rayon (1891).
The potential of silks in one of the toughest applications imaginable was realized in the late 19th century by a physician in Tombstone, Arizona: ‘In the spring of 1881 I was a few feet distant from a couple of individuals who were quarrelling,’ George Emery Goodfellow wrote in his diary. ‘They began shooting.’ Two bullets pierced the breast of one gunman, who expired from his wounds. But, on examining the body, Goodfellow found that, ‘not a drop of blood had come from either of the two wounds’. He noted that ‘from the wound in the breast a silk handkerchief protruded’. When he tugged on the handkerchief, it came out with a bullet wrapped inside. Evidently, the bullet had torn through the man’s clothes, flesh and bones but had failed to pierce his silk handkerchief. Intrigued by this discovery, Goodfellow began to document other cases of silk garments halting projectiles – including one incident in which a silk bandanna tied around a man’s neck kept a bullet from severing his carotid artery.
If silk was ever going to be used seriously for such applications it needed to be made in quantity. The mimicking of natural silks on a commercial scale began with the invention of nylon in 1937. Nylon is derived not from plant products but from very small chemical units, linked together to form long-chain molecules. Such compounds, now ubiquitous in modern civilization, are called polymers. In nylon, the link – the amide group – was the same as that in natural silks although the rest of the molecule was very different. Nylon has a much more regular structure than natural silks.
The first serious flak-jacket silk was kevlar, a tougher variant of nylon, invented in 1963. Even with nylon, kevlar and other fibres established as industrial staples, the superior properties of spider silk were alluring, but no bulk industrial or military use was proposed until very recently. The first serious modern application was very small scale. In the Second World War, single fibres of spider silk were used as cross-hairs for accurate range-finders – it came from black widows in the USA, garden spiders in the UK. Pioneer spider-silk researcher David Knight tells the story of the major US chemical company Du Pont, inventors of nylon and kevlar, who supplied a spider-silk sample to the US Army during the war, hoping for an order. Three years later, they politely enquired about the silk and asked whether the Army would be making an order. ‘Oh, we don’t need any more,’ they were told, ‘what you sent was fine.’
The picture changed dramatically, at least in prospect, with the arrival of genetic modification (GM) technologies in the late 1970s. In GM, a gene can be inserted into a foreign organism; the organism will function normally and produce the proteins programmed by that gene. So, in theory, if you took the gene for spider silk, and inserted it into an animal, you could make industrial quantities of silk.
Work began on this project in the 1980s and was bedevilled by nature’s cussedness. Spider-silk genes turned out to be harder to handle than the insulin gene, GM’s first great success. But, in June 2002, Nexia Biotechnologies in Quebec, Canada, claimed that they were able to produce industrial quantities of spider silk from the milk of genetically engineered goats. The story had a strange blend of hard military exploitation and New Age greenery. On the one hand, the US Army had been working on spider silk for many years; Nexia’s silk, named BioSteel®, was developed under an Army contract for flak jackets and one of the two herds of modified goats was kept on a former United States Air Force B52 bomber base at Plattburgh, New York State. On the other hand, Nexia’s President and CEO, Jeffrey Turner, waxed lyrical about this new fibre produced from meadows, goats, sun and water and spun at room temperature from a watery solution. Nylon and kevlar, the closest things we have to spider silk, are made using toxic chemicals and high temperatures and they generate toxic wastes. Turner said: ‘We use water and hay; to make nylon – which has a half-life of 5,000 years [which means it’s not biodegradable] – you have to sink a hole in the ground. That’s not the kind of world I want to leave my kids.’
If we could manufacture large quantities of spider silk and spin it the way the spider does we would have a very special material. But 16 months on from the excited press reports of June 2002 the spider-silk story looked very different. The US Army withdrew from its collaboration with Nexia because BioSteel, as it then was, could not meet their requirements for quality or quantity.
By mid-2004, Biosteel had been downgraded even further. Development of spinning for general yarn and fabric was suspended due to the ‘ongoing technical challenges of producing bulk, cost-competitive spider-silk fabrics with superior mechanical properties’. On 8 March 2005 this particular strand of the spider-silk story was fractured. Nexia’s principal asset Protexia® was taken over by an american company, PharmAthene, and CEO Jeffrey Turner resigned. BioSteel® remained as the rump of a much reduced operation. So what had gone wrong? How does the spider do it and why is it so hard to emulate?
Like wool and silkworm silk, spider silk is a protein. Although it is DNA that carries the instructions for making all living things, as materials scientist Mehmet Sarikaya says: ‘It is the proteins that are the workhorses in organisms. In the human body there are hundreds of thousands of different proteins; they all work at the same time in a concerted manner. You look, you eat, you think because of the interactions of these proteins.’
One of the curious things about DNA is that although it is responsible for you being a human being rather than a spider, the physical molecule is in many respects the same wherever it comes from: the arrangements of the four bases (adenine, cytosine, guanine, thymine) that make up the chain create their infinite patterns that carry the code of life, but because the bases match on complementary strands – a thymine always opposite an adenine, a cytosine always opposite a guanine – the helical structure is always the same. When, in 1953, Watson and Crick deduced its structure it did not matter where the DNA came from, its X-ray crystal pattern would always have been the same. This property is essential to its function. DNA has to be as neutral and unreactive as possible: it carries the code and must not get tangled up in extraneous reactions.
Proteins, by contrast, have many varied zones of attraction and repulsion on their surfaces and they get tangled up in all kinds of ways. Protein structure is very complicated, but the long and the short is that proteins can make almost any shape you like. Many are water-soluble globular masses, like egg white, some are water insoluble and fibrous or horny, like hair or nails – or spider silk. You get some insight into protein properties every time you boil an egg. Egg white is largely made of the protein albumen. When you crack an egg, you can see that the white is clearly some kind of very thick solution. The resistance of uncooked egg white to flowing suggests that, despite its transparency, it has some filamentous structure inside it. And indeed it does. But the filaments are strongly water-loving, at least in places, and these create the familiar jelly. When you heat the egg, water is expelled from the filaments and they start to curl up. Molecules from adjacent chains meet and form cross-links. The filaments quickly tangle up into a ball and the jelly is replaced by a rubbery solid.
Although we think of food proteins and structural proteins, such as hair and wool, as radically different substances, chemically they are similar. And in fact you can turn one into the other. Milk is a foodstuff but the protein it contains, casein, can be processed to make a useful plastic. Before the advent of synthetic plastics, milk plastic was used to make small objects such as buttons.
A crude milk plastic is easy to make in the kitchen. Add a small cup of white wine vinegar to half a pint of milk in a saucepan (quantities are not crucial) and warm it, stirring constantly. Flakes of a white solid separate out. Don’t boil but stir the mixture for five minutes or so over a moderate heat. Then pass through a coffee filter and wash with water to remove the vinegar. Dry the solid on kitchen roll.
The resulting product is soft and can be kneaded into a ball, or cast into shapes. As it dries, you will see how much water is trapped in the structure. The result is rather foamy, not very strong and the surface is greasy from the entrapped milk fats. To make a milk plastic for serious use, the water content needs to be controlled to allow it to escape without creating voids, the fats must be removed and the plastic hardened. Proteinaceous materials require very subtle processing to attain the right properties, but this simple experiment demonstrates that an industrial material can be made from an unlikely biological molecule. And if you can make buttons from cow’s milk why not flak jackets from the spider-silk proteins in a GM goat’s milk?
When still in the glands of the spider, spider silk is in jelly form and saturated with water molecules. But, like egg white, it nevertheless has a filamentous structure – in fact it is a liquid crystal. There are many kinds of liquid crystal, such as those in digital displays on watches, computers and the like, but the principle is straightforward. Although the substance in question behaves like a thick, sticky liquid, its molecules are all lined up in one direction. This affects the way the substance transmits and reflects light. In the liquid crystals used in display, a small voltage makes the molecules flip into a different orientation, thus dramatically changing their optical properties. The display consists of thousands of liquid crystal cells, appearing brighter or darker depending on the voltage applied to them, thus creating the contrast necessary for character formation. This is why the picture on a liquid crystal display disappears if you change the angle of view: the ‘picture’ is entirely due to the angle of reflected light.
In the case of spider silk, the liquid in the gland, known as ‘dope’,* has domains of strongly orientated molecules that act as liquid crystal zones. As water is removed and the acidity changes in the spinneret, the molecules become increasingly orientated, until they are aligned with the direction of flow. What emerges is a filament with the molecules strongly aligned. Pictures from the electron microscope show that the filament has a structure, rather like the large cables that hang suspension bridges: a core of bundled filaments is surrounded by a sheath made from a different protein.
Spiders spin their silks from watery solutions and it is this that so impresses commercial fibre spinners, used to working with harsh solvents. The question of proteins and water is complicated but, remembering the Lotus-Effect, it revolves around the two principles of water attraction and repulsion. Let’s call them here by their technical names – hydrophilicity (water-attracting) and hydrophobicity (water-repelling). Some of the amino acids of proteins are hydrophilic, some are hydrophobic. The way these are arranged along the chain affects the way the molecule folds up and how it behaves with water. Hydrophobic regions have a habit of doubling back on themselves to form a rigid crystalline pleated sheet folded like an accordion – the so-called beta sheet. The most plausible model for the structure of spider silk contains a mixture of hydrophilic portions (alpha strands) and beta-sheet crystals (fig. 3.3). This could account for the strength because it makes spider silk a composite.
A composite is a material that blends two substances with complementary properties. Fibreglass is the classic composite: glass is strong but brittle; resin is weak but stretchy. If a mat of glass has resin set around it the result is tougher and more resilient than either of the individual components. The thing about composites is that they don’t crack like monolithic substances – what makes things snap is the formation of cracks.*
In a nutshell, the stress at the tip of a crack is much greater than in the rest of the material: this means that, once started, a crack is likely to travel further. In fact, it is very difficult to stop cracks once they start and a whole history of industrial disasters stems from this fact, including the Comet air crashes in the 1950s and the Hatfield rail crash in 2000. At Hatfield, minute cracks had formed on a bend where the wheel flanges of high-speed trains had been forced against the rail – something known as gauge corner cracking. Short of replacing cracked rails, the remedy is to grind down the rail until the crack disappears. So long as any cracks remain, however tiny, there is a danger. Since the Space Shuttle Columbia’s disaster of February 2003, the Shuttle’s wings have been shown to be vulnerable to tiny cracks which can lead to catastrophic failure.
If you take a piece of fabric and give it a tug, usually nothing happens but if you make a small nick with scissors and then tug, the thing tears apart with a satisfying rip. In fact, the ripping noise signals that the failure of the material is catastrophic. This is positive feedback: the force increases as the tear lengthens and the tear lengthens as the force increases. The rails at Hatfield shattered into thousands of fragments for the same reason.
In spider silk, the brittle and the elastic components are different parts of the same molecule and this probably accounts for its outstanding toughness. A single molecule obviously has a greater structural integrity than two substances that are only physically mixed and not chemically bonded. A crystalline region in spider silk plays the role of the glass and a more elastic region plays the role of the resin.
It is one thing to recognize the superior properties of spider silk and quite another to know how to copy it. The silk exists as a fluid inside the spider and only becomes a solid filament when it emerges from the spinneret. So does the spider’s secret lie in the composition of the fluid or in what happens in the spinneret?
On the face of it, it would seem most likely that it is the composition and some scientists do believe that this is the key. Clearly, this has to be significant: a material for a demanding application cannot be made out of any old stuff. But textile spinners know very well the importance of the spinning process. Nylon was not immediately recognized as a superior fibre because when it was first made in 1933 it could not be spun. It was only discovered four years later, during work on another fibre (the first polyesters), that once nylon was formed it could be toughened immensely by stretching the cold fibre. A mechanical drawing-out process within the spinneret achieves something similar for spider silk.
Thanks to genetic engineering technology, the chemical composition of spider silk is far better understood than the processes in the spinneret, and this may have accounted for Nexia’s over-optimism. But even at the level of the chemistry, the spider keeps some secrets. The spider-silk protein is a very large molecule and because it does not have the precise molecular function that, say, the proteins insulin and haemoglobin do, its structure is much looser. There are many repetitive sequences and quite how precisely these have to be copied no one knows.
With a large gene such as that for spider silk, trying to make the gene work properly in a foreign organism is hard to achieve accurately: the process tends to stop short of the full chain, becoming scrambled and confused. David Kaplan, at Tufts University, near Boston, one of the researchers who has worked with spider silk from the start, explains the problems: ‘Whenever you move the gene out into E. coli or yeast or mammalian cells, you’re losing something. We spent years trying to clone the dragline silk and every time we put it into E. coli it would truncate down to 2.5 kD.’* Steve Arcidiacono at the US Army Soldier Center, Natick, Massachusetts, found the same thing. Steve showed me a 2.5 inch thread of GM spider silk, from E. coli, the first they produced – on 27 March 1998: ‘It was more than we expected, but then again so was everything we accomplished given all the difficulties encountered along the way.’
Although much larger than those early synthetic spider-silk molecules, the silk molecules produced by Nexia were not full size – the complete spider-silk sequence remains hidden. In the face of the difficulties encountered by Nexia there are two main schools of thought about the way forward: one suggests that the complete protein sequence is the key – get that right and you’ve got it. This approach has largely been followed by the US Army team at Natick, Randy Lewis’s laboratory at the University of Wyoming, and Nexia. The other approach does not deny the importance of the protein composition but stresses the chemical and physical changes that occur during the spinning process.
At Wyoming, Randy Lewis is continuing the quest for the total spider-silk structure. In October 2003 he said: ‘I certainly have to believe we can completely characterize spider silks. They are just proteins in a solid form and have to conform to the features of proteins. The fact that they are in the solid state makes things more difficult but not impossible.’
David Knight, who left Oxford University’s Zoology Department to found Spinox, a company dedicated to producing technical silks, has focused on the spinning process and has patented an apparatus that mimics some of the processes that occur in the spider’s spinneret. David Kaplan, at Tufts University, Boston, USA, has done a lot of work on varying the conditions of silk gels before they are spun and has also produced films and sponges of reconstituted silks.
Despite all the problems, industrial spider silk is a great prize and there is a race to bring it to market. David to Nexia’s Goliath is Knight’s company Spinox. David Knight was an Oxford University researcher until 2003, one of the world’s leading experts on spider silk, with his colleague Professor Fritz Vollrath. But, with help from Oxford University’s technology transfer unit, he decided it was time to jump from the cloistered academic world into the ‘nasty world out there’ of venture capitalism, patents and competition; hence a tiny two-room unit on an industrial estate in Berkshire.
Not just any old industrial estate. The spider-silk story seems to be ineluctably bound up with matters military. Spinox is located at the former Greenham Common Airbase. Greenham Common was the site of a US Air Force Cruise Missile squadron in the early 1980s. It became famous for the Women’s Peace Camp set up outside the base. The base was closed in 1992 following the collapse of the Soviet Union. Remarkably, the sequel is benign: the greening of Greenham Common. Although the runway still lies sinister and deserted, surrounded by barbed wire, and several of the large hangars and other buildings are still intact, the base is now a business park. Some of the units are new designer structures but many of the single-storey airbase buildings are now havens for alternative businesses.
David Knight came here in July 2003 to begin a revolution in tough fibres. Knight has a personal twist on the transformation of Greenham Common: he was last here as a member of Cruise Watch, the anti-missile protest group. He admits: ‘I am a Green, or partly Green. The Green aspect [of spider silk] interests me and motivates me but we have to keep very quiet about it, because people might think that we’re a bunch of Green lefties who are not really serious. We don’t push it because we’ve realized that it’s the wrong thing to say.’ This highlights one of the contradictions of bio-inspiration: in the first place, the idea of spider silk has to trap a venture capitalist in its sticky threads.
So, while Nexia was burning through Canadian $2.5 million a quarter in a hi-tech gamble on genetically engineered spider silk, David Knight was pitching for a £142,000 grant from the UK Government’s Department of Trade and Industry to perfect his patented spinning process.
When I met David Knight in October 2003, he was sceptical of Nexia’s approach: ‘Think of the cost of getting the stuff out of the milk…they based it on the cost of ordinary milk, but this is a very valuable genetically engineered herd of goats…They wanted us to test their material and they said: “We’d like to supply it in powder form.” We’d have to use very harsh solvents to get that protein back into solution.’
David Knight’s approach is different. He has worked a good deal with spiders and there are many of them in his laboratory; there are also plenty of other creatures. He sees himself in the business of spinning silk from ‘amphiphilic polymers’ (that is, polymers with a blend of water-attracting and -repelling properties) from whatever source.
Knight’s contribution has been in his understanding of what happens inside the spider’s spinneret, with water being extracted, ions exchanged, and the molecules lined up – for him, this is the key to the production of silk. He and his Oxford colleague Fritz Vollrath have patented a spinning nozzle that can handle many different solutions (fig. 3.4). It also allows for the vital ion exchange to occur while the unspun silk is still in the nozzle. In an interesting parallel with synthetic fibre spinning, the contents of a spider’s spinneret are acidified as the solution passes through – and many kinds of synthetic fibres are spun from an acid bath. Fortunately, the engineering of spinning nozzles comes more naturally to David Knight than it would to some biologists: his family were carpenters and he enjoys benchtop engineering.
Knight believes that, rather than flak jackets, the first markets for spider silk are likely to be biomedical, especially fibres for closing wounds and other surgical aids: ‘It’s particularly appropriate because the mark-up is high, it’s a market with rapid adoption of new technology, and they’re eager for new product.’ So how would Knight avoid the trap Nexia fell into, with BioSteel requiring time-consuming regulatory approval before it could be used in humans? He told me: ‘We have discovered another silk, which is a waste product, not cocoon silk. This other silk is from a totally different source – I only realized it was a silk in the summer. This material is completely free because it is being thrown out as a by-product. I can’t tell you any more but it has huge potential.’ I asked Knight whether he wouldn’t still have to go through the regulatory hoops, and he replied: ‘I can’t tell you any more but the silk we’re looking at is human.’
All the main players in spider silk became secretive at some stage in our conversations and to carry the story forward we must speculate. The main source of human fibrous protein that is thrown away in large quantities is hair. Hair is extruded slowly from the follicles, not spun rapidly like spider silk, but the structure of hair, composed of the protein keratin, is similar to silks. Silk researchers know a lot about reconstituting silks.
In 2000, Protein Polymer Technologies, Inc., a San Diego firm, took out a patent on a process for making various protein polymers, such as keratin and wool, soluble. If a good solution of keratin could be produced, Knight’s patented spinner might just be the job for turning it into something special.
But hair is not the only human protein that is thrown away. While some of the blood in blood banks is used whole, many specialized blood products are derived from the rest and the inevitable waste is discarded. One blood product is the protein fibrin that turns blood into a solid scab; it is produced when fibrinogen reacts with the coagulation factor thrombin. Fibrin is already used surgically as a tissue glue and the new knowledge of protein spinning could be applied to such a protein, perhaps producing a tough fibre or film.
David Knight’s final words at our interview were: ‘Maybe Spinox won’t succeed but somebody eventually is going to succeed with the idea of copying natural silks.’ Despite Knight’s affirmation, in October 2003 the spider-silk story looked none too rosy. On 31 October I learned from Steve Arcidiacono that the US Army’s collaboration with Nexia had ended. Then I saw Nexia’s Corporate Statement announcing their setbacks. And so, for the time being, David Knight was pursuing his mysterious human source of silk.
Nexia’s problems seemed to prompt a revisionist movement amongst some spider-silk specialists. At the Materials Research Society Conference in San Francisco in April 2004, a leading forum for bio-inspired materials scientists, Christopher Viney, from the University of California, Merced, suggested that the conventional laboratory tests for strength, stiffness and toughness do not represent how spider silk behaves in the real world. One reason is that although spider silk dries and hardens when it emerges from the spinneret, it retains an affinity for water: when wet it can shrink by up to half its length and become very elastic, stretching when pulled by up to 300% against the usual 30–40%. As a result, he concluded: ‘We cannot envisage natural silk serving as a long-term load-bearing material without modification. Natural silk will not rival steel as a means of suspending the deck of the Golden Gate Bridge.’
Viney has a neat way of demonstrating the effect of moisture on spider silk that you can try at home. If you surprise a spider high up it can quickly abseil to the ground on a thread. This is dragline silk. Collect a thread of at least 20 cm. Suspend it from a fixed support with a dab of super glue; the underside of the kitchen table will do. Now glue an office staple to the other end but at first support it for a couple of minutes so that the thread is not under stress. Then gently let the staple hang free. Now boil a kettle and pass the steam plume briefly across the thread a few times. Don’t leave the steam billowing at the thread or it will became soaked and burdened with the weight of water.
At first the thread contracts strongly because water plasticizes the less dense regions of the chains and allows them to curl up randomly. It does not just contract, it jerks visibly. This continues with repeated steamings until the contraction is 35–40%. But then the thread starts to extend slowly, a process known as creep. With heavier weights, the initial contraction is less and the creep more rapid. Viney has discovered that microwaving spider silk in a standard kitchen microwave can greatly reduce its sensitivity to water, and this might be the way forward. Whereas, in 2002, the way ahead seemed clear, it had now become very uncertain, with new routes through the maze opening up as old ones became blocked.
Not only is water a problem but spider silk’s most prized ability, that of absorbing impact, has been cast into doubt. The ability of a spider web to catch large objects such as flies without recoil suggested both the idea of the flak jacket and the aircraft carrier arrester wire. But there are obvious problems with each. Material made from fibres with 40% extension is no use for stopping a bullet: you want the back of the flak jacket not to deform at all, otherwise it is going to damage the body. The answer to this is simple: spider silk – if it could be produced in sufficient quantity – would be used in a composite with a more rigid material. As for the arrester wire, it is true that spider silk can absorb a vast amount of energy, but energy cannot be destroyed – it has to emerge somewhere. This is not a problem for the air-cooled gossamer threads of a spider web but a large, thickly coiled multi-strand arrester wire would get very hot indeed: perhaps hot enough to melt the wire.
But, in February 2004, when I met David Kaplan in his office at Tufts University, near Boston, he was remarkably bullish. As far back as 1994, Kaplan co-wrote an important paper on spider-silk protein sequences with David Knight and the two share the belief that the spider’s secret is weighted in the direction of the physical drawing process by which the thread is formed.
Kaplan has worked mainly with reconstituted silks. This has been the principal method of making research-grade silk: taking silk from the one existing industrial source, the silkworm, and recycling it. Silkworm silk is not as tough as spider silk (it is intended to make cocoons so it does not have to be as strong as a web that has to withstand heavy impacts) but it can be processed in ways that bring its properties closer to that of spider silk. First, it has to be reconstituted; spun silk is redissolved in water (it needs quite harsh solvents) until it forms a solution again. It is one of nature’s surprises that spider-silk ‘dope’ can be reconstituted in this way – it is as if you could redissolve paint once it has dried and cured.
David Kaplan’s innovation is that he can now control silk spinning through the water content alone. Reconstituted silks have usually been processed using methanol to convert them from the water-soluble state into the beta-crystalline sheet. But such fibres tend to be brittle. Kaplan says: ‘Now you can take the same approach but stay away from the methanol: you water-anneal them, let the water do it properly. And when you’ve done that you can stretch out these films 300%. They’re completely stable in water and yet there’s virtually no beta-sheet content…It’s really a cool material, apart from making fibres, you can make films, and stuff like the sponge you clean the kitchen sink with.’
David Kaplan was clearly excited about his work and not at all bogged down. Perhaps his techniques were the answer to the scepticism of the revisionists? He admitted that there was much he couldn’t tell me because he had grants from the US Air Force and the National Institutes of Health, and he was also working with an industrial specialist in spinning silks.
The reason for David Kaplan’s optimism became clear when I returned to England. The ending of the US Army’s contract with Nexia did not, as I first thought, mean the end of the dream of the spider-silk flak jacket. The US Defense Department’s website announces that in 2001 they awarded a Small Business Technology Transfer Program grant to David Kaplan and Foster-Miller Inc.,* a hi-tech firm in Waltham, Massachusetts, for Bio-inspired Fibers, Materials, and Properties ‘in an effort to produce films from silk that possess unique and tailorable properties for emerging Air Force applications’. The grant states that: ‘Ultimately, the material is likely to be well suited for highly optimized large space structures such as solar sails or space telescopes, applications where Foster-Miller is currently working on large deployable structures for the Air Force and NASA. In the commercial marketplace, preliminary target applications certainly include bulletproof vests as well as high-strength cords and straps, prosthetic devices, and highly abrasion-resistant textiles.’
In 2003, a follow-up grant to Foster-Miller, this time a Small Business Innovation Research Program, cited: ‘Large Scale Production of Spider Silk by Immortalized Spider Cells.’ Large-scale production is, of course, what Nexia had attempted with their goats. ‘Immortal spiders’ conjures up an image of the Arachne of legend, turned into a spider and condemned to spin for ever for presuming to be the equal of the goddess Athene. But what did it really mean in this context?
David Kaplan did not talk about immortal spiders when we met, but he did talk about stem cells. Stem cells are cells that retain the ability to develop into any kind of specialized cell, depending on which genes are switched on. Most cells can only produce daughters of themselves, but a stem cell can become any organ in the body.
Kaplan’s stem-cell work is mostly aimed at developing replacement tissues, especially things like ligaments, anything made from the kind of proteins he studies: collagen and elastin. And, of course, spiders have stem cells too. Between a spider producing its tiny but exquisite filaments and goats producing spider silk in their milk, there could be a third way: to create a mass of silk-producing cells from a spider’s stem cells. With Kaplan’s new understanding of how to get the water out of silk dope to produce strong films and fibres, the dream of large-scale fabrication using spider silk could be realized at last. It has been a long road, but although we are not there yet, no one who has lived with the problem for the last 10 years and more seriously doubts that a spider-silk fabric will eventually be achieved.
Before I met David Kaplan, I was thinking that industrial spider silk might be 15–20 years away; but I asked him the question anyway, as you do: ‘How long do you think it will take?’ He replied: ‘I tend to be a pretty optimistic guy and I think we’re very close because we understand many of the rules we didn’t understand before. One to five years I would like to say.’ At the time, not knowing about the immortal spiders, I was amazed. I had heard nothing but downbeat projections from everyone else.
The spider-silk story is far from over: it has been running for about 20 years now and the materials scientists are divided between those who think that spider silk is great stuff and well worth researching in its own right and that’s the end of it, and those who believe that it must inspire us to create something equally tough. It is worth remembering that for four years after its discovery nylon was thought to be useless as a technical fibre until the process of cold-drawing was discovered that dramatically increased its strength. Perhaps spider silk is still awaiting a similar boost.
In the early days of bio-inspiration, attention focused on a single outstanding attribute of a creature. The lotus has its leaves, the spider its silk, the mussel its amazing sets-under-water glue, abalone its tough composite shell – but, of course, any creature that has stayed the course during hundreds of millions of years of evolution must be versatile. Spiders are not only great web spinners, they are great climbers on any surface, especially their own webs. Like many other creatures, spiders have thousands of tiny structures on their feet that ensure good adhesion by means of forces that have only very recently begun to be understood. The fine structure of spider’s feet was only investigated in 2004, after pioneering work on the feet of geckos. Each one of the spider’s eight legs, it turns out, ramifies into no less than 624,000 fine hairs at the tip. The gecko has even more hairs per square millimetre than the spider, and thereby hangs an interesting tale: the bigger the creature, the more tightly packed the hairs on its feet. Why should that be?