NMR reveals the silky protein’s hidden strengths.

Picture a spider’s web glistening with morning dew and imagine the spider treading gingerly along the fine threads, putting not a foot wrong for fear of getting herself stuck on the way to a tasty fly meal. Those delicate threads from which she dangles are so thin and wispy—one might think there’s little strength within them. But wrapped up inside the cotton candy filaments of a spider’s silk is all the strength of one of the toughest of proteins.

Spider silk is nature’s high-performance protein fiber. Pound for pound, it has a breakage energy two orders of magnitude greater than steel, but is also lightweight and very stiff. Materials scientists and artificial fiber technicians have only been able to watch the spider spin her yarn with envy. Until now, that is.

Biotechnology holds the promise of allowing us to fabricate our own versions of this natural biopolymer for all kinds of high-tech applications, including tough but comfortable sportswear; wear-resistant shoes; safer seatbelts; artificial ligaments; and tough, lightweight automobile components.

So, why not just farm spiders as silk producers do with silkworms?

Well, the basic problem is that spiders tend to eat each other if you put them together, so each arachnid would have to have its own pen. This means an awful lot of spiders in an awful lot of pens. The other problem is quality control—spiders tend to vary the thickness of their silk as they make it, with results that would not pass muster by the high standards of fiber manufacturers.

To make a spider silk artificially, however, one needs to know the molecular architecture of the natural material, understand the genes that code for the silk proteins, and design a consistent method for spinning the raw material into threads. Disentangling the tough secret of spider silk requires powerful tools to handle its complex properties. NMR is just the tool to unravel the strands and reveal the underlying structure of a spider’s yarn.

Making Silk Is No Drag
Alanine is a major constituent of most silky materials. So, one might consider them to be polyalanine fibers. But nature is never so simpleminded, and spider silk proteins contain twists and sheets with several other amino acids, including glutamine, serine, leucine, valine, proline, and tyrosine.

In the mid-1990s, chemist Lynn Jelinski and colleagues at Louisiana State University (Baton Rouge) used deuterium NMR data from unoriented, oriented, and supercontracted fibers. Their studies indicated that the crystalline fraction of dragline silk is composed of two types of alanine-rich regions: one that is highly oriented and one that is poorly oriented and less densely packed.

They figured that the alanine-rich segments of the silk proteins are associated with crystalline regions, and that it is these sections and their random arrangement within the protein sequence that endow spider silk with its hidden strength.

The dragline silk from the golden orb-weaving spider, Nephila clavipes, is especially tough—a strand one ten-thousandth of an inch thick is as strong as a steel fiber of the same diameter. Combined with stiffness and elasticity on a par with rubber, this makes N. clavipes silk particularly intriguing.

“Spider silk is well designed to meet the demanding biological functions of catching prey and supporting the weight of a suspended spider,” explains Jelinski. (To see a movie of N. clavipes in action, visit http://jrscience.wcp.muohio.edu/movies/orbwebAlmonds.mov.)

A Widow's Strength

Although the golden orb-weaving spider’s webs are interesting, Anne Moore, while working at The University of the Pacific (Stockton, CA) in 1997, discovered that the deadly black widow spider produces silk some 25% more stretchy and twice as strong as steel, making the fiber very tough. Newspaper claims that it could beat Kevlar in a gunfight were highly exaggerated, says Moore. “Spider silk would be particularly ill-suited for a bullet-proof vest because it is so stretchy,” she explains. “It may not rip when the bullet hits, but the bullet would still go through the person, carrying the stretched silk with it.” Moore is looking at the material properties of black widow silks as well as those made by other spiders. “I am doing this in order to figure out how the material properties are modified when spiders use different kinds of traps for catching insects,” she explains. She now works with NMR specialist Barbara Lawrence at Eastern Illinois University (Charleston).

NMR Stretches Spider Silk
Although Jelinski and others have revealed at least some of the secrets of spider silk, such as the -sheet conformation of polyalanine crystallites within each strand, limited structural information is available. Diffraction, the mainstay of protein studies, cannot see past the bulk properties. Classical NMR approaches also fail because of the high molecular weights, the repetitive primary structures, and the structural heterogeneity of solid silk.

Jelinski, now conducting research at Cornell University (Ithaca, NY), and physicist- and materials scientist-colleagues have designed NMR equipment to examine the strengths and weaknesses of the dragline silk secreted from the major ampullate gland of N. clavipes. They hand-feed their spiders deuterium-based food (deuterated alanine in the standard glucose solutions used to feed laboratory microbes) so that the amino acids in the spider silk are prelabeled for their experiments. “They were very hungry after being ‘silked’, and slurped up the stuff,” muses Jelinski.

Solid-state ¹³C NMR measurements were performed on a home-built spectrometer operating at 90.56 MHz. They relied on a triply tuned magic angle spinning probe used in double-resonance mode. Samples were consistently spun at 5 kHz, and chemical shifts were referenced to deuterated adamantane. Cross-polarization experiments were performed with a Hartmann-Hahn matching field of 60 kHz and a mixing time of 2 m.

Previously, Jelinski and her colleagues developed a technique to spin fiber from dissolved natural dragline silk. The process resembles the way polymer chemists might spin a strand of nylon or another artificial fiber from a solution. The team found that the mechanical properties of this regenerated spider-silk fiber were very dependent on the precise processing involved, which, Jelinski points out, highlights the importance of control in the production of artificial protein materials.

Figure 1. Drawn and Tightened. Fresh (unprocessed) regenerated spider silk, as seen by scanning electron microscopy, appears spongy (a) while single-drawn (b) and double-drawn strands (c) are tighter and up to 100-fold stronger.

Initially, the regenerated fibers were rather spongy, but drawing them out resulted in an increase in density and a loss of the porosity common to most wet-spun artificial fibers (Figure 1 at right). A second drawing resulted in a 100-fold increase in the tensile strength of the fiber. The strongest sample of regenerated fiber had a tensile strength of 320 MPa and a modulus (stress-to-strain ratio) of 8 GPa. This is not as strong as the native silk—875 MPa and 10.9 GPa—Jelinski admits. However, she believes that fine-tuning the spinning scheme will ultimately lead to materials competitive with, or perhaps even superior to, natural spider silk. She emphasizes that these “natural” regenerated silks are stronger than the artificial analogs that have been synthesized by other researchers, such as Stephen Fahnestock and colleagues at DuPont.

The NMR studies revealed that the presence of water during the spinning process is crucial to the final properties of the regenerated silk. Without water, the fibers are brittle and break at very low strains: their strength is just 4–8% of that of the hydrated samples.

The bulk properties of any protein fiber depend on the crystallinity, three-dimensional polymer architecture, and overall degree of molecular orientation of the material’s protein components, explains Jelinski. She believes that the improvements the researchers observed in mechanical properties during post-spinning processing are likely to be due to processing-induced modifications at the molecular level, along with a reduction in porosity and a density increase.

Testing Its Strengths
The researchers investigated the influence of the first drawing step of their fiber regeneration using solid-state NMR spectroscopy. ¹³C cross-polarization magic angle spinning NMR spectra were deemed the most suitable methods for analyzing a series of spectra from single-drawn regenerated silk samples. The spectra were recorded and samples differed only in that the length of the post-spinning draw was varied for each.

To demonstrate the fundamental effect of drawing on the spider silk protein structure, the team exploited the dependence of the chemical shift of ¹³C atoms on the local conformation of the protein. In their spider silks, any changes in secondary structure induced by the drawing process would be revealed in the chemical shifts in the 14–22 and 48–53 ppm regions, which correspond to the C_b and the C_a carbons of the amino acid alanine, respectively.

Spinning Out the Study
According to Jelinski and her colleagues, there are basically four important factors that result in the characteristic tensile properties of a regenerated spider dragline silk fiber: the porosity, the fraction of alanine residues adopting the -sheet conformation, the degree of crystallinity, and the overall degree of molecular orientation. They are determined by the post-spinning processing. Water acts to plasticize the silk fibroin and stabilize the processing.

“The future success of any rational design of artificial protein materials will depend on the generality of fundamental studies such as the present one,” concludes Jelinski.

One characteristic that NMR cannot reveal, though—the detailed randomness of the protein sections within each spider thread—is perhaps the underlying property that makes the fibers so tough. Materials scientists hoping to reproduce the spider silk’s strength could do worse than to bear that in mind.

Further Reading

1. Seidel, A.; Liivak, O.; Calve, S.; Adaska, J.; Ji, G. Macromolecules 2000, 33, 775–780.
2. Simmons, A. H.; Michal, C. A.; Jelinski, L. W. Science 1996, 271, 84–87.
3. Yang, Z.; Liivak, O.; Seidel, A.; LaVerde, G.; Zax, D. A. J. Am. Chem. Soc. 2000, 122, 9019–9025.

David Bradley is a freelance science writer based in Cambridge, UK. Send your comments or questions regarding this article to tcaw@acs.org or the Editorial Office 1155 16th st N.W., Washington, DC 20036.