蜘蛛如何将“丝液”纺成自然界最坚韧的材料

By probing its molecular structure, biologists are uncovering how a tangle of unstructured proteins turns into a strong, flexible fiber.

Introduction

A spider, when knocked off a wall, doesn’t fall for long. Within milliseconds, it spins a soup of proteins into sturdy climbing gear. Using a thin, shimmering strand of silk, the spider hauls itself back up.

The strand is tougher than Kevlar, the synthetic fiber in bike tires and bulletproof vests, and stronger than steel. It’s flexible and biocompatible with human bodies. Scientists dream of making spider silk into high-performing materials for seat belts, helmets, naval materials, and even scaffolds to help repair brain cells. But attempts to make spider silk in the lab have fallen short because researchers don’t know enough about how a spider spins “silk dope,” a liquid blend of proteins, water, and salt, from its silk glands into a remarkable solid.

Mapping out the molecular mechanisms behind the process is “really the holy grail” for the field, said Gregory Holland, a chemist and spectroscopist at San Diego State University. Recently, he led research that described how the molecules in liquid spider dope reorganize into semiliquid droplets called condensates before turning into solid silk fiber. The result is an atomic-level look into the spider’s silk factory.

The findings are also a practical guide to the formation of condensates, which nature has used repeatedly to organize molecules for genetic processes and for the creation of impressively strong biological materials.

“It’s becoming clear that [condensates are] a common strategy that’s being used across a range of different organisms,” said Matthew Harrington, a biological materials scientist at McGill University who was not involved in the study. By unpacking how nature uses condensates, scientists hope to re-create some of its most rugged materials.

Silk Factory

Spiders are incredible natural factories. They can spin up to seven types of silks, and the same spider can often spin several of them. Aciniform silk wraps up insects for later snacking. Rubbery flagelliform silk reinforces the capture spiral to absorb shock. Dragline silk anchors webs and spiders to walls. Each type of silk has different properties depending on the species and related to how its molecules are assembled. And silks of the same type can have slightly different properties depending on the species.

In 1907, scientists discovered that spider silk is made of long, modular proteins, later named spidroins, which are in the same group of fiber proteins as collagen and keratin. More than half a century later, X-rays started to reveal the fiber’s underlying structure. The silk is made of flat sheets of protein, known as beta sheets, that are embedded in a less ordered, non-crystalline matrix.

In the early 2000s, as they inspected the spidroins in solution, researchers didn’t see organized, fixed structures like those seen in most proteins. Instead, the spidroins featured intrinsically disordered regions that fluttered around, forming and re-forming various shapes. Researchers then started to untangle how this soup of unstructured proteins could organize into strong, solid fibers.

One early theory suggested that proteins in the dope start to align with one another, as in a crystal, while remaining dynamic, in a transitional state known as “liquid crystal.” Another considered spider silk to be like silkworm silk, which has proteins organized into micelles, or small spheres, which then compress into silk fiber.

Then, in 2007, researchers noted that spidroins engineered in the lab formed liquidlike globs. Soon these kinds of structures would become inescapable in modern biology. In 2009, scientists at the Marine Biological Laboratory in Massachusetts and the Max Planck Institute in Germany discovered droplets of aggregated RNA molecules in fertilized worm eggs. These were condensates — liquid- or gel-like bubbles of concentrated proteins suspended in solution. Though observed for more than a century, condensates started to be seen as a critical organizational tool for molecular processes across life.

The researchers realized that condensates concentrate molecules, pre-organizing them into a flexible microenvironment without locking them in place like a solid. “Because they’re fluid, they’re malleable,” Harrington said, “so you [i.e., biology] can kind of shape them into whatever you want.”

More research showed that living cells use condensates everywhere, to organize DNA, regulate gene expression, and package proteins. At a macro scale, they help species build interesting materials. They help mold squid beaks into a gradient of stiff parts. They form strong adhesives that allow mussels to attach themselves to rocks, and marine cement that sandcastle worms use to build protective tubes.

“It’s been a hot area in biology,” Holland said.

Though researchers had described condensate-like blobs forming in lab-made spidroins more than a decade earlier, they weren’t thought of as condensates until 2020, when Keiji Numata, a biomaterials chemist at Kyoto University, suggested that spider silk undergoes a condensate phase during its transition from liquid dope to solid thread.

Condensates form by “liquid-liquid phase separation,” a process through which molecules in a liquid can start to interact weakly with one another, easily forming bonds and breaking them, as a result of a stimulus such as a change in pH, chemicals, or temperature. Numata proposed that phosphate in the ampulla, the middle section of a spider’s silk glands, might initiate these interactions. But no one had described the atomic-scale mechanisms that drive their formation until Hannah Johnson, then a graduate student in Holland’s lab, decided to take a look.

Inside a Black Widow

Every few months throughout 2023 and 2024, a package of female black widow spiders arrived at the San Diego State University chemistry department. The arachnids, jet black with a dash of red, cultivate a powerful venom that can be painful to humans. Except for that minor inconvenience, they make good subjects for studying spider silk. Their dragline silk spidroin proteins are some of the best-characterized in all spiders. And they’re available year-round, unlike some spiders that spin their webs seasonally.

Six times over the course of two weeks, Johnson donned two pairs of protective gloves. Over the course of an hour, she and several students “silked” black widow spiders, which had been knocked unconscious by carbon dioxide. With tweezers, the scientists gently pulled the strand of silk from the spider’s spinnerets — the abdominal spigots that fibers emerge from — and reeled it in by winding it around a vial with a spinning motor.

“The spiders who give a ton of silk are our favorites,” Johnson said, “but I have a soft spot for the rebels who don’t want to cooperate.” Throughout the process, the researchers fed the black widows a cocktail of amino acids (like “protein shakes,” Johnson said) enriched with particular isotopes that they could detect in the lab.

Nuclear magnetic resonance (NMR) spectroscopy can map the structure of a molecule. It sends out radio frequencies to disturb the nuclei inside atoms, which then move in a way characteristic of the type of atom. Johnson used NMR on the liquid spider dope, as well as on condensates collected from the spiders’ ampullae. Meanwhile Kevin Chalek, then a postdoctoral researcher in Holland’s lab (now a researcher at the University of California, Los Angeles), mapped the atomic structure of the proteins in the solid fiber.

“We were trying to bridge the gap between what is happening when the protein is inside the spider and when it’s outside the spider,” Chalek said. “These are the same protein, [so] how can they be so different? And what exactly is happening between the two?”

Johnson observed that after she added phosphate to the liquid silk dope to start the condensation process, side groups on two amino acids (arginine and tyrosine) in the intrinsically disordered regions of the spidroin proteins started to interact weakly with one another. This interaction displaced water, organizing the disordered proteins into a denser, gel-like material — a condensate. The interaction persisted in the solid fiber, suggesting that the amino acids serve as “molecular stickers” that guide the disordered proteins into condensates and ultimately into fibers.

Others on the team used AlphaFold, an artificial intelligence tool that models protein structures, and molecular dynamics software to predict this process computationally. Researchers know that spider silk’s strength comes from the sturdy beta-sheet protein structures in its final fiber. The team’s simulations showed that these amino acids line up right up at the bends in the beta sheets, suggesting that they are helping to guide the spindroins into their final structure.

“Understanding how those side groups interact and how they can develop structure that’s greater than the sum of its individual parts … is fantastic,” said Chris Holland, a biologist at the University of Sheffield who was not involved in the study and is not related to Gregory Holland. And they used real spiders, instead of synthetic silks, “which is brilliant.”

Their findings are “direct evidence” that condensates can form from native silk, which is “quite significant,” said Daiwen Yang, a biologist at the National University of Singapore who was not part of the study.

Their description of condensates doesn’t rule out earlier theories of spider silk formation. Gregory Holland, who had previously shown the presence of micelles in native spider silk, suggested that all three — micelles, condensates, and liquid crystals — could be stages of the same phase transition. Micelles that form in the liquid dope prevent the messy, disordered proteins from clumping and getting tangled. The proteins can also pre-assemble into condensates, or align into structures like liquid crystals before they are released from spinnerets as highly ordered fibers. But the actual process is likely “more dynamic and heterogenous” than sequential, Gregory Holland said, with several of these structures forming in the mixture at the same time.

Harrington agreed. “I don’t think these are mutually exclusive,” he said. “You might have different sorts of phase transitions that are occurring at different stages of the processing.”

Still, there’s more to understand about how these interactions add up on this factory line. There are likely other significant amino acid interactions that the researchers haven’t yet been able to map out in detail. They also want to look for triggers that initiate the process beyond phosphate. “There could be a range of other things that are driving structural changes in these proteins,” Chris Holland said. “We don’t know all of the switches yet.”

Superior Materials

Describing the entirety of the intricate silk-spinning process will take a lot of effort, but the payoff could be huge. Spiders produce small amounts of usable silk compared to silkworms, which have been reared for their cocoons for more than 4,000 years. It takes thousands of silkworms to make a silk shirt; it would take around 1 million spiders, in part because they don’t produce cocoons that can be easily processed. There is a more practical reason: Their cannibalistic nature makes them difficult to keep because each spider needs its own container.

Producing spider silk in the lab could result in a new range of strong yet flexible materials. Kevlar, a synthetic fiber, is currently our toughest material, but it requires processing at very high temperatures and acidic conditions. “Here, you have an organism like a spider or a mussel,” Harrington said, “which is basically making a superior material under ambient conditions in seawater or in just plain air.”

Humans haven’t figured out how to match those organisms yet. When scientists synthesize spider silk in the lab, using the exact sequence of amino acids to build a spidroin protein, the resulting silk is not as strong as natural silk. “Something is happening there that we’re also struggling to mimic,” Chalek said. He thinks additional strength must be provided by something beyond the protein’s beta sheets, arising from other aspects of silk’s structure, or perhaps from parts of the spidroin protein frequently left out of synthetic silk solutions.

“I think once we really understand those blueprints,” Chris Holland said, “we’ll be able to do more stuff with silk.”

Next article

How Voting Is Like a Klein Bottle