For thousands of years, humans have prized silk for its strength, luster, and versatility. Today, silk proteins are making their way into cutting-edge applications ranging from flexible bioelectronics and sustainable water-treatment materials to 3D-bioprinted tissues. Yet despite centuries of use, a fundamental question has persisted: how do silkworms transform a liquid protein solution stored in their silk glands into one of nature’s toughest fibers—seemingly without the protein prematurely solidifying?
Silk is composed of two main proteins. Silk fibroin (SF) forms the structural core of the fiber, while silk sericin (SS) coats the fibroin like a layer of glue. For decades, SS was viewed as a passive, secondary ingredient—helpful for sticking fibers together, but not much more. However, scattered evidence hinted that SS might be more than a bystander: higher SS concentrations slowed SF aggregation, though contradictory reports left its role unclear.
To resolve the mystery, a team led by Professors Mingying Yang and Yajun Shuai at Zhejiang University, together with Professor Chuanbin Mao at The Chinese University of Hong Kong, designed a biomimetic system that recreates the chemical environment inside the silkworm’s silk gland during the critical fifth larval instar—the stage when silk production peaks. Using regenerated silk fibroin (RSF) and sericin (RSS), they tracked how the two proteins interact over six days of incubation.
The results revealed a striking picture. Under salt-assisted conditions mimicking the weakly alkaline environment of the silk gland, RSS promoted liquid-liquid phase separation (LLPS)—a process in which the protein solution spontaneously partitions into dense, liquid-like droplets called coacervates. Cryo-electron microscopy confirmed these droplets remain fully hydrated and dynamic, capable of fusing, budding, and reorganizing within seconds. Confocal imaging showed a distinct core-shell architecture: fibroin concentrated in the core while RSS localized to the periphery, forming a natural "reaction chamber".
Even without added salts, RSS rapidly formed aggregate compartments that captured RSF nanoparticles. Over six days, these confined RSF proteins matured into highly ordered, parallel-aligned nanofibrils, with β-sheet content nearly tripling from ~11.2% to 27.1% and crystallinity markedly increasing. When the team applied mechanical shear—mimicking the forces silk proteins experience as they pass through the spinneret—the compartments disassembled and reorganized, giving rise to ultra-long nanofibrils.
The study fundamentally revises the textbook view of RSS. Rather than a passive lubricant, RSS acts as an endogenous regulator that: (1) locally concentrates and stabilizes RSF through compartmental confinement, preventing premature and disordered aggregation; and (2) guides the oriented arrangement of RSF nanofibrils through compartment-driven maturation. This dual role bridges the gap between molecular-scale self-assembly and the macroscopic properties of silk fibers.
Beyond silk biology, the findings offer a blueprint for designing programmable, bioinspired materials. The multi-stage phase transition framework—from soluble monomers through coacervates, aggregates, and gels to ordered fibrils—demonstrates how nature avoids disordered precipitation by stepping down the free-energy landscape through discrete intermediate states. Similar principles could guide the development of synthetic protein-based fibers, smart hydrogels, and hierarchically structured biomaterials for tissue engineering and beyond.
"Sericin is far more than glue—it’s the conductor of a symphony," said Dr. Yajun Shuai, the study’s corresponding author. "Our work shows that by creating confined, dynamic compartments, sericin provides the spatial and temporal control needed to transform disordered protein solutions into the exquisite hierarchical structure of silk. Understanding this natural assembly line opens the door to engineering materials with similarly elegant architectures."
Science Bulletin
Experimental study