Engineered Spider Silk Proteins for Biomimetic Spinning of Fibers with Toughness Equal to Dragline Silks

Abstract

Spider silk is the toughest fiber found in nature, and bulk production of artificial spider silk that matches its mechanical properties remains elusive. Development of miniature spider silk proteins (mini-spidroins) has made large-scale fiber production economically feasible, but the fibers' mechanical properties are inferior to native silk. The spider silk fiber's tensile strength is conferred by poly-alanine stretches that are zipped together by tight side chain packing in β-sheet crystals. Spidroins are secreted so they must be void of long stretches of hydrophobic residues, since such segments get inserted into the endoplasmic reticulum membrane. At the same time, hydrophobic residues have high β-strand propensity and can mediate tight inter-β-sheet interactions, features that are attractive for generation of strong artificial silks. Protein production in prokaryotes can circumvent biological laws that spiders, being eukaryotic organisms, must obey, and the authors thus design mini-spidroins that are predicted to more avidly form stronger β-sheets than the wildtype protein. Biomimetic spinning of the engineered mini-spidroins indeed results in fibers with increased tensile strength and two fiber types display toughness equal to native dragline silks. Bioreactor expression and purification result in a protein yield of ≈9 g L^-1 which is in line with requirements for economically feasible bulk scale production.

Keywords: biomimetic materials; biomimetic spider silk fibers; fibers; protein engineering; recombinant protein production.

Figure 1

Schematic representation of the designed constructs. A) NT2RepCT (A₁₅‐A₁₄) is composed of an N‐terminal domain (NT, red; PDB: 4FBS), a repeat region with two poly‐Ala blocks (green and yellow), and a C‐terminal domain (CT, blue, PDB 3LR2). Both subunits of the soluble NT2RepCT dimer are shown (one is shaded). B) Protein sequence alignment of the repetitive region from A₁₅‐A₁₄ and engineered constructs thereof. Note that all constructs contain NT, a repeat part, and CT. Substitutions in the poly‐Ala blocks are indicated in orange.

Figure 2

Rosetta energy profiles of A) A₁₅‐A₁₄ and (A₃I)₃‐A₁₄ (profiles for all designed proteins are found in Figure S1 and Table S2, Supporting Information). Bars show Rosetta energies for moving hexapeptides (indicated at the first residue of each hexapeptide), red bars indicate Rosetta energies equal or below −23 kcal mol⁻¹ (dashed line). Green bars indicate Rosetta energies above the threshold and are unlikely to form steric zippers (https://services.mbi.ucla.edu/zipperdb/).^{[ 64 ]} B) Bars indicate the Rosetta energy of the hexapeptide with the lowest predicted energy from A₁₅‐A₁₄ and the engineered mini‐spidroins (all hexapeptides are shown in Table S2, Supporting Information). C) Hypothetical zipper structure of two β‐sheets composed of hexapeptides AAAAAA from A₁₅‐A₁₄ and AIAAI derived from (A₃I)₃‐A₁₄, respectively.

Figure 3

CD spectroscopy of purified engineered mini‐spidroins. A) Initial spectra at 20 °C and B) molar ellipticity measured at 222 nm from 20 to 90 °C was converted to fraction natively folded (%) and then normalized. CD spectroscopy of different constructs C) heated to 90 °C and D) after cooling to 20 °C.

Figure 4

Mechanical properties of spinnable engineered mini‐spidroins in comparison with A₁₅‐A₁₄. A) Photographs of the biomimetic spinning set‐up; a video of the spinning can be found in Video S1, Supporting Information. B) Photographs of spun fibers. C) Strength, D) strain at break, E) toughness modulus, dashed line indicates toughness modulus of a native dragline silk,^{[ 10 ]} and F) Young's modulus. Whiskers show standard deviation. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Representative stress–strain graphs for all spinnable engineered mini‐spidroins are shown in Figure S5A–G, Supporting Information. The diameters of the fibers are shown in Figure S5H, Supporting Information. The values and corresponding standard deviations are shown in Table S3, Supporting Information.

Figure 5

FTIR spectroscopy of engineered fibers. Normalized and baseline‐subtracted absorbance spectrum in the amide I region of A) A₁₅‐A₁₄, (A₃V)₃‐(A₃V)₃, (A₃V)₃‐A₁₄, and (A₃T)₃‐(A₃T)₃ and B) A₁₅‐A₁₄, (A₃I)₃‐A₁₄, A₁₅‐(A₃I)₃, and (A₂I)₄‐A₁₄. C) Percent secondary structure content determined by cofitting the absorbance spectrum and the second derivative. Horizontal line indicates β‐sheet content of A₁₅‐A₁₄. Fits of absorbance spectra and second derivative of fibers spun are shown in Figure S7, Supporting Information.

Figure 6

Solid‐state NMR ¹³C‐¹³C correlation spectra (aliphatic region) of A₁₅‐A₁₄ (blue) and (A₃I)₃‐A₁₄ (red) fibers. The Cα/Cβ correlations of Ala and Ile in α‐helical and β‐sheet conformation are indicated.