Relating the structure of insect silk proteins to function

Abstract

Silks are extracorporeal fibrous protein materials. Classically, silkworm (Bombyx mori) and orb-spiders (Arachnida: Araneidae) have served as model organisms in which to investigate silk protein structure-function relationships. However, silk production has evolved multiple times in insects. The silk proteins of many insects do not fold into the beta-sheet structures found in silkworm and spider silks but into coiled-coils, collagen helices or polyglycine helices. Therefore, the structure-function relationships elucidated for silkworm and spider silk proteins may be too narrow to apply to insect silk proteins generally. To increase the available data, I examined silk production by raspy crickets (Orthoptera: Gryllacrididae), silverfish (order Thysanura), praying mantises (order Mantodea), glow-worms (Diptera: Keroplatidae), and sawflies (Hymenoptera: Tenthredinidae). Silk protein primary structures were investigated using transcriptomics, mass spectrometry, and amino acid analysis; secondary and tertiary structures were investigated by infrared and Raman spectroscopy, nuclear magnetic resonance, circular dichroism spectroscopy, and bioinformatics. Novel features of silk production were related to idiosyncrasies of each insect group, while features found in multiple silk-producing groups were associated with general mechanisms of silk production. A comparative analysis of silk proteins revealed a correlation between predominant secondary structure type and more general architectural features such as length and repeat regularity: silk proteins that fold into coiled-coils and collagen helices had low molecular weights and high repeat regularity, suggesting they fold into short semi-rigid rods; beta-sheet-forming silk proteins were found to be more variable in molecular weight and have lower repeat regularity. Based on these data, I propose three major mechanisms of silk fabrication by insects: a) mesogenic ordering of short rod-like proteins, a process for which the coiled-coil and collagen structures are well-suited; b) molecular extension of long flexible protein chains to promote intermolecular bonding, which is suitable for the formation of beta-sheet-rich silks; and c) entanglement of protein chains, which is suited to silks with a high degree of disorder. Thus, many features of insect silk proteins are adaptations for material fabrication. In a few cases, particular structural motifs constituted adaptations conferring a mechanical property required for the silk's function in the solid state. However more often proteins were observed to have features promoting dense protein packing in a general way. I explain these data by consideration of how silk mechanical behaviour relates to the fitness advantage conferred to individual insects by silk production. Specifically, I suggest protein features ensuring structural homogeneity and molecular orientation result in silk materials with mechanical properties sufficient for most purposes. Further increases in properties such as strength lead to little or no fitness increase. Local maxima in the fitness landscape associated with distinct protein secondary structures or fabrication mechanisms trap silk proteins in one of several states. Overall, silk protein evolution can to a large extent be understood as convergence of a number of independently co-opted proteins of other functions toward one of several distinct functional archetypes.