Eri silkworm spins mechanically robust silk fibers regardless of reeling speed

Silkworms produce proteinous fibers with biocompatibility and biodegradability. Lightweight and lustrous silk fibers are traditionally used as expensive textile materials with excellent strength and ductility (Omenetto and Kaplan, 2010; Vollrath and Porter, 2009). Hence, in the context of sustainable development goals, silk is regarded as a candidate for promising bio-based fibers (Agnieray et al., 2021). Generally, silk is commercially produced by domesticated silkworms, which have been used for centuries (Ude et al., 2014). Wild silkworm silk fibers are used much less than domesticated silkworm silk fibers because the number of wild silkworms is scarce, and it is difficult to extract silk fibers from the cocoons of wild silkworms because of their mineral reinforcement (Rozet et al., 2019; Rozet and Tamada, 2019; Zhou and Wang, 2020). It has been reported that demineralization using a chelating agent allows the reeling of commercial lengths of wild silk fibers with good quality, which contributes to the development of the wild silk industry (Gheysens et al., 2011).

A lot of attention has been paid to a variety of wild silkworms not only for the production of silk but also for their benefits in structural and medical materials (Kundu et al., 2012). This is not only because wild silkworm silk fibers have their own natural color but also because they often have excellent mechanical properties compared with those of the domesticated silkworm (Guan et al., 2017; Malay et al., 2016). The mechanical properties of silk fibers are largely influenced by their crystal structures (Guo et al., 2017). It is known that the crystal region of silk mainly determines its mechanical strength. However, the amorphous region of silk contributes to its extensibility (Fang et al., 2016). The crystal region of wild silkworm silk consists of polyalanine sequences, which is the case with spider silk (Malay et al., 2016). The average length of the polyalanine motifs of wild silkworm silk is often longer than that of spider silk (Numata et al., 2015; Tsubota et al., 2016). In contrast, the crystal region of domesticated silkworm silk has alternate sequences of glycine and alanine (Malay et al., 2016). We have reported that dragline silk fibers, which are used as mechanical lifelines of spiders, maintain their structural and mechanical properties irrespective of the reeling speed during forced spinning (Yazawa and Sasaki, 2021). The mechanically robust nature of spider dragline silk contrasts with the case of domesticated silkworm (Bombyx mori) silk fibers. The molecular orientation and mechanical strength of domesticated silkworm silk fibers are known to be significantly influenced by the reeling speed (Du et al., 2006; Khan et al., 2008; Mortimer et al., 2013; Pérez-Rigueiro et al., 2001b; Riekel et al., 2000; Shao and Vollrath, 2002; Wu et al., 2009). Domesticated silkworm silk fibers show higher degrees of orientation and crystallinity with an increase in reeling speed, enabling improvement of the tensile strength of silkworm silk. As wild silkworms need to spin their silk fibers regardless of the weather, it is speculated that the spinning mechanism of silk fibers might differ between wild and domesticated silkworms. In addition, wild silkworms hasten to spin their silk fibers in the case of emergencies, such as abrupt falling and sudden invasion by their predators. Accordingly, we hypothesized that wild silkworms might be endowed with a higher level of adaptation against environmental disturbances compared with domesticated silkworms.

To verify the above-mentioned hypothesis, in this study, we used the larvae of an eri silkmoth, Samia cynthia ricini, as a model of wild silkworms, and we forcibly reeled silk fibers from a live wild silkworm. The body size of eri silkworms is the third largest after B. mori and tussah, and the lifecycle of eri silkworms is shorter than that of B. mori. Additionally, the cocoon size of eri silkworms is larger than that of B. mori. Bombyx mori silkworms can only be fed with mulberry leaves, while eri silkworms can be fed with the leaves of several trees. These features make it easier to rear eri silkworms. It is known that the fracture surface of wild silkworm silk has a porous structure, which can be a structural defect (Akai, 2005). Therefore, it is also interesting to know how the fracture surface of eri silk fibers changes after forced reeling. Accordingly, we investigated the effects of reeling speeds on the structural, mechanical and morphological properties of eri silkworm silk.

Eri silk fibers were forcibly collected from wild silkworms, Samia cynthia ricini (S. c. ricini) (Boisduval 1854), which were reared at Shinshu University in Nagano, Japan. Diet can affect the amino acid composition of silk fibers (Cheng et al., 2018; Ito et al., 1995). Accordingly, the eri silkworms used in the experiment were only fed with castor bean leaves to minimize the effects of dietary conditions. The eri silk fibers were forcibly and directly reeled from the late fifth instar larvae of S. c. ricini restrained by two pieces of sponge and fixed with rubber bands (Fig. 1). The forced reeling process was performed in an experimental chamber, where the temperature was set at 25°C and the relative humidity (RH) was set at 40%. The eri silk fibers were reeled onto a spool that was rotated by a motorized apparatus (Harada Corp., Nagano, Japan). The reeling speed was set at the following constant values: 3, 15, 30, 45, 60 and 75 mm s⁻¹, covering most of the natural spinning speeds of silkworms (Miura et al., 1993). For each speed, the silk samples from the start and end of reeling were avoided. All the collected silk fibers were left to dry in lightproof boxes with silica gel to minimize the structural change of silk fibers derived from the variation of extrinsic factors (Boulet-Audet et al., 2016; Yazawa et al., 2020). After the forced reeling process, the wild silkworms produced cocoons.

A FT-IR device (IR Prestige-21, Shimadzu Corp., Kyoto, Japan) equipped with an attenuated total reflectance accessory (DuraSamplIR, Smith Detection, London, UK) was used to examine the peaks derived from the secondary structures of the eri silk fibers. For each measurement, 128 scans were accumulated with a resolution of 4 cm⁻¹ at 25°C under a RH ranging from 35% to 45%.

A synchrotron WAXS measurement was performed on bundles of eri silk fibers using an X-ray energy of 12.4 keV (wavelength: 0.1 nm) at the BL-10C beamline of Photon Factory (Tsukuba, Japan) (Shimizu et al., 2019). The sample-to-detector distance for the WAXS measurement was 238 mm, and the exposure time for each scatter pattern was 10 s at 25°C under a RH ranging from 30% to 40%. The obtained scattering data were converted into one-dimensional and radially integrated profiles using Fit2D software (Hammersley et al., 1996). The data were corrected for background scattering. Also, the crystallinity was calculated by dividing the area of the crystal peaks by the total area of the crystal peaks and amorphous halo by fitting the Gaussian function using Igor Pro 8.03 (WaveMetrics, Inc., Portland, OR, USA). Additionally, the degree of orientation was evaluated from the (210) peak representing the crystalline β-sheet structure based on the full width at half-maximum (FWHM) value, according to previous procedures (Malay et al., 2016; Yazawa et al., 2020).

The eri silk single fiber obtained by forced reeling was subjected to tensile tests using a test apparatus (RTC-1250A, A&D Company Ltd, Tokyo, Japan) at 25°C under a RH ranging from 35% to 40%. The initial length of the eri silk single fiber was set at 5 mm, and the deformation rate was set at 10 mm min⁻¹ using a load cell with 1 N. The silk fiber was carefully fixed with the same force each time. The fracture surface and diameter of the tensile test specimens of the eri silk fibers were individually examined by scanning electron microscopy (SEM; see below). The cross-sectional area of the tensile test specimen of the eri silk fibers was individually evaluated by tracing the edge contours of the silk fibers. Twenty samples per silking speed of the eri silk fibers were tested. All data were processed in Microsoft Excel by excluding the slack regions from the initial stages of the tensile tests. The tensile strength, elongation at breaking, Young's modulus and toughness were evaluated based on the resultant stress–strain curves.

The fracture surface, side-view morphology and diameter of the tensile test specimens of the eri silk fibers were individually examined by SEM (SU1510, Hitachi High-Tech Corporation, Tokyo, Japan) at 25°C. The eri silk fibers were mounted on an aluminium stub with conductive tape and then sputter-coated with gold for 1 min using a smart coater (JEOL, Tokyo, Japan) prior to SEM visualization at 15 kV. The cross-sectional area of the individual tensile test specimen of the eri silk fibers was evaluated by tracing the edge contours of the silk fibers using ImageJ (NIH, Bethesda, MD, USA).

Statistical analysis was performed using Student's t-test with Microsoft Excel to determine whether the averages were significantly different. The differences were considered significant when the P-values were less than 0.05.

FT-IR spectroscopy was applied to evaluate the bending vibrations and stretching associated with amide bonds originating from secondary structures of eri silkworm silk fibers. WAXS measurement can detect the coherent interference of scattered X-rays passing through the crystal (Fig. 2), enabling us to monitor the crystal regions of eri silkworm silk fibers. Therefore, the protein secondary structure of the eri silkworm silk fiber was monitored by FT-IR, while the crystallinity was evaluated from the WAXS data. Forced reeling of eri silk fibers was performed at different reeling speeds. The obtained eri silk fibers were examined by FT-IR to analyze the secondary structures of the silk fibers (Fig. 3A). A peak centered at a wavenumber of 1625 cm⁻¹ indicates the presence of crystalline β-sheets in the eri silk fibers (Moseti et al., 2019a,b). The FT-IR spectra of the eri silk fibers displayed no observable peak shifts based on the reeling speed (Fig. S1). The crystallinity and degree of molecular orientation were examined based on the WAXS data. Two-dimensional WAXS images of the eri silk fibers (Fig. 2) were converted into one-dimensional plots using radial integration (Fig. 3B). The (210) and (020) peaks were detected in the one-dimensional WAXS profiles, indicating the formation of a crystalline β-sheet structure (Moseti et al., 2019a,b). Based on the WAXS profiles, the positions of the peaks originating from the crystalline β-sheets were maintained regardless of the reeling speed (Fig. 3B). Thus, the crystal structure of the eri silk fibers did not change. Small peaks in WAXS data from the different reeling speeds (3, 30, 45 and 60 mm s⁻¹) were derived from calcium oxalate that covers the cocoons of wild silkworms (Gheysens et al., 2011).

The one-dimensional WAXS plot was deconvoluted into crystal and amorphous fractions using Gaussian functions to examine the crystallinity of the eri silk fibers (Fig. S2), which was maintained irrespective of the reeling speed (Fig. 4A). The effects of the reeling speeds on the molecular orientation degree were also analyzed based on the FWHM values that were calculated by the azimuthal integration of the (210) peak in the two-dimensional WAXS profiles (Fig. 4B). The FWHM values did not significantly change with the different reeling speeds. These results indicate that the crystal structure and molecular alignment of the forcibly reeled eri silk fibers were not influenced by the reeling speed. Additionally, the diameter of the eri silk fibers was evaluated by SEM measurements (Fig. 4C). The results showed that the diameter was not affected by the reeling speed.

The structural features of the eri silk fibers were not influenced by the reeling speed. Next, the effects of the reeling speed on the mechanical parameters of the eri silk fibers were examined through tensile testing. The eri silk fibers presented tensile strength values ranging from 0.15 to 0.20 GPa and extensibility values ranging from 15% to 35% under the reeling speeds used in this study (Fig. 5). These tensile strength and extensibility values are within the range of naturally spun eri silk fibers (Malay et al., 2016). Thus, the stress–strain curves of the eri silk fibers were not significantly influenced by the reeling speed. The tensile strength, elongation at break, Young's modulus and toughness of the eri silk fibers are summarized in Fig. 6, based on the stress–strain curves. The changes in the mechanical parameters were negligible for the given reeling speeds.

The morphological features of eri silk fibers obtained by forced reeling under different reeling speeds were visualized by SEM imaging, as shown in Fig. 7. Eri silk fibers comprise a semicrystalline fibroin core and an outer layer of sericin. The surface of the eri silk fibers was smooth regardless of the reeling speed. However, the fracture surface of the naturally spun eri silkworm silk fibers had voids (Fig. S3). Furthermore, the diameter of the silk samples from the start of reeling tended to increase (Fig. S4). In addition, smaller diameters and cracks were detected in the eri silk samples from the end of forced reeling for producing a continuous thread of several hundred meters in length (Fig. S5), probably originating from the lack of protein solution in the spinning dope or from defects in the sericin layers formed during silk fiber formation. These morphological features suggest that wild silkworm S. c. ricini spin silk fibers with a constant diameter irrespective of the reeling speed.

The effects of reeling speed on the structural and mechanical properties of forcibly spun silk fibers were examined using S. c. ricini as a model of Saturniidae wild silkworms. The crystal structure, surface morphology and mechanical parameters of the eri silk fibers were not significantly changed by the reeling speed. The obtained results agree with those of a previous study, in which forced reeling of dragline silk fibers from the web-building spider Trichonephila clavata was performed at different speeds (Yazawa and Sasaki, 2021). Dragline silk fibers ensure robustness irrespective of the spinning speed. Day and night, insects and animals living in the wild should adapt to environmental changes. It has been reported that the physical properties of domesticated silkworm cocoons are influenced by environmental temperature, wind speed and humidity. Environmental temperature affects the speed of the figure-of-eight movement of domesticated silkworms and changes the morphological properties of cocoons (Offord et al., 2016). Moreover, humidity conditions vary the speed of water removal from newly spun silk fibers, which influences the crystallization behavior (Boulet-Audet et al., 2016), and wind loading might affect the spinning speed of silkworms (Vollrath et al., 2001). Such environmental disturbances might also change the physical properties of the silk fibers produced by wild insects and animals. Accordingly, it is essential for insects and animals in their natural habitats to maintain the mechanical toughness of their silk fibers. Pérez-Rigueiro et al. (2001a, 2005) have demonstrated a strong correlation between the silking force and tensile behavior of spider silk. They found that silk fibers spun at high silking force are stiff, while silk fibers spun at low silking force are extensible. Considering that eri silkworms produced structurally and mechanically robust silk fibers in this study, eri silkworms might add a particular level of tensile load to their silk fibers by adjusting their ventral and dorsal muscles near the spinneret. To reveal the details of the muscle manipulation, we are now investigating the electric potential that is exerted near the spinneret of wild silkworms and spiders during forced silking to compare it with the electric potential exerted by domesticated silkworms. It should be noted that the variation in physical properties between eri silkworm and B. mori silk fibers might be derived from the different genus and species. Further studies are needed using other species of wild silkworms to elucidate the influence of genus.

Akai (2005) reported that the fracture surfaces of wild silkworm silk fibers have abundant voids. These voids are thought to be derived from the lysosomes in silk glands and to play an important role in maintaining certain temperature and humidity values inside cocoons. However, such voids can be structural defects and can result in weaker mechanical strength. After forced reeling, the fracture surfaces of eri silk fibers were visualized by SEM (Fig. 8; Fig. S6). The fracture surfaces of eri silk fibers displayed dense structures without voids. Thus, forced spinning is useful in decreasing the structural defects of silk fibers and in contributing to the production of mechanically tough silk fibers.

The effects of various reeling speeds on the structural, morphological and mechanical properties of forcibly spun silk fibers were analyzed using S. c. ricini as a model of wild silkworms. The crystal structures and mechanical properties of the obtained eri silk fibers were not significantly changed by the reeling speed. In addition, the morphological aspects of the eri silk fibers demonstrated fixed diameters and flawless surfaces although the diameter tended to decrease, and cracks in the surface layers were detected at the end of the forced reeling process. These results indicate that S. c. ricini silkworms have evolved to spin silk fibers with a particular level of mechanical toughness so that they can be used as protective shells for their pupae. Wild silkworms might exert a constant tensile force by adjusting the ventral and dorsal muscles near their spinnerets, enabling them to spin mechanically robust silk fibers. This indicates that the insects and animals living in the wild have evolved to spin silk fibers with robustness to the reeling speed. Overall, the results of the forcibly spun wild silkworm silk fibers in this work can help in understanding the natural spinning mechanism of silkworms and pave the way to determining the best spinning conditions for synthesizing artificial silk fibers with comparable mechanical properties to those of their natural counterparts. We are now developing an automated rearing system of silkworms from eggs to cocoons in germ-free conditions for the bulk-scale production of silk fibers. As eri silkworms are resistant to reeling speed in comparison to domesticated silkworms, eri silkworms may be promising candidates for producing silk fibers using the automated rearing system.

The WAXS measurement was performed under the approval of the Photon Factory Program Advisory Committee (proposal no. 2021G006).