260
W. Qiu and X.-Y. Liu
spectra; the β-strands in the crystallites were aligned along the fibrous axis in silk
fibers but were perpendicular to the long axis of amyloid fibrils. Statistical results
show that most force-extension curves (94%) of RSF nanofibrils exhibit sequential unfolding events involving random peak forces; however, no clear trend was
observed. For the rest of the trajectories (6%), another characteristic pattern was
identified [18]. The highest unfolding force peak was followed by a series of events,
which shows a general upward trend in unfolding forces. Given the fact that hydrogen
bonds between β-strands and hydrophobic interactions between β-sheet layers are
responsible for the stability of β-sheets and β-crystallites, the two distinct trajectories
imply two possible pathways for unfolding β-strands from β-crystallites. Pathway 1
involves the sequential unzipping of β-strands directly from β-crystallites (without
affecting the inter-β-sheet interactions), whereas Pathway 2 involves the peeling of a
β-sheet off the crystallite first, followed by the unzipping of β-strands from the peeledoff β-sheet. In the case of Pathway 1, at each step of the unzipping of the β-strands,
the force is mostly applied onto the strand directly connected to the AFM tip, while
an equal amount of force is exerted on the other strands in the β-crystallites, owing
to the integrated response of the β-crystallites. Hence, the strands directly linked to
the AFM tip should experience a force much stronger than in other locations, so they
are unzipped beforehand. This is then repeated for the next adjacent strand. As the
strength of collective hydrogen bonding between different β-strands is different, the
corresponding force pattern should result in peaks with different heights. According
to the slab-segment model, the force distribution among β-strands in the β-sheet is
similar to that in β-crystallites; hence, the force pattern should be similar to Pathway
1 [36]. However, if the β-crystallites are cross-linked with each other, as seen in a
fishnet network, the peeled-out β-sheet will be stretched out between two anchor
points, i.e., the AFM tip and the remaining β-crystallites. In this condition, an equal
amount of stretching force is applied on all the β-strands within the peeled-off β-sheet
because they are connected in a series. Thus, the weakest β-strands are unzipped first,
followed by the others in the order of strength [18]. This can explain the existence
of the 6% trajectories. Furthermore, Liu et al. [18] carried out a series of simulations to compare the relative strength of networks according to both pathways. The
simulation results confirm that both the calculated possibility and the contour length
change (of a β-sheet being peeled off from β-crystallites) according to the fishnet
structure are in agreement with the AFM results [18]. This again indicates that the
β-crystallites in the SF nanofibrils adopt a nano-fishnet topology. Furthermore, it is
reported that a similar force pattern for recombinant spider eggcase silk fibrils has
been observed, suggesting the prevalence of the molecular fishnet structure in animal
silk nanofibrils of different types (Fig. 6.12e) [18]. It is noted that the adoption of
the fishnet structure is the natural choice for obtaining outstanding toughness in the
crystallite network. In fact, when using the Monte Carlo approach to simulate the
breaking stress of different network topologies, it was calculated that the fishnet
W. Qiu and X.-Y. Liu
spectra; the β-strands in the crystallites were aligned along the fibrous axis in silk
fibers but were perpendicular to the long axis of amyloid fibrils. Statistical results
show that most force-extension curves (94%) of RSF nanofibrils exhibit sequential unfolding events involving random peak forces; however, no clear trend was
observed. For the rest of the trajectories (6%), another characteristic pattern was
identified [18]. The highest unfolding force peak was followed by a series of events,
which shows a general upward trend in unfolding forces. Given the fact that hydrogen
bonds between β-strands and hydrophobic interactions between β-sheet layers are
responsible for the stability of β-sheets and β-crystallites, the two distinct trajectories
imply two possible pathways for unfolding β-strands from β-crystallites. Pathway 1
involves the sequential unzipping of β-strands directly from β-crystallites (without
affecting the inter-β-sheet interactions), whereas Pathway 2 involves the peeling of a
β-sheet off the crystallite first, followed by the unzipping of β-strands from the peeledoff β-sheet. In the case of Pathway 1, at each step of the unzipping of the β-strands,
the force is mostly applied onto the strand directly connected to the AFM tip, while
an equal amount of force is exerted on the other strands in the β-crystallites, owing
to the integrated response of the β-crystallites. Hence, the strands directly linked to
the AFM tip should experience a force much stronger than in other locations, so they
are unzipped beforehand. This is then repeated for the next adjacent strand. As the
strength of collective hydrogen bonding between different β-strands is different, the
corresponding force pattern should result in peaks with different heights. According
to the slab-segment model, the force distribution among β-strands in the β-sheet is
similar to that in β-crystallites; hence, the force pattern should be similar to Pathway
1 [36]. However, if the β-crystallites are cross-linked with each other, as seen in a
fishnet network, the peeled-out β-sheet will be stretched out between two anchor
points, i.e., the AFM tip and the remaining β-crystallites. In this condition, an equal
amount of stretching force is applied on all the β-strands within the peeled-off β-sheet
because they are connected in a series. Thus, the weakest β-strands are unzipped first,
followed by the others in the order of strength [18]. This can explain the existence
of the 6% trajectories. Furthermore, Liu et al. [18] carried out a series of simulations to compare the relative strength of networks according to both pathways. The
simulation results confirm that both the calculated possibility and the contour length
change (of a β-sheet being peeled off from β-crystallites) according to the fishnet
structure are in agreement with the AFM results [18]. This again indicates that the
β-crystallites in the SF nanofibrils adopt a nano-fishnet topology. Furthermore, it is
reported that a similar force pattern for recombinant spider eggcase silk fibrils has
been observed, suggesting the prevalence of the molecular fishnet structure in animal
silk nanofibrils of different types (Fig. 6.12e) [18]. It is noted that the adoption of
the fishnet structure is the natural choice for obtaining outstanding toughness in the
crystallite network. In fact, when using the Monte Carlo approach to simulate the
breaking stress of different network topologies, it was calculated that the fishnet