Xue Li1,2, Xingmei Qi3, Yu-Ming Cai4, Yuan Sun2, Rui Wen2, Rui Zhang5, Jan Johansson6, Qing Meng2, Gefei Chen6. 1. Department of Medical Ultrasound, Shanghai Tenth People's Hospital, Ultrasound Research and Education Institute, Tongji University Cancer Center, Shanghai Engineering Research Center of Ultrasound Diagnosis and Treatment, Tongji University School of Medicine, 200092 Shanghai, China. 2. Institute of Biological Sciences and Biotechnology, Donghua University, 201620 Shanghai, China. 3. The Jiangsu Key Laboratory of Infection and Immunity, Institutes of Biology and Medical Sciences, Soochow University, Suzhou 215123, China. 4. Institute for Life Sciences, University of Southampton, SO17 1BJ Southampton, Hampshire, U.K. 5. Department of Pulmonary Circulation, Shanghai Pulmonary Hospital, Tongji University School of Medicine, Shanghai 200433, China. 6. Department of Biosciences and Nutrition, Karolinska Institutet, 14157 Huddinge, Sweden.
Abstract
Spider flagelliform silk shows the best extensibility among various types of silk, but its biomimetic preparation has not been much studied. Herein, five customized flagelliform spidroins (FlSps: S and NTDFl-Sn-CTDFl, n = 1-4), in which the repetitive region (S) and N-/C- terminal domains (NTDFl and CTDFl) are from the same spidroin and spider species, were produced recombinantly. The recombinant spidroins with terminal domains were able to form silk-like fibers with diameters of ∼5 μm by manual pulling at pH 8.0, where the secondary structure transformation occurred. The silk-like fibers from NTDFl-S4-CTDFl showed the highest tensile strength (∼250 MPa), while those ones with 1-3 S broke at a similar stress (∼180 MPa), suggesting that increasing the amounts of the repetitive region can improve the tensile strength, but a certain threshold might need to be reached. This study shows successful preparation of flagelliform silk-like fibers with good mechanical properties, providing general insights into efficient biomimetic preparations of spider silks.
Spider flagelliform silk shows the best extensibility among various types of silk, but its biomimetic preparation has not been much studied. Herein, five customized flagelliform spidroins (FlSps: S and NTDFl-Sn-CTDFl, n = 1-4), in which the repetitive region (S) and N-/C- terminal domains (NTDFl and CTDFl) are from the same spidroin and spider species, were produced recombinantly. The recombinant spidroins with terminal domains were able to form silk-like fibers with diameters of ∼5 μm by manual pulling at pH 8.0, where the secondary structure transformation occurred. The silk-like fibers from NTDFl-S4-CTDFl showed the highest tensile strength (∼250 MPa), while those ones with 1-3 S broke at a similar stress (∼180 MPa), suggesting that increasing the amounts of the repetitive region can improve the tensile strength, but a certain threshold might need to be reached. This study shows successful preparation of flagelliform silk-like fibers with good mechanical properties, providing general insights into efficient biomimetic preparations of spider silks.
Orb-weaving
spiders produce up to seven types of silks with various
mechanical properties for different biological purposes (Figure A).[1−3] Spider silk as a biomaterial possesses outstanding mechanical properties,[1,4] and excellent biocompatibility and biodegradability, holding great
potential in biomedicine.[5,6] Due to spiders’
innate territorial behavior and limited amounts of silks from a spider
web, it is not realistic to obtain spider silk on a large scale via
farming spiders. Recently, expressing trimmed spider silk proteins
(spidroins) followed by silk-like fiber preparation has become the
prevalent strategy for generating artificial spider silk-like fibers.[4,5] Among different spider silks, dragline silk and flagelliform silk
have attracted considerable attention attributed to their best tensile
strength and extensibility, respectively.[2]
Figure 1
Orb-web
spider silks and spinning strategy in this study. (A) Seven
types of silk from orb-web spider, corresponding spidroins, and biological
applications. (B) Scanning electron microscopy (SEM) images of native
spider flagelliform silk generated in this study. The scale bars are
10 μm (left) and 2 μm (right), respectively. (C) Facile
manual pulling strategy utilized for spider flagelliform silk-like
fiber preparation from a recombinant protein droplet at pH 8.0.
Orb-web
spider silks and spinning strategy in this study. (A) Seven
types of silk from orb-web spider, corresponding spidroins, and biological
applications. (B) Scanning electron microscopy (SEM) images of native
spider flagelliform silk generated in this study. The scale bars are
10 μm (left) and 2 μm (right), respectively. (C) Facile
manual pulling strategy utilized for spider flagelliform silk-like
fiber preparation from a recombinant protein droplet at pH 8.0.Each spider silk consists of a certain type of
spidroins, which
are responsible for specific mechanical properties and performance.
For instance, dragline silk is made up of major ampullate spidroins
(MaSps) with poly (A) motifs contributing
to strength by forming β-sheet crystal structures,[2] while flagelliform silk is composed of flagelliform
spidroins (FlSps) that are rich in GPGGX (X is mainly for S or Y)
motifs, resulting in intriguing extensibility, possibly through spring
structures.[3,7] The FlSp contains a large core repetitive
region franked with NTDFl (FlSp N-terminal domain) and
CTDFl (FlSp C-terminal domain). The core repetitive region
is made up of 11 repetitive units that are built by iterations of
the GPGGX, GGX (X is mainly for S, Y, or A), and hydrophobic spacer
motifs.[7] These functional motifs confer
flagelliform silk the best extensibility among other spider silks
as well as excellent strength, and it has been shown that the GPGGX,
GGX, and spacer motifs contribute to extensibility, toughness, and
strength, respectively, in synthetic silk-like fibers.[8−10] While several studies have looked into the preparation of artificial
flagelliform silk, only a few chimeric FlSp/MaSp spidroins were successfully
constructed for producing spider silk-like fibers,[11−14] which might be due to the fact
that FlSp contains highly repetitive units rich in Gly and Pro, making
it difficult to be expressed in exogenous systems. Meanwhile, the
mechanism of molecular self-assembly or the structural/functional
investigations of FlSp remains largely missing, in particular the
terminal domains, NTDFl and CTDFl. NTD and CTD
are highly conserved with regard to the tertiary structure, where
NTD adopts a five α-helix structure that dimerizes into an antiparallel
dimer at low pH[15−23] and CTD is a parallel dimer with α-helical bundles.[24−28] NTD allows spidroin storage in silk gland and progressively forms
stable dimers in a pH-dependent manner.[18,21] CTD, on the
other hand, can unfold and assemble into amyloid-like fibrils which
can then trigger silk formation or regulate silk formation and improve
mechanical properties.[25,27,28] To obtain high-quality artificial spider silk-like fibers, it is
probably essential to include both NTD and CTD. Indeed, some designed
spidroins, such as mini-MaSp[29,30] and mini-AcSp (aciniform
spidroin),[31] have been recombinantly generated
and spun into spider silk-like fibers with decent mechanical properties
without using an organic solvent. Furthermore, the terminal domains,
in particular the CTD from different spidroins, display various roles
in the silk formation process; for example, CTD from MaSp is important
for the correct structure of nanofibers,[32] while AcSp CTD is important for maintaining silk mechanical properties
and uniquely modulates silk fiber properties.[24,33] Hence, customized spidroins with the repetitive regions fused with
its natural terminal domains might benefit the biomimetic preparation
of specific type of silk-like fibers.Here, five customized
spidroins [S, and NTDFl-Sn-CTDFl, (NTSnCT) with n = 1–4] were generated
from the full-length FlSp of spider Araneus ventricosus, that is, the NTDFl, repetitive region S, and CTDFl are all from the same spidroin, FlSp, and the same spider
species (A. ventricosus). The customized
spidroins were recombinantly produced in Escherichia
coli and spun into spider silk-like fibers directly
from physiological buffer at pH 8.0. Furthermore, mechanical properties
and secondary structure transformation from soluble protein solutions
to solid silk-like fibers were investigated by tensile testing, circular
dichroism (CD) spectroscopy, and attenuated total reflection Fourier-transform
infrared (ATR-FTIR) spectroscopy. This study has prepared artificial
spider flagelliform silk-like fibers via a facile manual-method which
simultaneously covers FlSp-derived repetitive region and teriminal
domains, and the results contribute to the development of novel potential
biomaterials.
Materials and Methods
Plasmid
Construction
Five plasmids that code for A.
ventricosus FlSp spidroins, S, NS1C, NS2C, NS3C,
and NS4C, were constructed, resulting in customized spidroins with
different numbers of repetitive region S optionally flanked with A. ventricosus FlSp terminal domains—NTDFl and CTDFl. The NTDFl, repetitive region
S and CTDFl amino acid sequences are shown in Figure . The DNA sequences
for A. ventricosus FlSp NTD, CTD, and
S were obtained through screening the previously constructed A. ventricosus genomic library[34] and further confirmed by phylogenetic analysis. Polymerase
chain reaction primers used in this study are listed in Table S1. The gene fragment encoding the repetitive
region S was optimized for efficient expression in E. coli BL21. All the constructs were confirmed by
sequencing.
Figure 2
Architecture of different customized FlSps and SDS-PAGE analysis
of recombinant preparations. (A) Amino acid sequences of the Flag
repetitive region S, NTDFl, and CTDFl used in
this study. (B) Architecture of customized FlSps (S, NS1C, NS2C, NS3C,
and NS4C). NTDFl and CTDFl indicate N- and C-
terminal domains from A. ventricosus flagelliform spidroin, respectively. S stands for the repetitive
region trimmed from A. ventricosus flagelliform
spidroin. (C) SDS-PAGE analysis of the purified customized FlSps—S
(right), NS1C, NS2C, NS3C, and NS4C (left). Lane M is the protein
marker.
Architecture of different customized FlSps and SDS-PAGE analysis
of recombinant preparations. (A) Amino acid sequences of the Flag
repetitive region S, NTDFl, and CTDFl used in
this study. (B) Architecture of customized FlSps (S, NS1C, NS2C, NS3C,
and NS4C). NTDFl and CTDFl indicate N- and C-
terminal domains from A. ventricosus flagelliform spidroin, respectively. S stands for the repetitive
region trimmed from A. ventricosus flagelliform
spidroin. (C) SDS-PAGE analysis of the purified customized FlSps—S
(right), NS1C, NS2C, NS3C, and NS4C (left). Lane M is the protein
marker.
Recombinant Customized
FlSp Spidroin Preparation
Customized
FlSps were expressed in E. coli BL21.
A single colony of each construct was cultured in 10 mL of LB medium
with 100 μg/mL ampicillin at 37 °C overnight and then transferred
to 1 L of fresh LB medium with 100 μg/mL ampicillin. When OD600 reached 0.6–0.8, the incubation temperature was
cooled down to 25 °C and 0.3 mmol/L (final concentration) isopropyl
β-d-1-thiogalactopyranoside was added. Cells
were incubated for another 16 h and collected by centrifugation at
4500 rpm for 15 min and further resuspended in 30 mL of 20 mmol/L
Tris pH 8.0. The cells were lysed by a pressure homogenizer JN-3000
plus (JNBIO) under 1 200 bar. The supernatant was collected after
centrifuged at 10 000 rpm (4 °C) for 40 min and applied to Ni-NTA
column, followed by a 1 h incubation. The column was washed with 20
mmol/L Tris pH 8.0 containing 5 mmol/L imidazole, and target proteins
were subsequently eluted out by 20 mmol/L Tris pH8.0 with 100 mmol/L
imidazole that was eventually removed by dialysis against 20 mmol/L
Tris pH 8.0.
CD Spectroscopy
Customized FlSps
with a final concentration
of 0.5 mg/mL in 20 mmol/L phosphate buffer (pH 7.5, 6.5, and 5.5)
were subjected to CD measurements. For each protein, the samples at
different pHs were diluted from the same protein stock to maintain
the same concentration. CD spectra were recorded in 1 mm path length
quartz cuvettes at 25 °C from 260 to 190 nm in a J-810 spectropolarimeter
(JASCO, Japan) at room temperature, and the main parameters were as
follows: wavelength step 0.5 nm, response time 1 s, and bandwidth
1 nm. The individual spectrum shown for each was the average from
three continuous scans with background (20 mmol/L phosphate buffer)
subtracted.
Spider Silk-like Fiber Generation
As previously reported,[35] spider silk-like
fibers were manually drawn
by hand (manual pulling) from recombinant customized FlSp solutions
in 20 mmol/L Tris pH 8.0 (∼1 mg/mL) with a syringe needle (26G)
at room temperature, as shown in Figure C. Briefly, 100 μL of recombinant spidroin
solution was dropped onto a glass slide, and spider silk-like fibers
were pulled from the solution with a continuous speed of ∼10
mm/s.
Scanning Electron Microscopy
The native flagelliform
silk collected from the web of wild A. ventricosus and artificial spider silk-like fibers spun from customized FlSp
spidroins were fixed onto SEM plates and coated with gold for 45 s.
Images were taken by a SEM system (Phenom-WorldBV) with 10 kV acceleration
voltage at room temperature.
Mechanical Property Test
Mechanical
properties of silk-like
fibers were tested with UTM T150 (Agilent). Prior to mechanical tests,
the uniform thickness of the individual silk-like fiber was confirmed
via light microscopy, during which the diameters of different silk-like
fibers were measured. For strain–stress measurements, silk-like
fibers were fixed into a 1 cm frame and mechanical test was performed
with starting load strength being 750 μN at 20 °C and 50%
humidity. For each type of the silk-like fibers, 10 samples were supposed
to be tested; however, some of them were broken before applying the
starting load strength, which subsequently lead to seven (for NS1C,
NS2C, and NS4C) or nine (for NS3C) valid tests for each type of silk-like
fibers. Statistical analysis was conducted by using one-way analysis
of variance (ANOVA), followed by Tukey’s post-hoc test for
multiple comparisons. *p < 0.05, **p < 0.01.
ATR-FTIR Analysis
Nicolet 6700 spectroscopy
system
with ATR fitting (Thermo Fisher) was used to record ATR-FTIR spectra
of silk-like fibers from 600 to 4 000 cm–1 at room
temperature.[12,29,31] Sixty-four scans were collected for each spectrum with a resolution
of 4 cm–1. The background spectrum of a blank was
subtracted. Each type of the silk-like fiber was analyzed three times
(for each type, a mass of silk-like fibers was fixed for recording
ATR-FTIR spectra, as a single silk-like fiber was not suitable for
ATR-FTIR measurement). The transmittance (T) was
transformed into absorbance (A) according to the
function A = 2–log(% T).
Spectral decomposition of the amide I region ranging from 1 600 to
1 700 cm–1 was performed using the peak-fit program
(Cranes Software International Ltd.), and secondary structure composition
was subsequently evaluated. A seven-point Savitsky-Golay second-derivative
function was implemented to separate overlapping bands in the amide
I region, which was followed by curve-fitting (Gaussian peak shape).
Before curve fitting, a linear baseline was subtracted in the amide
I band from 1600 to 1700 cm–1. Different secondary
structures present distinct characteristic peaks in the amide I region
(R2 values are nearly 1); characteristic peak near 1 639
cm–1 is for random coil, 1 656 cm–1 for α-helix, 1 669 cm–1 for β-sheet,
and 1 684 and 1 698 cm–1 for β-turn.[31]
Results
Preparation of Customized A. ventricosus FlSps
The FlSp repetitive
region (S) containing typical
FlSp motifs was selected to fuse with its native NTDFl and
CTDFl (Figure A,B). The repetitive region S is rich in Gly (55%) (Figure S1A), and the Gly content increased when
more repetitive region S were included. With 10% of Ile, Leu, and
Val, the repetitive region S showed several hydrophobic stretches;
however, hydrophilic patches were also observed (Figure S1B). The customized A. ventricosus FlSps containing different numbers of repetitive region S with and
without NTDFl and CTDFl were designated as S,
NS1C, NS2C, NS3C, and NS4C (Figure B). These five customized FlSps were successfully expressed
in E.coli and mainly present in the
soluble fraction after cell lysis and centrifugation. After Ni-NTA
affinity column purification, the purity of customized FlSps reached
90% as judged by sodium dodecyl sulfate-polyacrylamide gel electrophoresis
(SDS-PAGE) (Figure C). The final yields for S, NS1C, NS2C, NS3C, and NS4C were ∼10,
∼70, ∼66, ∼25, and ∼13 mg/L, respectively,
indicating that the terminal domains can promote the recombinant preparation,
whereas the length of the repetitive region is inversely proportional
to the final yield.
Secondary Structure of Customized FlSps at
Different pHs
As pH gradient is necessary for spider silk
formation, hence, we
investigated pH effects on the potential secondary structure transformation
of customized FlSps by CD spectroscopy at different pHs (7.5, 6.5,
and 5.5, Figure ).
At pH 7.5, the recombinant repetitive region S mainly adopted random
coil and partially helical structures, which did not display obvious
changes when the pH decreased to 5.5 (Figure E). All the other customized FlSps (NS1C,
NS2C, NS3C, and NS4C) showed overall α-helix conformation, indicated
by two typical negative peaks at 208 and 222 nm (Figure A–D). However, the strong
α-helix signal could be probably derived from the NTDFl and CTDFl domains because NTD and CTD domains from other
spidroins mainly consist of five α-helixes and these two domains
are highly conserved among different types of spidroin.[18,19,21,24,25,27,28] With pH decreased to 5.5, the overall shapes of the
CD spectra of NS1C, NS2C, and NS3C were not affected significantly,
although lower amplitude was observed (Figure A–C), which might result from the
decline of the amount of protein in solution induced by acidification.
Interestingly, pH decrease did not substantially influence the amplitude
of NS4C (Figure D),
which could be probably due to that NS4C predominantly consists of
the core repetitive region (four S regions) which maintained its secondary
structure at different pHs (Figure E).
Figure 3
CD spectra of customized FlSps under different pHs. CD
spectra
of customized FlSps in 20 mM phosphate at different pHs (7.5, 6.5,
and 5.5) were recorded at room temperature. (A) NS1C. (B) NS1C. (C)
NS3C. (D) NS4C. (E) S. The Y axis is given by mdeg.
CD spectra of customized FlSps under different pHs. CD
spectra
of customized FlSps in 20 mM phosphate at different pHs (7.5, 6.5,
and 5.5) were recorded at room temperature. (A) NS1C. (B) NS1C. (C)
NS3C. (D) NS4C. (E) S. The Y axis is given by mdeg.
Spider Silk-like Fiber Preparation from Customized
FlSps and
Characterization
To prepare flagelliform silk-like fibers
from the recombinant customized FlSps, a straightforward spinning
method was pursued, by which spider silk-like fibers were directly
generated by manual pulling from different customized FlSp solutions
with a syringe needle (Figure C). The recombinant single repetitive region S was soluble
and could not assemble into silk-like fibers, at least by direct manual
pulling from 20 mmol/L Tris pH 8.0. Interestingly, all the other customized
FlSps with the terminal domains were able to form long solid silk-like
fibers (>10 cm) as “spun” by manual pulling at pH
8.0,
suggesting that the terminal domains are essential for silk formation
under the spinning conditions. Neutral pHs are used for spidroin storage
in the spider silk gland, and acidic pHs are essential for spider
silk protein solidification.[28] The observations
here suggest that pH is probably not the only determinant for solid
silk formation from soluble spidroins. Compared to native spider flagelliform
silk (diameter as ∼2 μm), all silk-like fibers pulled
from NS1C, NS2C, NS3C, and NS4C droplets displayed similar diameters
(2–5 μm) (Figures B and 4C). These silk-like fibers all
presented smooth surface morphology and no obvious grooves or pores
were observed under SEM (Figure ).
Figure 4
SEM observation of different silk-like fibers from customized
FlSps.
Spider silk-like fibers were prepared via the manual-pulling method
and observed under SEM. (A) NS1C. (B) NS1C. (C) NS3C. (D) NS4C. The
scale bars are 10 μm (upper panel) and 2 μm (lower panel),
respectively.
SEM observation of different silk-like fibers from customized
FlSps.
Spider silk-like fibers were prepared via the manual-pulling method
and observed under SEM. (A) NS1C. (B) NS1C. (C) NS3C. (D) NS4C. The
scale bars are 10 μm (upper panel) and 2 μm (lower panel),
respectively.
Mechanical Properties and
Secondary Structures of Silk-like
Fibers
Spider silk-like fibers derived from customized FlSps
(NS1C, NS2C, NS3C, and NS4C) were subjected to mechanical property
assessment at room temperature with 50% humidity. The mechanical properties
are shown in Figures and S2. All the silk-like fibers showed
outstanding tensile strength (>180 MPa, Figure A,B). The largest protein studied here, NS4C
fibers broke at the highest stress, 254 MPa on average (Figure A,B). In addition, the silk-like
fiber from the NS4C fiber also showed the best toughness (24 MJ/m3) and Young’s modulus (10 GPa) (Figure C), suggesting a positive relationship between
the molecular weight (the number of repetitive region S) and the corresponding
mechanical properties. However, there were no significant differences
of the maximum tensile strength between NS1C, NS2C, and NS3C (Figure B), indicating there
is probably a molecular weight threshold for positive correlations
between molecular weight and mechanical properties. In terms of extensibility,
no obvious differences were noticed among the different silk-like
fibers, and the average extension for all the silk-like fibers reached
10–12% (Figure C), which is considerably lower compared to the native spider flagelliform
silks (>200%).[7]
Figure 5
Mechanical property of
silk-like fibers from the customized FlSps.
(A) Stress–strain curves of silk-like fibers from NS1C, NS2C,
NS3C, and NS4C. (B) Tensile strength comparison, where ns is for no
significant difference and ** for p < 0.01. (C)
Mechanical properties of silk-like fibers. The mean values were calculated
from independent tests of each silk type, and data were shown as mean
± standard deviation.
Mechanical property of
silk-like fibers from the customized FlSps.
(A) Stress–strain curves of silk-like fibers from NS1C, NS2C,
NS3C, and NS4C. (B) Tensile strength comparison, where ns is for no
significant difference and ** for p < 0.01. (C)
Mechanical properties of silk-like fibers. The mean values were calculated
from independent tests of each silk type, and data were shown as mean
± standard deviation.Generally, during spider silk formation, spidroins will experience
structure transformation from disordered and partly helical structures
to β-sheets and “amorphous” conformations.[36] Here, the secondary structures of spider silk-like
fibers were explored with ATR-FTIR. Despite roughly consistent ATR-FTIR
spectra of silk-like fibers from NS1C, NS2C, NS3C, and NS4C (Figure ), the ATR-FTIR spectrum
of NS1C silk-like fiber presented a slightly left-shift peak near
1630 cm–1 compared to other types of silk-like fibers,
which suggests that it contains different secondary structure compositions,
for example, more β-sheet structure.[29] Furthermore, spectral decomposition of the amide I region from 1600
to 1700 cm–1 was performed with peak-fit software
to evaluate the secondary structure composition (Figure S3). Compared to other types of silk-like fiber, NS1C
silk-like fiber adopted more β-sheet structure (43%) but less
β-turn conformation (27%) (Table S2). Although NS1C silk-like fibers contained the highest amount of
β-sheet structure, it did not show a relative better tensile
strength compared to other silk-like fibers. The silk-like fibers
from NS4C broke at the highest stress point, but the secondary structure
profile evaluated from ATR-FTIR spectrum was similar to that of silk-like
fibers from NS2C, NS3C (Table S2), suggesting
that the mechanical properties cannot be perfectly reflected by secondary
structure composition and there might be other determinants. The results
here support that recombinant spidroins with larger molecule weight
generally were leading to silk-like fibers with better strength.[31,37]
Figure 6
Secondary
structure evaluation of silk-like fibers from customized
FlSps by ATR-FTIR.
Secondary
structure evaluation of silk-like fibers from customized
FlSps by ATR-FTIR.
Discussion
In
this study, five customized FlSps, S, NS1C, NS2C, NS3C, and
NS4C, with terminal domains and repetitive region from the same spidroin
and spider species, were successfully produced recombinantly. The
repetitive region S was soluble and could not assemble into a silk-like
fiber induced by manual pulling at pH 8.0, while all the other customized
FlSps formed spider silk-like fibers with outstanding strength and
decent extensibility.Spider flagelliform silk owns the best
extensibility among different
types of spider silk, which could have unique application potential,
for example, cardiac tissue engineering applications.[38] The repetitive region and terminal domains of spidroin
function differently,[8,14,18,21,27,28] therefore, in this study besides the single repetitive
region S, four customized FlSps containing NTDFl, S, and
CTDFl which are all from A. ventricosus FlSp were generated for biomimetically preparing spider flagelliform
silk-like fibers. These recombinant customized FlSps all showed considerable
expression yield, even though the molecular weight of the largest
customized FlSp NS4C reached 96 kDa (Figure ). All the recombinant customized FlSps were
in the soluble fraction during purification, and the purified proteins
also showed excellent solubility. For the customized FlSps with NTDFl, high solubility is expected because NTD is extremely soluble
and has been implemented to produce membrane proteins, enzymes, and
hydrophobic amyloid prone peptides.[39−44] Interestingly and surprisingly, the recombinant single repetitive
region S without NTDFl and CTDFl was also present
in the soluble fraction. It has been shown that repetitive regions
from MaSp,[37] MiSp,[26] TuSp (tubuliform spidroin),[45] and PySp[46] were expressed in inclusion body and needed
to undergo denaturing and refolding processes, and the results here
suggest that different repetitive regions from various spidroins have
different biochemical properties.Without NTDFl and
CTDFl, recombinant S did
not self-assemble into silk-like fibers by direct pulling at pH 8.0,
indicating that terminal domains, NTDFl and CTDFl, are essential for spidroin assembly. However, this study does not
exclude the possibility that the single repetitive region S could
form fibers under some other specific conditions. The detailed structure/function
correlations of N-/C-terminal domains of FlSp are poorly understood,
although they are conserved between different spidroins.[4] The C-terminal domains from MaSp and AcSp both
adopt an identical α-helical structure and can fold into dimer
conformations, but they play distinct functional roles. For instance,
a recombinant designed spidroin MaSp [(AQ)12NR3] containing
MaSp CTD formed regular well-defined fibers, whereas (AQ)24 only without the CTD formed random aggregates;[27] for the CTD from AcSp, it mainly contributes to mechanical
properties of synthetic fibers, and no difference was found between
the macro morphologies of fibers with or without CTD.[24,25] The repetitive domain S mainly maintained random coil and partially
helical conformations (Figure E), which is quite different from that of other spidroin repetitive
regions, for example, AcSp repetitive domain contains seven[47] or five α-helixes[48] depending on the spider species, while TuSp repetitive domain has
six α-helixes.[49] This suggests that
the silk formation mechanism of FlSp can be different from other spidroins.[47] Hence, for customized spidroin designing, the
NTD, repetitive region, and CTD from the same spidroin and even the
same spider species can benefit a specific type of biomimetic silk-like
fiber production.Chimeric mini-spidroins have shown the capability
to be spun into
silk-like fibers by wet spinning or hand-pulling method at different
pHs. NTMaSp-2Rep-CTMiSp was spun to silk-like
fibers at pH 5.0, where high protein concentration was used (100–500
mg/mL).[29] Similarly, NMaSp2-R7-CMaSp1 at 300 mg/mL was spun into silk-like fibers at pH 5.0.[30] Chimeric spidroins NTMaSp-Rep-CTMiSp at 100–300 mg/mL were spun into silk-like fibers
from pH 2.0 to 11.[50] Chimeric spidroins
WAcSp2CMaSp4CTMaSp and WAcSp2CMaSp8CTMaSp formed silk-like fibers by hand-pulling
at pH 7.5 at rather low concentration (0.4 mg/mL). Recently, silk-like
fibers were pulled manually from NTMaSpRMaSp/FlSpnCTMaSp (n indicates the number of R in combination of
MaSp and FlSp repeats)[11] and from NTMaSpWAcSpnCTMiSp at concentrations of
(0.8–1.0 mg/mL) at pH 8.0.[31] It
is also interesting that silk-like fibers can be drawn from only two
repetitive regions from AcSp, WAcSp2 (20–200 μmol/L)
at pH 7.5.[48] In this study, the customized
FlSp with terminal domains here could form silk-like fibers at rather
low concentrations (∼1 mg/mL) via manual pulling at pH 8.0
(Figure ). Taken together,
the above phenomenon suggests that pH is probably not the sole determinant,
but there are other factors that can affect spidroin-silk transformation.The silk-like fibers from customized FlSps, NS1C, NS2C, NS3C, and
NS4C displayed excellent tensile strength, ranging from 180 to 250
MPa, and reasonable extensibility of 10–12% (Figure C). Silk-like fibers pulled
from chimeric spidroin NTMaSpWAcSpnCTMiSp (n = 1–4) with the manual pulling method
presented a tensile strength range from 51 to 245 MPa,[31] while silk-like fibers from NTMaSpWAcSp1CTMiSp (45.8 kDa) presented ultimate
tensile strength of 51 MPa, extensibility of ∼12%, a Young’s
modulus of 2.8 GPa, and a toughness of 4.8 MJ/m3. Silk-like
fibers from the smallest customized FlSp NS1C (44 kDa) in this study
broke at a maximum strength of 180 MPa, a maximum extensibility of
10%, a high Young’s modulus of 6 GPa, and a toughness of 14
MJ/m3 (Figure C). The mechanical properties of silk-like fibers from customized
FlSps are also comparable to synthetic silk-like fibers produced via
wet-spinning strategy. Recombinant fibers from chimeric spidroins
NTMaSp-RepMaSp-CTMiSp (44 kDa) showed
a tensile strength of about 50 MPa but a high extensibility of ∼90%.[50] The tensile strength of synthetic fibers from
chimeric spidroins NTMaSp-2RepMaSp-CTMiSp (33 kDa) reached 162 MPa and these silk fibers had an extensibility
of 37%.[29] The customized FlSp silk-like
fibers in the current study revealed an outstanding tensile strength
comparable to the wet-spinning silk-like fibers, but a relatively
lower extensibility, which was not expected. As spider flagelliform
silk is the most elastic among different types of spider silk, contributed
by its spidroin amino acid composition, these silk-like fibers spun
from different FlSp versions would be expected to have an excellent
extensibility. This discrepancy can probably result from the spinning
method that is very different from the spider spinning system and
the wet-spinning method used in literature where shear force, spinning
rate control, highly concentrated dope, and coagulating bath (normally
at low pH) were all involved.[29] On the
other hand, the amino acid sequence of repetitive S selected and the
limited size of recombinant customized FlSps are also possible explanations
for the weak extensibility compared to the native fibers, as the size
of spidroin can affect fibril properties.[31,37] The main secondary structures of the customized FlSps in phosphate
buffer were α-helix and random coil, while the dominant secondary
structures of the silk-like fibers were β-sheet and β-turn.
These results are consistent with the previous study, where the solid
silk formation is accompanied by structure transformation of spidroins.[1] However, the secondary structure content of the
flagelliform silk-like fibers is significantly different from the
native flagelliform silk,[9] which may explain
the difference in mechanical properties of the recombinant flagelliform
silk-like fibers in this study. For future artificial silk-like fiber
preparation, the spinning process and/or the protein constructs should
promote the final product to recapitulate corresponding native silks
to obtain high-performance artificial fibers.
Summary
Recombinant
spidroin-like proteins are normally dissolved in an
organic solvent to make dopes for spinning artificial spider silks,
whereas customized spidroins make it possible that artificial spider
silk-like fibers can be spun from biological buffers at different
pHs. To our knowledge, this study is the first to design customized
FlSps with employing the NTD, repetitive region, and CTD from the
same spidroin and same spider species. The customized FlSps were able
to form artificial spider silk-like fibers with outstanding mechanical
strength and decent extensibility from “spidroin” solutions
with a rather low concentration (∼1 mg/mL) at pH 8.0. Our results
shed light on the fact that FlSp terminal domains are essential for
silk formation, and the new customized FlSp-derived silk-like fibers
hold application potentials as novel biomaterials.
Authors: Florence Teulé; Bennett Addison; Alyssa R Cooper; Joel Ayon; Robert W Henning; Chris J Benmore; Gregory P Holland; Jeffery L Yarger; Randolph V Lewis Journal: Biopolymers Date: 2011-10-20 Impact factor: 2.505
Authors: Sherry L Adrianos; Florence Teulé; Michael B Hinman; Justin A Jones; Warner S Weber; Jeffery L Yarger; Randolph V Lewis Journal: Biomacromolecules Date: 2013-05-22 Impact factor: 6.988