biomolecules

Article
Effects of Ultraviolet Light Irradiation on Silk Fibroin Films
Prepared under Different Conditions
Sora Lee, Soo Hyun Kim, You-Young Jo, Wan-Taek Ju, Hyun-Bok Kim and HaeYong Kweon *

Sericultural and Apicultural Materials Division, National Institute of Agricultural Sciences, RDA,
Wanju-gun 55365, Korea; [email protected] (S.L.); [email protected] (S.H.K.); [email protected] (Y.-Y.J.);
[email protected] (W.-T.J.); [email protected] (H.-B.K.)
* Correspondence: [email protected]

Abstract: Silk fibroin (SF)-based materials are exposed to both natural and artificial ultraviolet (UV)
light during preparation or administration. However, the effects of UV irradiation on SF films
prepared under different conditions have not yet been described in detail. In this study, four SF films
with different molecular weight (MW) distribution were fabricated using SF solutions, which were
prepared by dissolving degummed SF for 0.5–24 h. We observed UV (365 nm) irradiation on SF films
induced the increase of yellowness and absorbance at 310 nm of SF films, indicating the formation
of new photo-products and di-tyrosine bonds by photo-oxidation. Due to di-tyrosine cross-links
between SF chains, UV-irradiated SF films were not fully dissociated in urea solution. In addition to
formation of new products, UV reduced the crystallinity of SF films by breaking hydrogen bonds of
β-sheet conformation. Unlike the UV-induced decomposition of physical interactions, UV did not
affect the covalent bonds (i.e., peptide bonds). Through these experiments, we could expect that SF
with higher MW was more susceptible and SF with lower MW was more resistant to UV-induced
photo-oxidation and photo-degradation. These results provide useful information about UV-induced
aging of SF-based materials under natural sunlight and UV irradiating conditions.






Keywords: silk fibroin film; ultraviolet aging; photo-oxidation; photo-degradation
Citation: Lee, S.; Kim, S.H.; Jo, Y.-Y.;
Ju, W.-T.; Kim, H.-B.; Kweon, H.
Effects of Ultraviolet Light Irradiation
on Silk Fibroin Films Prepared under
1. Introduction
Different Conditions. Biomolecules
2021, 11, 70. https://doi.org/
Silk fibroin (SF) is one of well-known natural protein materials obtained from silk-
10.3390/biom11010070 worms. The primary structure of SF is highly unique and composed of repetitive glycine-
alanine-glycine-alanine-glycine-serine (GAGAGS) [1]. These repeating sequences form the
Received: 30 October 2020 crystalline region with β-sheet conformation maintained by various molecular interactions
Accepted: 4 January 2021 such as hydrogen bonds, van der Waals interaction, and hydrophobic interaction. The
Published: 7 January 2021 crystalline regions contribute to the strength of SF. In contrast, amorphous regions of
relatively hydrophilic blocks composed of random/α-helix conformations contribute to
Publisher’s Note: MDPI stays neu- the elasticity of SF.
tral with regard to jurisdictional clai- Because of the high crystallinity of the degummed SF, SF has been generally regarded
ms in published maps and institutio- as water-insoluble polymer. Instead of pure water, several chaotropic solvents [2] have been
nal affiliations. used for solubilization and application of SF in a variety of forms such as films, sponges,
hydrogels, or coating materials [3]. During the solubilization process of degummed SF, the
structure of SF can be affected by various conditions such as the type of solvent, dissolution
Copyright: © 2021 by the authors. Li-
time, and temperature, resulting in differences in physical and chemical features of SF. For
censee MDPI, Basel, Switzerland.
example, different solvent systems that have different ionic forces can affect the molecular
This article is an open access article
weight (MW) and structure of SF [4]. Wang and colleagues compared the MW of SF
distributed under the terms and con- solutions dissolved in 9.3 M LiBr and Ajisawa’s reagent (CaCl2 /H2 O/EtOH (molar ratio
ditions of the Creative Commons At- 1/8/2)) by using Rouse-like rheological behavior, which enables the determination of
tribution (CC BY) license (https:// molecular weight of SF [2]. MW of LiBr-dissolved SF was 179 kDa, whereas MW of Ajisawa’s
creativecommons.org/licenses/by/ reagent-dissolved SF was 128 kDa, indicating that LiBr solvent system was milder. In fact,
4.0/). Yamada et al. also found that Ajisawa’s reagent caused degradation of the heavy chain of

Biomolecules 2021, 11, 70. https://doi.org/10.3390/biom11010070 https://www.mdpi.com/journal/biomolecules

Biomolecules 2021, 11, 70 2 of 11

SF [5]. In addition, the fast protein liquid chromatogram of Ajisawa’s reagent-dissolved

SF showed a strong peak indicating 450 kDa and weak shoulder peak indicating 150 kDa.
The peak of 150 kDa increased with the increase of dissolution time. Moreover, peak shift
from 150 kDa to 110 kDa and new peak formation indicating 16 kDa were observed when
dissolution time was 180 min [6]. Such a different MW distribution resulted in different
particle size, rheological properties, and degree of transparency. Meanwhile, according
to the literature, the gelling property of SFs triggered by ultrasonic treatment was poor
in Ajisawa’s reagent system compared to LiBr system, although the similar MW and
distribution of SFs were acquired by the control of dissolution time [7]. It can be expected
that these two solvent systems might affect not only MW distribution of SF but also its
structure of hydrophobic segments. Hence, the dissolution conditions must be set carefully
considering the target application fields.
In addition to various dissolution conditions, there are various external parameters
that can affect the physicochemical properties of SF, such as heat, oxygen, volatile organic
compounds, light, or absolute humidity [8]. Among them, ultraviolet (UV) light is a
highly important parameter because during preparation and administration, SF-based
fibers, medical devices, and additives for foods and cosmetics are generally exposed to
natural or artificial UV light. Especially, UV sterilization of SF-based biomaterials is the
required procedure before implantation [9]. Therefore, understanding the physicochemical
stabilities of SF-based material under UV light is essential for a wide range of applications.
For several decades, many researchers have tried to understand the UV-induced
photo-aging process of silk fabrics, solutions, and films. The accumulated results have
demonstrated that SF fiber is susceptible to UV-induced changes in color and bright-
ness [10,11] because SF molecules contain numerous aromatic amino acids such as tyrosine
(5.1%), tryptophan (0.1%), and phenylalanine (0.4%) [8]. In addition, it is revealed that
photo-destruction of the ordered structures by UV irradiation makes silk fibers more brittle
than those that are nontreated [12]. To reduce UV-induced weakening effect on mechan-
ical properties, researchers fabricated silk fibers coated with TiO2 as UV-absorbers and
confirmed more stable tensile properties [13]. Sionkowska and colleagues also found that
thin films from a mixture of chitosan and silk fibroin were affected by UV light with a
wavelength of 254 nm, showing the decrease of surface roughness that was detected by
atomic force microscopy [14].
Although photo-induced oxidation and degradation are highly important reactions for
using SF-based materials, to the best of our knowledge, the effect of dissolution time and
the resultant different MW distribution on the resistance of SF-based materials against the
photo-induced reaction have not been investigated in detail. Therefore, in this study, four
types of SF solutions were prepared by dissolving degummed SF fibers for different times
(0.5–24 h). Subsequently, SF films were fabricated and irradiated by UV light. In addition,
spectroscopic and structural properties of UV-irradiated SF films were investigated.

2. Materials and Methods

2.1. Preparation of SF Aqueous Solutions
Degumming of Bombyx mori cocoons was performed twice with 0.5% o.w.f. (on the
weight of fiber) of sodium oleate (C18 H33 NaO2 , Sigma Aldrich, MO, USA) and 0.3% o.w.f.
of solution sodium carbonate (Na2 CO3 , Sigma Aldrich) with a bath ratio of 1:50 at 100 ◦ C
for 1 h. The silk fibers were rinsed in warm water and air-dried at 50 ◦ C. The degummed
silk was dissolved in Ajisawa’s reagent solution containing calcium chloride (CaCl2 , Showa,
Japan), distilled water, and ethanol with molar ratio of 1:8:2 at 87 ◦ C for 0.5 h, 2 h, 8 h,
and 24 h. In 0.5 h-dissolution group, some undissolved fibers remained. However, no
residual fibers were observed in the other groups. Next, SF solutions were dialyzed using
a cellulose acetate tube (MWCO: 12–14 kDa) for 3 days. To remove insoluble components,
filtration processes were performed before and after dialysis. The final concentration of
aqueous SF solution was measured by weighing the remaining solid after drying at 130 ◦ C.
Biomolecules 2021, 11, 70 3 of 11

Six milliliters of 1.6 wt.% regenerated SF solutions were poured into 6-well plates and dried
in fume hoods to prepare regenerated SF films.

2.2. Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE) and Silver Staining
Protein concentration of each SF solution was measured using Bradford assay (Bio-Rad
laboratories, Hercules, CA, USA) and the solutions were mixed with 2X Laemmle buffer
(Bio-Rad) and 2.5% β-mercaptoethanol. Denatured SF samples at 95 ◦ C for 2 min were
loaded into the lanes of acrylamide gel (Bio-Rad) and electrophoresis was performed. Next,
the gel was stained using a silver staining kit (Invitrogen, Carlsbad, CA, USA) according
to the manufacturer’s guide. Using the analyze line graph tool of the image processing
software (ImageJ), the gray value profile was created along each lane of the gel image. The
representative profile of three replicates was selected and shown herein.

2.3. UV Irradiation on SF Films

SF films were irradiated in air at room temperature using a UV lamp (LF215LM,
UVITEC, Cambridge, UK) which emits light at a wavelength of 365 nm. The average
intensity of radiation was 2.37 ± 2.12 mW/cm2 at 5 cm from the light source. The dose of
radiation during 1 h exposure was 8538 ± 453 mJ/cm2 . The intensity of the incident light
was measured using an UV light meter (YK-35UV, Lutron, Taiwan). Irradiation experiments
were carried out on a well-plate and the location of films were rotated every 15 min for the
uniform exposure of light.

2.4. Color Measurement

Color change of SF films was measured using a spectrophotometer (Color i7, X-Rite,
Grand Rapids, MI, USA). A white standard color plate (L = 94.27, a = 0.13, and b = 2.80)
was used as a background and a standard for color measurements. Hunter color values (L:
lightness; a: redness; b: yellowness) were averaged from more than 10 measurements from
each sample. The total color differences (∆E) were calculated as follows:

∆E = [(∆L)2 + (∆a)2 + (∆b)2 ]0.5 (1)

where ∆L, ∆a, and ∆b are the difference between each color values of film specimen and
the standard.

2.5. UV–Vis Absorbance and Fluorescence Spectroscopy

The UV–vis absorption spectra of SF film before and after UV-irradiation were
recorded with a Multiskan GO plate reader (Thermo Scientific, Waltham, MA, USA) in
the range of 270 to 360 nm. Films were fixed inside of the cuvette cell. Absorbance of SF
molecules released from SF films during immersion in 6 M urea solution at 10 mg/mL
was also collected. The fluorescence intensity (excitation: 360/40 nm; emission: 460/40
nm) of the SF molecules in 6 M urea solution was measured by microplate reader (HT-
Synergy, Bio-Tek, Winooski, VT, USA). For film and solution samples, the measured
values of empty cuvette and the same volume of 6 M urea solution were used for blank
subtraction, respectively.

2.6. Attenuated Total Reflection Fourier Transform Infrared (ATR-FTIR) Spectroscopy

ATR-FTIR experiments were carried out and the spectra of silk films were recorded
with a Fourier transform spectrometer (S100, Perkin Elmer, Waltham, MA, USA) in the
range of 1000 to 4000 cm−1 . Thirty-two scans were averaged with resolution 4 cm−1 and
the experiments were repeated more than 6 times. Amide I and III crystallinity index were
determined by calculating the absorbance intensity ratios of 1620 cm−1 /1652 cm−1 for
amide I and 1264 cm−1 /1230 cm−1 for amide III crystallinity index.
Biomolecules 2021, 11, x FOR PEER REVIEW 4 of 11

Biomolecules 2021, 11, 70 determined by calculating the absorbance intensity ratios of 1620 cm−1/1652 cm−1 for amide
4 of 11
I and 1264 cm−1/1230 cm−1 for amide III crystallinity index.

2.7. Assay of 2,4,6-Trinitrobenzene Sulfonic Acid (TNBS)

2.7. Assay of 2,4,6-Trinitrobenzene Sulfonic Acid (TNBS)
To determine the relative content of primary amines in SF film, TNBS assay was per-
To determine
formed according tothethe
relative content
established of primary
protocol aminesmodification
with minor in SF film, TNBS
[15]. SFassay
filmswas
(2.5
performed according to the established protocol with minor modification
mg) were mixed with 250 μL of 200 mM sodium phosphate buffer (pH 8.2) and 250 μL of [15]. SF films
(2.5
0.5%mg)of were
TNBSmixed withThe
solution. 250 µL of 200mixture
reaction mM sodium was phosphate
heated at 40buffer (pH48.2)
°C for h inand 250 µL
a shaking
of 0.5% of TNBS solution. The reaction mixture was heated at 40 ◦ C for 4 h in a shaking
incubator. To stop the reaction, 250 μL of 6 N HCl was added into the mixture. Next, the
incubator.
mixture was Toautoclaved
stop the reaction,
at 120 °C250
for µL of 6 to
30 min Ndissolve
HCl wasany
added into the
insoluble mixture.
matter. Next,
The absorb-
the mixture was autoclaved at 120 ◦ C for 30 min to dissolve any insoluble matter. The
ance of 100 μL of mixture was read at 340 nm using a microplate reader (Multiskan GO)
absorbance
and all valuesof 100 of mixtureby
µLsubtracted
were was
theread at 340
value nm using
of blank a microplate
solution preparedreader (Multiskan
with same proce-
GO) and all values
dure without SF film. were subtracted by the value of blank solution prepared with same
procedure without SF film.
3. Results and Discussion
3. Results and Discussion
3.1. Characterization
3.1. Characterization of of Molecular
Molecular Weight
Weight Distribution
DistributionofofSF SFDissolved
Dissolvedfor forDifferent
DifferentTime
Time
Although Ajisawa’s
Although Ajisawa’s reagent
reagent hashas been
been usedused for
for dissolving
dissolving the the degummed
degummed SF SFfibers
fibers
with high utility, it can make damages in the intrinsic structure of SF,
with high utility, it can make damages in the intrinsic structure of SF, resulting in molecular resulting in molec-
ular scissions
scissions [5,6].[5,6]. As expected,
As expected, SDS-PAGE
SDS-PAGE resultsresults
shown shown in Figure
in Figure 1A indicate
1A indicate 0.5 h0.5
andh and
2h
2 hgroups
SF SF groupsthatthat
were were dissolved
dissolved forfor a relatively
a relatively short
short timeshowed
time showedstrong strongband
bandintensity
intensity
at high
at high MWMWregion
region(>100
(>100 kDa).
kDa). Unlike
Unlike the the low
low band
band intensity
intensity of of 0.5
0.5 hhSFSFgroup
groupbetween
between
100and
100 and3535kDa,
kDa,thethe band
band intensity
intensityforfor the
the 22 hh SF
SF group
group was was high
high between
between the the same
same range,
range,
indicating the wide distribution of MW for the 2 h SF group. In the case of the 8 hh SF
indicating the wide distribution of MW for the 2 h SF group. In the case of the 8 SF
group,
group, bandband intensity was gradually increased in the range from 180
gradually increased in the range from 180 kDa to the bottomkDa to the bottom of
thethe
of gel. For
gel. For the
the2424h hSFSFgroup,
group,which
whichwas wasdissolved
dissolved for for the
the longest time, aa strong
strong band
band
wasshown
was shownonlyonlyat atthe
thelower
lowerMW MWregion
region(<35 (<35kDa).
kDa). Molecular
Molecular distributions
distributionsof ofSF
SFgroups
groups
dissolved for
dissolved for different
different times
times were
were also
also observed
observed in in the
the gray
gray value
value profile
profileof ofeach
eachlane
laneof of
SDS-PAGEgel
SDS-PAGE gel(Figure
(Figure1B). 1B).Considering
ConsideringSF SFhad
hadaaheavy
heavychainchainofof391
391kDakDaandandaalight
lightchain
chain
of
of 26
26 kDa [16], we could could expect
expect the
thebands’
bands’upper upperregion
regionofof245 245kDakDaand and near
near region
region of of
25
25
kDakDa were
were contributed
contributed bybythe
theheavy
heavyand andlight
lightchain
chain of of SF, respectively.
respectively. Therefore,
Therefore, thethe
bands
bands from
from245245kDa
kDato to25
25kDa
kDaandandbelow
below25 25kDa
kDawere
werethought
thoughtto tobebethe
thedegraded
degradedheavyheavy
and
andlight
lightchain,
chain,respectively.
respectively.According
According totothe results
the resultsshowing
showing thetheweakweakband intensities
band of
intensities
the 8 and
of the 24 h24
8 and groups,
h groups,we could conclude
we could that dissolution
conclude for longer
that dissolution than 8 hthan
for longer could8 halmost
could
critically degradedegrade
almost critically the heavy thechains
heavyofchains
SF fibers.
of SF fibers.

Figure1.
Figure 1. SDS-PAGE
SDS-PAGEresults
resultsof
offour
foursilk
silkfibroin
fibroin(SF)
(SF)solutions.
solutions.(A)
(A)Silver-stained
Silver-stainedgel
gelimage
imageshowing
show-
ing difference
the the difference of SFs
of SFs in molecular
in molecular weight
weight distribution.The
distribution. Thelane
lane on
on the
the right
right side
side indicates
indicatesthe
the
molecular weight marker. The other four lanes represent SF with increasing dissolution time, from
left to right. (B) The gray value profiles of protein bands on the SDS-PAGE gel. The set of intensity
values is taken from each lane, from the top to the bottom, and plotted in the chart.
 3.2. Effect of UV Irradiation on Color Change of SF Films
The difference in MW distribution of SF can generally affect molecular structu
Biomolecules 2021, 11, 70 physicochemical properties. Given the SDS-PAGE results, we hypothesized 5 of 11 t
groups dissolved for different times would show different susceptibility to UV irrad
To confirm this, dose-dependent color changes were evaluated. Total color diffe
3.2. Effect
(∆E) of UV Irradiation
calculated from valueson Colorof Change
Hunter of L
SF(lightness),
Films a (redness), and b (yellowness)
SF films were plotted
The difference in MW with yellowness
distribution of SF can index (b) asaffect
generally a function
molecular ofstructure
UV irradiation
and ti
physicochemical properties. Given the SDS-PAGE results, we hypothesized
depicted in Figure 2, ∆E was not affected by UV irradiation time, regardless of disso that SF groups
dissolved for different times would show different susceptibility to UV irradiation. To
time. In contrast, yellowness index tended to increase with UV irradiation. Especia
confirm this, dose-dependent color changes were evaluated. Total color differences (∆E)
yellowing
calculated fromeffect on the
values 0.5 SFLgroup
of Hunter wasastronger
(lightness), (redness),thanand bother groups,
(yellowness) forshowing
the SF mo
a 2-fold increase of yellowness index after 4 h of UV irradiation. In contrast, the yello
films were plotted with yellowness index (b) as a function of UV irradiation time. As
depictedof
indexes in Figure
the other2, ∆Egroups
was not were
affected by UVthan
lower irradiation
the 0.5time, regardless
h group even of dissolution
after 4 h of irrad
time. In contrast, yellowness index tended to increase with UV irradiation. Especially,
Yellowing is considered as one of the most important photo-oxidation markers
the yellowing effect on the 0.5 SF group was stronger than other groups, showing more
teins.
than aThe development
2-fold of yellowindex
increase of yellowness colorafterof SF 4 hfilms
of UVmight be because
irradiation. aromatic
In contrast, the amin
produced
yellowness new indexesproducts containing
of the other groups were chromophores
lower than the 0.5 under UVeven
h group light [17].
after 4 h Utilizing
of
irradiation. Yellowing is considered as one of the most important
siproteomic approach, Dyer and colleagues identified chromophoric photo-produ photo-oxidation markers
in proteins.
rived fromThe development
tyrosine of yellow color and
and tryptophan, of SF proposed
films might be thebecause aromatic amino
photo-oxidation pathway
acids produced new products containing chromophores under UV light [17]. Utilizing a
Therefore, these color measurement results indicate the
quasiproteomic approach, Dyer and colleagues identified chromophoric photo-products
MW of SF can affect the su
bility
derivedto from
UV-induced
tyrosine and production
tryptophan,of andchromophores. We also expected
proposed the photo-oxidation pathways that UV irra
[18].
can be an these
Therefore, ecosustainable technique
color measurement resultsfor modifying
indicate the MWcolorof SF properties
can affect theof SF film with
suscepti-
bility to UV-induced production of chromophores. We also expected
ing chemical reagents. Similarly, Sagnella and colleagues modified the color prope that UV irradiation
can be an ecosustainable technique for modifying color properties of SF film without using
silk fiber and silk fibroin by doping RhoB in the artificial diet during silkworm
chemical reagents. Similarly, Sagnella and colleagues modified the color properties of silk
[19].
fiberWith
and silkthe increase
fibroin of feeding
by doping RhoB in time of RhoB-added
the artificial diet from
diet during silkworm 0 to[19].
raising 72 h,Withchromatic
increased and hue angle (h°) decreased, indicating the color
the increase of feeding time of RhoB-added diet from 0 to 72 h, chromaticity (C) increased change from between
and hue angle ◦
(h ) decreased, indicating
green and yellow components to anthe color change
almost from between
red-purple bluish-green and
component.
yellow components to an almost red-purple component.

A. 0.5 h B. 2h
3 3 3 3
Yellowness Yellowness
(normalized by std.)

(normalized by std.)

ΔE ΔE
Yellowness index
Yellowness index

2 2 2 2 ΔE
ΔE

1 1 1 1

0 0 0 0
0 1 2 3 4 0 1 2 3 4
UV irradiation time (h) UV irradiation time (h)

C. 8h D. 24 h
3 3 3 3
Yellowness Yellowness
(normalized by std.)

(normalized by std.)

ΔE ΔE
Yellowness index

Yellowness index

2 2 2 2
ΔE

ΔE

1 1 1 1

0 0 0 0
0 1 2 3 4 0 1 2 3 4
UV irradiation time (h) UV irradiation time (h)

Figure 2. Yellowness and total color differences (∆E) of UV-irradiated SF films (1–4 h) prepared with
Figure
four SF2. Yellowness
solutions anddissolved
that were total color
for differences
different times(ΔE) of h;
(A: 0.5 UV-irradiated SF24films
B: 2 h; C: 8 h; D: h) (n(1–4
> 10;h) prep
with ± SEM).
meanfour SF solutions that were dissolved for different times (A: 0.5 h; B: 2 h; C: 8 h; D: 24 h
10; mean ± SEM).
x FOR PEER REVIEW 6 of 11

Biomolecules 2021, 11, 70 6 of 11

3.3. Effect of UV Irradiation on UV–Vis and Fluorescence Spectroscopic Properties of SF Films
Changes of UV–vis absorbance spectra of SF films can detect the formation of new
photo-products. As3.3.
shown
Effect in Figure
of UV 3A–D,
Irradiation the main
on UV–Vis andpeak around
Fluorescence 280 nm is
Spectroscopic attributed
Properties to
of SF Films
aromatic amino acids that are present in SF molecular chains such as tyrosine (5.1%),
Changes of UV–vis absorbance spectra of SF films can detect the formation of new phe-
nylalanine (0.4%), and tryptophanAs(0.1%)
photo-products. shown[1,8]. After
in Figure UVthe
3A–D, irradiation, light absorbance
main peak around at
280 nm is attributed
to aromatic amino acids that are present in SF molecular chains
310 nm tended to increase for all SF groups. These results are probably due to the for- such as tyrosine (5.1%),
phenylalanine (0.4%), and tryptophan (0.1%) [1,8]. After UV irradiation, light absorbance
mation of new photo-products by UV irradiation such as new cross-links and/or an oxi-
at 310 nm tended to increase for all SF groups. These results are probably due to the
dized form of aromatic amino
formation acids.
of new Similarly, Sionkowska
photo-products and such
by UV irradiation colleagues also observed
as new cross-links and/or an
the changes in absorbance of SFofsolutions
oxidized form at around
aromatic amino 300–310Sionkowska
acids. Similarly, nm and they explained
and colleagues alsothat
observed
the changes in
3,4-dihydroxylphenylalanine absorbance
(DOPA) of SF solutions
might be formed at around 300–310 nm and they
by photo-oxidation explainedbe-
reaction that 3,4-
dihydroxylphenylalanine (DOPA) might be formed by photo-oxidation reaction between
tween phenylalanine and tyrosine, affecting the absorbance property [1].
phenylalanine and tyrosine, affecting the absorbance property [1].

A. 0.5 h B. 2h
1.2 0.20
UV 1.2 0.20
UV
OD (normalized by A280)

OD (normalized by A280)
0h 0h
1.0 0.15 1h 1.0 0.15 1h
0.8
0.10 2h 0.8
0.10 2h
0.05 3h 0.05 3h
0.6 4h 0.6 4h
0.00 0.00
300 305 310 315 320 300 305 310 315 320
0.4 0.4

0.2 0.2

0.0 0.0
270 290 310 330 350 270 290 310 330 350
Wavelength (nm) Wavelength (nm)

C. 8h D. 24 h
1.2 0.20
UV 1.2 0.20
UV
OD (normalized by A280)

OD (normalized by A280)

0h 0h
1.0 0.15
1h 1.0 0.15
1h
0.8
0.10 2h 0.8
0.10 2h
0.05 3h 0.05 3h
0.6 4h 0.6 4h
0.00 0.00
300 305 310 315 320 300 305 310 315 320
0.4 0.4

0.2 0.2

0.0 0.0
270 290 310 330 350 270 290 310 330 350
Wavelength (nm) Wavelength (nm)

Figure 3. Absorbance spectra of UV-irradiated SF films (1–4 h) prepared with four SF solutions that were dissolved for
Figure 3. Absorbance spectra of UV-irradiated SF films (1–4 h) prepared with four SF solutions
different times (A: 0.5 h; B: 2 h; C: 8 h; D: 24 h).
that were dissolved for different times (A: 0.5 h; B: 2 h; C: 8 h; D: 24 h).
To know the presence of newly formed cross-links between SF chains, we immersed
To know the presenceinof6 M
SF films urea solution
newly formedatcross-links
10 mg/mL and wait for SF
between 24 h. Becausewe
chains, it isimmersed
well known that
concentrated urea solution can easily break hydrogen bonds in SF, we expected SF films
SF films in 6 M urea solution at 10 mg/mL and wait for 24 h. Because it is well known that
would be dissolved. In fact, intact SF films were almost dissolved, remaining few solid-
concentrated urea solution can easily
state components, breakinhydrogen
as shown Figure 4A.bonds
However,in with
SF, wethe expected
increase of SF
UV films
irradiation
would be dissolved.time,
In fact,
larger amounts of undissolved and swollen films were observed for allsolid-
intact SF films were almost dissolved, remaining few SF groups,
state components, as shownSFinchains
indicating Figure
were4A. However, with
photo-cross-linked the increase
by chemical of UV of
bonds instead irradiation
hydrogen bonds.
time, larger amounts of undissolved and swollen films were observed for all SF groups,
indicating SF chains were photo-cross-linked by chemical bonds instead of hydrogen
bonds. The chemical bonds formed by UV irradiation are thought to be di-tyrosine bonds.
 Biomolecules 2021, 11, 70 7 of 11

x FOR PEER REVIEW 7 of 11

The chemical bonds formed by UV irradiation are thought to be di-tyrosine bonds. To

confirm the formation of di-tyrosine bonds in released SF molecules, fluorescence intensity
405 nm was detected when samples
at 440–480 were excited
nm was measured at 320 nm
at the excitation due to di-tyrosine
wavelength of 320–380 nm.bonds [20].to the
According
Similarly, for all SFliterature,
samples,forthe
UVfluorescence wasthe
irradiated insulin, increased with
progressive the increase
formation of a peakofaround
UV irra- 405 nm
was detected when samples were excited at 320 nm due to di-tyrosine
diation time, indicating the formation of di-tyrosine bonds (Figure 4B). These di-tyrosine bonds [20]. Similarly,
for all SF samples, the fluorescence was increased with the increase of UV irradiation
bonds might be formed upon tyrosyl radical isomerization and enolization reaction [21].
time, indicating the formation of di-tyrosine bonds (Figure 4B). These di-tyrosine bonds
It is especially worth nothing
might thatupon
be formed SF groups with higher
tyrosyl radical MW showed
isomerization a higher
and enolization increase
reaction [21]. It is
of fluorescence upon increase
especially of UV
worth irradiation
nothing time. with
that SF groups Thishigher
fluorescence
MW showed profile is similar
a higher increase of
fluorescence
to the color measurement upon
result in increase
Figure of UV irradiation
2A–D and filmtime. This fluorescence
morphologies shown profile is similar to
in Figure
the color measurement result in Figure 2A–D and film morphologies shown in Figure 4A.
4A. Therefore, we can conclude the MW of SF can influence the susceptibility to UV in
Therefore, we can conclude the MW of SF can influence the susceptibility to UV in terms
terms of formation of offormation
new photo-products. These MW
of new photo-products. Theseeffects might
MW effects be be
might related
relatedto
tothe con- of
the content
tent of aromatic amino acids
aromatic because
amino acids longer
because heat
longer treatment timetime
heat treatment could damage
could damagemoremorearo-
aromatic
matic residues during dissolution
residues of degummed
during dissolution SF fibers.
of degummed SF fibers.

Figure 4. Morphologies of the remaining four SF films (A) and fluorescence intensity of the released SF molecules from
Figure 4. Morphologies of the remaining four SF films (A) and fluorescence intensity of the re-
SF films (B) after immersion of films in 6 M urea solution for 24 h (ex. 360/40; em. 460/40) (n = 3; mean ± SD; 1-fold: 0 h
leased SF molecules from SF films (B) after immersion of films in 6 M urea solution for 24 h (ex.
UV irradiation).
360/40; em. 460/40) (n = 3; mean ± SD; 1-fold: 0 h UV irradiation).
3.4. Effect of UV Irradiation on Structural Stability of SF Films
3.4. Effect of UV Irradiation on Structural Stability of SF Films
In addition to formation of new chemical bonds by UV, the high energy of UV can
In addition to induce
formation of new chemical
photo-degradation bonds
of hydrogen byand
bonds UV, the high
disulfide bonds.energy
Therefore,of UV can
we evaluated
by FT-IR the structural stability of four different SF films
induce photo-degradation of hydrogen bonds and disulfide bonds. Therefore, we evalu- after exposure to UV. FT-IR is very
useful tool for understanding the chemical and physical structure of silk proteins. In the
ated by FT-IR the structural stability of four different SF films after exposure to UV. FT-
literature, a semiquantitative approach was chosen by calculating the relative intensities
IR is very useful tool for understanding
of amide bands to know the theextent
chemical and physical
of SF degradation structure
[8]. The ratios ofoftwosilk pro-band
amide
teins. In the literature, a semiquantitative approach was chosen by calculating
intensities that indicate random/α-helix and β-sheet conformations have been used as the relative
intensities of amide one of degradation
bands to knowmarkers by comparing
the extent the crystallinity.
of SF degradation [8].Therefore,
The ratios we ofacquired
two FT-
IR spectra and calculated the ratio of 1620 cm−1 (β-sheet) to 1652 cm−1 (random coil)
amide band intensities that indicate random/α-helix and β-sheet conformations have been
for amide I crystallinity index (CI) and 1264 cm−1 (β-sheet) to 1230 cm−1 (random coil)
used as one of degradation
for amide IIImarkers by comparing
CI. The FT-IR the crystallinity.
spectra of nonirradiated Therefore,
and 4 h irradiated we ac-
SF films showed
quired FT-IR spectra and calculated
different absorbance the ratio of
intensities 1620
at the cm wavenumbers
−1
specific (β-sheet) to 1652 (cm cm
− 1 −1
), which(random
indicated β-
coil) for amide I crystallinity index coil
sheet and random (CI)structures
and 1264 cm−1 5).
(Figure (β-sheet) to of
In the case 1230 cm−1prepared
SF films (random bycoil)
a shorter
for amide III CI. The dissolution time (0.5of
FT-IR spectra h and 2 h SF groups),
nonirradiated the 4absorbance
and h irradiated valuesSFatfilms
1652 and 1230 cm−1
showed
indicated random coil structure increased after 4 h UV irradiation, whereas the values at
different absorbance intensities at the specific wavenumbers (cm−1), which indicated β-
1620 and 1264 cm−1 indicated β-sheet structure decreased or increased slightly. On the
sheet and random coil otherstructures
hand, in the(Figure 5).8In
case of the the 24
h and case ofgroups,
h SF SF films prepared
there by a shorter
were no significant changes
dissolution time (0.5 h and 2 h SF groups), the absorbance values at 1652 and 1230 cm−1
indicated random coil structure increased after 4 h UV irradiation, whereas the values at
1620 and 1264 cm−1 indicated β-sheet structure decreased or increased slightly. On the
other hand, in the case of the 8 h and 24 h SF groups, there were no significant changes in
 ture was thought to be affected by breakage of hydrogen bonds involved in β-she
formation (Figure 6A). In addition, this result can explain the previously reported
Biomolecules 2021, 11, 70 ened mechanical properties of aged silk fibers under UV exposure [12,13]. 8 of Meanwh
11

2 h SF group showed a slight increase of amide I CI by 1 h UV irradiation and a g

decrease after 1 h (Figure 6B). The newly formed di-tyrosine cross-links could domi
contribute to an increase
in the absorbance values, thusof indicating
CI. However, when theEspecially,
both structures. dose of UV bothlight was
values accumulat
at 1264
and 1230 cm −1 of the 24 h SF group increased similarly after UV irradiation, meaning
decomposition effects of UV might reduce the crystallinity by breaking hydrogen
there were less compositional changes of β-sheet and random coil that occurred with UV
like the 0.5 SF group. For the 8 h and 24 SF groups, the CI values were not drama
irradiation. Moreover, as shown in Figure 6, we could compare the calculated CI values
changed
upon the (Figure 6C,D).
irradiation We could
time. Like expectforthat
the tendency SF chains
formation with higher the
of photo-products, MW 0.5were
SF exten
affected by UV
group showed energy
a strong and the
decrease physical
of both amide interactions
I and III CI. Thewere easilyofbroken
destruction because a
crystalline
structureof
number was thought tobonds
hydrogen be affected
were byinvolved
breakage of inhydrogen
long SF bonds
chainsinvolved
than ininshorter
β-sheet chains
conformation (Figure 6A). In addition, this result can explain the previously reported
From the TNBS assay results, we could not observe any remarkable differences
weakened mechanical properties of aged silk fibers under UV exposure [12,13]. Meanwhile,
content
the 2 h SFofgroup
ε-amine
showed group
a slightorincrease
newlyofformed
amide I CIN-terminal amine groups
by 1 h UV irradiation (Figure 7).
and a gradual
results
decreaseindicate
after 1 h that UV6B).
(Figure did
Thenot induce
newly chain
formed scission
di-tyrosine and decrease
cross-links MW for all SF g
could dominantly
contribute to an increase of CI. However, when the dose of UV light
Because peptide bonds are covalent, unlike the crystallinity that is maintained was accumulated, the by ph
decomposition
molecular effects of UV
interactions, might reduce
stronger energythe may
crystallinity by breaking
be needed hydrogen
to break bonds, bonds
the peptide
like the 0.5 SF group. For the 8 h and 24 SF groups, the CI values were not dramatically
sidering that several other studies reported the molecular degradation by UV [22
changed (Figure 6C,D). We could expect that SF chains with higher MW were extensively
with higher
affected by UV energy
energyor longer
and irradiation
the physical timewere
interactions thaneasily
the light
broken used for aour
because study mi
larger
duce
numberchain scissions
of hydrogen andwere
bonds decrease theinMW
involved long of
SF SF.
chains than in shorter chains.

A. Amide I region B. Amide III region

1652 1620 0h 1264 1230 0h
4h 4h
0.5 h

1680 1660 1640 1620 1600 1280 1260 1240 1220 1200

1652 1620 0h 1264 1230 0h

4h 4h
2h

1680 1660 1640 1620 1600 1280 1260 1240 1220 1200

1652 1620 0h 1264 1230 0h

4h 4h
8h

1680 1660 1640 1620 1600 1280 1260 1240 1220 1200

1652 1620 0h 1264 1230 0h

4h 4h
24 h

1680 1660 1640 1620 1600 1280 1260 1240 1220 1200

Wavenumber (cm−1) Wavenumber (cm−1)

5. FT-IR
Figure 5.
Figure FT-IRspectra
spectraof SFof films nonirradiated
SF films and irradiated
nonirradiated with UV light
and irradiated withforUV
4 h;light
(A) amide
for 4Ih; (A) am
region (1680–1600 cm −1 ), (B) amide III region (1280–1200 cm−1 ).
region (1680–1600 cm ), (B) amide III region (1280–1200 cm ).
−1 −1
moleculesBiomolecules FOR11,PEER
2021, 11, x2021, 70 REVIEW 9 of 11 9o

Figure 6. IAmide
Figure 6. Amide I and
and III III crystallinity
crystallinity index
index calculated from
calculated fromFT-IR spectra
FT-IR of UV-irradiated
spectra SF films prepared
of UV-irradiated SF filmswith four SF with four
prepared
solutions that were dissolved for different times (A: 0.5 h; B: 2 h; C: 8 h; D: 24 h) (n > 6; mean
SF solutions that were dissolved for different times (A: 0.5 h; B: 2 h; C: 8 h; D: 24 h) (n > 6; mean ± SEM). ± SEM).

From the TNBS assay results, we could not observe any remarkable differences in the
content of ε-amine group or newly formed N-terminal amine groups (Figure 7). These
0.5
A. results indicate h UV did not induce
that B. 2 and
chain scission h decrease MW for all SF groups.
0.8 0.8
Because peptide bonds are covalent, unlike the crystallinity that is maintained by physical
Absorbance at 340 nm

Absorbance at 340 nm

molecular interactions, stronger energy may be needed to break the peptide bonds. Consid-
0.6 that several other studies reported
ering 0.6the molecular degradation by UV [22], light with
higher energy or longer irradiation time than the light used for our study might induce
0.4 scissions and decrease the MW of
chain 0.4
SF.

0.2 0.2

0.0 0.0
0 1 2 3 4 0 1 2 3 4
UV irradiation time (h) UV irradiation time (h)

C. 8h D. 24 h
0.8 0.8
Absorbance at 340 nm

Absorbance at 340 nm

0.6 0.6

0.4 0.4

0.2 0.2
Figure 6. Amide I and III crystallinity index calculated from FT-IR spectra of UV-irradiated SF
Biomolecules 2021, 11, 70
films pre
10 of 11
SF solutions that were dissolved for different times (A: 0.5 h; B: 2 h; C: 8 h; D: 24 h) (n > 6; mean ± SEM).

A. 0.5 h B. 2h
0.8 0.8

Absorbance at 340 nm

Absorbance at 340 nm
0.6 0.6

0.4 0.4

0.2 0.2

0.0 0.0
0 1 2 3 4 0 1 2 3 4
UV irradiation time (h) UV irradiation time (h)

C. 8h D. 24 h
0.8 0.8
Absorbance at 340 nm

Absorbance at 340 nm
0.6 0.6

0.4 0.4

0.2 0.2

0.0 0.0
0 1 2 3 4 0 1 2 3 4
UV irradiation time (h) UV irradiation time (h)

Figure 7. Absorbance at 340 nm obtained by 2,4,6-trinitrobenzene sulfonic acid (TNBS) assay with
Figure 7. Absorbance
UV-irradiated SF films preparedatwith
340four
nmSFobtained bySFs
solutions. The 2,4,6-trinitrobenzene
were dissolved for differentsulfonic
times aci
(A: 0.5 h; B: 2 h; C: 8 h; D: 24 h) (n = 3; mean ± SD).
UV-irradiated SF films prepared with four SF solutions. The SFs were dissolv
(A: 0.5 h; B: 2 h; C: 8 h; D: 24 h) (n = 3; mean ± SD).
4. Conclusions
In this study, investigation on the effect of UV irradiation on the physicochemical
properties of SF films prepared under different conditions was carried out. UV irradiation
on SF films not only formed of new bonds by photo-oxidation, but also damaged crys-
tallinity of SF by photo-degradation. These UV-induced photo-reactions were confirmed
by measuring color, UV–vis absorbance spectra, fluorescence, or FT-IR spectra. In addition,
we could observe that SF with higher MW was more susceptible to these UV irradiation
effects. To explain the exact mechanism about MW effect on photo-resistance of SF films,
further studies in the molecular levels may be required. In conclusion, we expect that our
study would provide useful information for preparing and applying SF-based materials
under natural sunlight or UV irradiating conditions.

Author Contributions: S.L., conceptualization, formal analysis, investigation, visualization, and

writing—original draft preparation; S.H.K., resources; Y.-Y.J., methodology and validation; W.-T.J.
and H.-B.K., supervision, validation, and writing—review and editing; H.K., conceptualization,
supervision, project administration, funding acquisition. All authors have read and agreed to the
published version of the manuscript.
Funding: This work was carried out with the support of “Research Program for Agriculture Sci-
ence and Technology Development (Project No. PJ01502201)” Rural Development Administration,
Republic of Korea.
Biomolecules 2021, 11, 70 11 of 11

Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Sionkowska, A.; Planecka, A. The influence of UV radiation on silk fibroin. Polym. Degrad. Stabil. 2011, 96, 523–528. [CrossRef]
2. Wang, Q.; Chen, Q.; Yang, Y.; Shao, Z. Effect of various dissolution systems on the molecular weight of regenerated silk fibroin.
Biomacromolecules 2013, 14, 285–289. [CrossRef] [PubMed]
3. Koh, L.-D.; Cheng, Y.; Teng, C.-P.; Khin, Y.-W.; Loh, X.-J.; Tee, S.-Y.; Low, M.; Ye, E.; Yu, H.-D.; Zhang, Y.-W. Structures, mechanical
properties and applications of silk fibroin materials. Prog. Polym. Sci. 2015, 46, 86–110. [CrossRef]
4. Cheng, G.; Wang, X.; Tao, S.; Xia, J.; Xu, S. Differences in regenerated silk fibroin prepared with different solvent systems: From
structures to conformational changes. J. Appl. Polym. Sci. 2015, 132. [CrossRef]
5. Yamada, H.; Nakao, H.; Takasu, Y.; Tsubouchi, K. Preparation of undegraded native molecular fibroin solution from silkworm
cocoons. Mater. Sci. Eng. C 2001, 14, 41–46. [CrossRef]
6. Cho, H.J.; Ki, C.S.; Oh, H.; Lee, K.H.; Um, I.C. Molecular weight distribution and solution properties of silk fibroins with different
dissolution conditions. Int. J. Biol. Macromol. 2012, 51, 336–341. [CrossRef]
7. Kim, H.; Song, D.; Kim, M.; Ryu, S.; Um, I.; Ki, C.; Park, Y. Effect of silk fibroin molecular weight on physical property of silk
hydrogel. Polymer 2016, 90, 26–33. [CrossRef]
8. Koperska, M.; Pawcenis, D.; Bagniuk, J.; Zaitz, M.; Missori, M.; Łojewski, T.; Łojewska, J. Degradation markers of fibroin in silk
through infrared spectroscopy. Polym. Degrad. Stabil. 2014, 105, 185–196. [CrossRef]
9. Okahisa, Y.; Narita, C.; Yamada, K. Fabrication and characterization of a novel silk fibroin film with UV and thermal resistance.
Mater. Today Comm. 2020, 101630. [CrossRef]
10. Vilaplana, F.; Nilsson, J.; Sommer, D.V.; Karlsson, S. Analytical markers for silk degradation: Comparing historic silk and silk
artificially aged in different environments. Anal. Bioanal. Chem. 2015, 407, 1433–1449. [CrossRef]
11. Liang, Z.; Zhou, Z.; Dong, B.; Wang, S. Fabrication of Superhydrophobic and UV-Resistant Silk Fabrics with Laundering Durability
and Chemical Stabilities. Coatings 2020, 10, 349. [CrossRef]
12. Liu, H.; Zhao, S.; Zhang, Q.; Yeerken, T.; Yu, W. Secondary structure transformation and mechanical properties of silk fibers by
ultraviolet irradiation and water. Text. Res. J. 2019, 89, 2802–2812. [CrossRef]
13. Aksakal, B.; Koç, K.; Yargı, Ö.; Tsobkallo, K. Effect of UV-light on the uniaxial tensile properties and structure of uncoated and
TiO2 coated Bombyx mori silk fibers. Spectrochim. Acta A 2016, 152, 658–665. [CrossRef] [PubMed]
14. Sionkowska, A.; Płanecka, A. Surface properties of thin films based on the mixtures of chitosan and silk fibroin. J. Mol. Liq. 2013,
186, 157–162. [CrossRef]
15. Kale, R.; Bajaj, A. Ultraviolet spectrophotometric method for determination of gelatin crosslinking in the presence of amino
groups. J. Young Pharm. 2010, 2, 90–94. [CrossRef]
16. Zhou, C.Z.; Confalonieri, F.; Jacquet, M.; Perasso, R.; Li, Z.G.; Janin, J. Silk fibroin: Structural implications of a remarkable amino
acid sequence. Proteins 2001, 44, 119–122. [CrossRef]
17. Sashina, E.; Bochek, A.; Novoselov, N.; Kirichenko, D. Structure and solubility of natural silk fibroin. Russ. J. Appl. Chem. 2006, 79,
869–876. [CrossRef]
18. Dyer, J.; Bringans, S.; Bryson, W. Characterisation of photo-oxidation products within photoyellowed wool proteins: Tryptophan
and tyrosine derived chromophores. Photochem. Photobiol. Sci. 2006, 5, 698–706. [CrossRef]
19. Sagnella, A.; Chieco, C.; Di Virgilio, N.; Toffanin, S.; Posati, T.; Pistone, A.; Bonetti, S.; Muccini, M.; Ruani, G.; Benfenati, V.; et al.
Bio-doping of regenerated silk fibroin solution and films: A green route for biomanufacturing. RSC adv. 2014, 4, 33687. [CrossRef]
20. Correia, M.; Neves-Petersen, M.T.; Jeppesen, P.B.; Gregersen, S.; Petersen, S.B. UV-light exposure of insulin: Pharmaceutical
implications upon covalent insulin dityrosine dimerization and disulphide bond photolysis. PLoS ONE 2012, 7, e50733. [CrossRef]
21. Kerwin, B.A.; Remmele, R.L., Jr. Protect from light: Photodegradation and protein biologics. J. Pharm. Sci. 2007, 96, 1468–1479.
[CrossRef] [PubMed]
22. Sionkowska, A.; Płanecka, A.; Lewandowska, K.; Michalska, M. The influence of UV-irradiation on thermal and mechanical
properties of chitosan and silk fibroin mixtures. J. Photochem. Photobiol. B Biol. 2014, 140, 301–305. [CrossRef] [PubMed]