Hydrolysis Processing Impacts SDP Biochemistry and Surface Wetting Properties
The architectural protein contaminant sericin was successfully extracted from the silkworm cocoons in which the resulting fibroin fiber was reacted under high heat, temperature and pressure over time in LiBr brine to produce the SDP hydrolysis product, which was then purified using tangential flow filtration. The resulting hydrolysate protein distribution was compared on an SDS-PAGE gel to regenerated fibroin solution. The results show that the hydrolysis processing significantly shifts the average molecular weight of fibroin from 200,000 Da to under 50,000 Da for SDP (Figure 1A). In addition, the molecular weight distribution was also decreased, and spans primarily between 10,000 – 100,000 Da. Of particular note was the absence of the fibroin light chain, which is located at 26,000 Da, in the SDP after hydrolysis processing. These results indicate that the fibroin light chain portion of the protein was also hydrolyzed during processing. Correspondingly, fibroin amino acid composition was also altered in SDP as a result of the hydrolysis process (Table 1) [32]. In particular, SDP was found to contain 40% less serine relative to fibroin control samples (Figure 1B), with a corresponding increase in observed levels of glycine and alanine (Figure 1C). Furthermore, cysteine levels fell below the limits of assay detection and corroborated with the disappearance of the fibroin light chain around 26 kDa, which is covalently attached to the fibroin heavy chain by cysteine-mediated disulfide bonding [33].
Table 1. Amino acid content (g AA / kg material) for native fibroin and SDP.
Amino Acid
|
Fibroin
|
Silk-Derived Protein (SDP)
|
L-Glycine
|
33.17%
|
36.28%
|
L-Alanine
|
27.62%
|
30.77%
|
L-Serine
|
13.69%
|
6.11%
|
L-Tyrosine
|
11.20%
|
12.06%
|
L-Valine
|
2.74%
|
3.24%
|
L-Aspartic acid
|
2.20%
|
2.79%
|
L-Glutamic Acid
|
1.85%
|
2.48%
|
L-Threonine
|
1.33%
|
0.69%
|
L-Phenylalanine
|
1.30%
|
1.47%
|
L-Isoleucine
|
0.92%
|
1.06%
|
L-Arginine
|
0.88%
|
1.03%
|
L-Isoleucine
|
0.92%
|
1.06%
|
L-Leucine
|
0.73%
|
0.82%
|
L-Tryptophan
|
0.59%
|
0.00%
|
L-Proline
|
0.54%
|
0.65%
|
L-Lysine
|
0.52%
|
0.34%
|
L-Histidine
|
0.33%
|
0.21%
|
L-Methionine
|
0.19%
|
0.00%
|
L-Cysteine
|
0.19%
|
0.00%
|
Notably, the changes in SDP composition had a profound impact on formulation surface wetting properties as demonstrated by adding 1.0% SDP (wt./vol.) to saline solution. As expected, when saline is placed on a hydrophobic wax surface the solution will remain self-contained as a droplet and unable to wet the underlying surface. However, when SDP is added the formulation will readily wet paraffin wax in response to applied mechanical force across the surface (Figure 1D), thus providing surfactant-like properties due to the proteins’ amphiphilic nature, which interacts with both hydrophobic surface and hydrophilic solution chemistries. Similarly, a saline drop will readily roll down an inclined paraffin wax surface at a 40° incline, while with the addition of 1.0% SDP to formulation will allow the solution to adhere and hang at a 90° angle without movement (Figure 1E).
Impact of the Hydrolysis Process on SDP Stability Profile Versus Regenerated Fibroin
Stability of SDP in water was further evaluated against regenerated fibroin solutions. SDP demonstrated no gelation, protein aggregation, or significant increases in viscosity even at protein concentrations up to 25% w/w (Figure 2A). SDP viscosity remained below 10 cP at protein concentrations where regenerated fibroin could no longer stay in solution, and SDP was capable of remaining homogeneous at concentrations exceeding 40% w/w and at a viscosity of 140 cP. In contrast, the viscosity of regenerated fibroin rises precipitously when solution concentration exceeds 15% w/w, and as anticipated these samples could not be concentrated above 25 % w/w without becoming insoluble. Collectively, these results clearly demonstrate that the process-related protein transformations described herein for the preparation of SDP are needed for the production of a highly-concentrated, low viscosity protein solution.
Next, dried films of both SDP and fibroin solution were produced to evaluate how the introduction of the hydrolysis process impacted post-drying properties of the material. This resulted in the formation of solid protein films that exhibited stark differences in appearance between the two starting solutions (Figure 2B). Specifically, the SDP material demonstrated a darker yellow translucency when compared to the transparent films produced from fibroin. In addition, the SDP material formed a dried surface that prevented the lower region of the volume from completely dehydrating, resulting in the film remaining partially dissolved. This was not the case for the fibroin material, which was completely dried and physically distorted due to inherent tensional effects due to drying. These results indicate significant changes to the SDP mechanical properties, and thus chemical interactions, as a result of hydrolysis processing.
To assess solubility as a function of hydrolysis, both dried material samples were reconstituted in water. For the SDP material the dried outer surface later was peeled off and weighed, while for the regenerated fibroin a portion of the material was broken off and weighed. For both samples, 20 mg were added to deionized water, and then vortexed at high speed for 10 minutes. Interestingly, the SDP material completely dissolved in water, while the regenerated fibroin material demonstrated minimal dissolution. These results indicate the material solubility was distinctly changed between the SDP and regenerated fibroin materials due to hydrolysis processing.
To further elucidate changes to protein secondary structure, samples from both film materials were processed using a humidified vacuum chamber (i.e., water-annealing) to instigate secondary β-sheet structures [30]. ATR-FTIR spectral analysis revealed that SDP and fibroin films produced similar IR signatures before processing; however, water annealing of the SDP material lacked β-sheet secondary structure formation as indicated by the absence of absorption peaks in the Amide I and II regions at 1624 and 1510 cm-1, respectively (Figure 2C). The secondary structure formation findings represent a significant difference in material composition of the two samples, which provides functional significance to the altered amino acid composition imbued by the hydrolysis processing of fibroin into SDP (Table 1).
To investigate the impact of reduced molecular weight on material stability, fibroin solution was enzymatically digested with trypsin enzyme for up to a 6 hours. Samples from different enzyme digestion time points were then subjected to sonication, which is a well characterized method to initiate β-sheet formation and cause fibroin solution gelation [29]. The digested fibroin samples exhibited accelerated gelation kinetics and denser appearing gels when compared to untreated fibroin controls (Figure 3A). Specifically, 1 hour of trypsin treatment induced gel formation by 40 minutes following sonication, while control samples treated with deactivated trypsin exhibited slower gelation kinetics with a reduced density appearance. In contrast, SDP showed no tendency toward instability during this time frame, evidenced by a minimal and unchanging absorbance at 550 nm. These data provide an indicator of secondary structure formation in regenerated fibroin or SDP solutions over time. Accumulating β-sheet formation, indicated by an increasing absorbance at 550 nm, occurred within 30 minutes after sonication of fibroin samples. Sonication post enzymatic cleavage of fibroin with trypsin enhanced instability and accelerated gel formation. Conversely, SDP exhibited no tendency towards secondary structure formation based on the absence of absorption at 550 nm. These results indicate that fractionation of fibroin by enzymatic cleavage, without amino acid transformation, are ineffective and in fact counter-productive to forestall β-sheet formation, instability, and gel formation.
The impact on the fibroin light chain, which is linked through a disulfide bond [33], was assessed by adding the reducing agent DTT to fibroin solution samples (Figure 3B). The presence of DTT decelerated instability as indicated by increasing 550 nm absorbance over time relative to untreated control samples. These effects were further pronounced with higher concentrations of DTT, but still ineffective to abolish instability. In contrast, SDP exhibited no tendency toward instability following sonication, indicated by an unchanging baseline absorbance, which was unaffected by the addition of DTT. The results demonstrate that the reduction of disulfide bonds in regenerated fibroin improves stability but does not prevent gel formation, whereas instability is abolished in SDP. Reduction of disulfide bridges with DTT slowed β-sheet and subsequent gel formation relative to untreated fibroin solution, but the material ultimately forms secondary β-sheet structures. In contrast, SDP solutions formed no secondary structures based on the lack of absorbance at 550 nm and remained stable over the testing period. These results demonstrate that the fibroin disulfide bridge participates in the mechanisms underlying regenerated fibroin instability, but their reduction is insufficient to maintain solution stability alone without the use of the added hydrolysis process.
Lastly, to assess the impact of heat on producing enhanced stability during hydrolysis, SDP was sonicated and the solution absorbance at 550 nm monitored longitudinally to compare the kinetics of gelation. The application of increasing heat to fibroin solution reduced absorbance and instability relative to non-heated control samples proportionally (Figure 3C). However, solution heating caused an increase in baseline absorbance at 550 nm, which increased with heating duration relative to non-heated controls (Figure 4D). In contrast, SDP exhibited no change in absorbance following sonication throughout the duration of the experiment. Furthermore, the duration of heat exposure from 30 to 120 minutes had an inversely proportional impact on basal solution absorbance and gelation time, and that absorbance at 550 nm continued to escalate in all of the heated regenerated fibroin solutions over time, but did not change from baseline in sonicated SDP solution samples (Figure 3D). All fibroin solution samples showed an increase in absorbance over time, indicating change in protein properties and demonstrate that heat treated samples were undergoing β-sheet formation and therefore becoming unstable. These results indicate that heat-mediated hydrolysis of fibroin in the absence of LiBr is insufficient to drive the necessary amino acid transformations that facilitate enhanced protein stability as seen with SDP.
SDP Inhibits Activation of the NF-kB Inflammatory Pathway
Previous research has demonstrated that fibroin affects both NF-kB driven inflammation and wound healing [5,6,22]; therefore, the propensity for SDP to activate the canonical NF-kB signaling pathway was investigated as a baseline for bioactivity. The nuclear transcription factor p65 is part of the NF-kB complex, which translocates into the cell nucleus upon activation to facilitate pro-inflammatory gene expression, including TNF-α and MMP-9 [34]. The NF-kB pathway was stimulated within HCLE cultures using TNF-α, and then treated with increasing doses of SDP. Following a 1 hour incubation period, both the cell nuclei and p65 were stained and assessed by fluorescent confocal microscopy. The results demonstrated that p65 was localized primarily to the cytoplasm for unstimulated and untreated control cultures, as expected for cells in a non-inflammatory state (Figure 4A). As expected, p65 staining was confined to the nucleus for untreated cell cultures that were challenged with TNF-α in the culture medium, indicating that activation of the NF-kB inflammatory pathway had taken place (Figure 4B). Interestingly, p65 staining for SDP-treated cells was largely confined to the cytoplasm and demonstrated a dose-dependent sequestration whereby less nuclear localization was exhibited with cells dosed with higher SDP concentrations (Figure 4C-D). This indicates that the SDP protein inhibits the NF-kB inflammatory response in HCLE cultures dose-dependently.
Further confirmation of NF-kB pathway activation was carried out by looking at gene expression of endogenous pro-inflammatory proteins, TNF-α and MMP-9, whose expression are driven primarily by activated NF-kB. It was observed that the addition of SDP caused no change in basal gene expression of TNF-α or MMP-9; however, stimulation with TNF-α evoked a significant rise in expression of both genes (Figure 4E). Importantly, treatment with SDP at the time of TNF-α stimulation evoked a 6-fold reduction in expression of both TNF-α and MMP-9, thereby demonstrating potent inhibition of NF-kB gene expression by TNF-α in the presence of SDP. These results corroborate with the earlier p65 immunofluorescent findings, and collectively support that SDP inhibits NF-kB pathway activation and its inflammatory signaling consequences.