Fig 1.
Comparison between reduced (KTN) and oxidized (KOS) keratin biomaterials.
KTN can form disulfide linkages to produce a stable scaffold; but KOS cannot, due to the sulfonic acid modification of thiol groups. Consequently, the electrostatic properties are also altered, resulting in more negatively-charged KOS scaffolds, prone to more rapid hydrolytic degradation than KTN scaffolds. 10 mM NaOH solvent was used for wetting and soaking.
Fig 2.
The KTN extract is comprised of monomeric (A, A’), dimeric (AA’), and tetrameric (2(AA’)) keratins as well as keratin associated proteins (B) indicated by peaks corresponding to their respective molecular weights using gel filtration chromatography. One of the protein monomers detected by western blot technique is keratin-31 (K31). Both keratins and keratin associated proteins have free thiols (R-SH) that react with gold and form strong gold-sulfur bonds. Au-S covalent interactions, as well as some physisorption, enable the formation of a stable KTN monolayer.
Fig 3.
Gold surfaces treated with A) solvent only (10 mM NaOH) and with KOS (oxidized keratin) solution have relatively smooth profiles compared to those with B) KTN (reduced keratin) extract. C) Partial surface adsorption of KTN increased the roughness, Rq, whereas full coverage by overnight KTN incubation led back to a smoother surface. *p < 0.05 compared to each of the other groups.
Fig 4.
A) Wide-scan XPS spectra of gold surfaces treated with 10 mM NaOH solvent, KOS, and KTN. Overnight incubation of KTN led to no detectable Au signals. The inset graph shows that, compared to the solvent group, KOS has very similar concentration levels of carbon, nitrogen, oxygen, sulfur, and gold, while KTN has elevated amounts of protein elements (C, N, and O) but decreased Au. B) Near-scan analysis displays the formation of an amide (O = C-N) peak at 288 eV, corresponding to KTN protein deposition on gold. Unbound and gold-bound KTN thiols were also detected at 163.6 and 162.5 eV, respectively. Partial adsorption of KTN on gold shifted the Au4f peaks to slightly lower energies.
Fig 5.
Adsorption kinetics of KTN on gold analyzed through A) SPR and B) QCM-D methods. After establishing baseline readings using 10 mM NaOH solvent, KTN solution was flowed on gold sensor chips until almost full saturation. A) The irreversible adlayer at SPR response (Δn) = 1410 μRIU was retained after solvent, 2% SDS, and solvent washings, and its surface concentration (Γ) = 2.49 mg/m2. B) The normalized changes in dissipation factor (ΔD) for the adsorbate = 0.46 × 10–6 (red curve, using the right y-axis), suggest rigidity. The changes in frequency per overtone (Δf/n) of the adsorbed KTN = -12.08 Hz (green curve, using the left y-axis), correspond to a surface concentration (Γ’) = 2.14 mg/m2. A comparison between Γ (SPR) and Γ’ (QCM-D) showed statistical similarity (p = 0.4273).
Fig 6.
A) Representative SPR sensogram of KTN-KTN association and dissociation with 10 mM NaOH, generated by increasing concentrations of the analyte (0.6, 1.3, 2.6, 5.1, 10.2, and 22.9 μM) and fitted with Langmuir curves (red trace). B) Linear regression plot of kobs vs [KTN] was used to obtain the slope = ka, y-intercept = kd, and KD = kd/ka.
Fig 7.
KTN gel bulk degradation in vitro at A-B) constant pH = 7.4, and C-D) constant [KCl] = 154 mM for over a period of 28 days at 37°C. Faster degradation occurred at longer time points, lower [KCl], and higher pH levels.
Fig 8.
Association (a) and dissociation (d) electrostatic interaction profiles of BMP-2 and the KTN monolayer in PBS and in water. BMP-2 analytes were successively flowed, at incrementally increasing concentrations: A) in PBS (pH 7.4) at 0.2, 0.4, 0.9, 1.7, 3.5, and 6.9 μM, B) in PBS (pH 4.5) at 0.03, 0.05, 0.1, 0.2, 0.4, 0.9, and 1.7 μM, and C) in water (pH 7) at 0.03, 0.05, 0.1, 0.2, and 0.4 μM. KTN-BMP-2 electrostatic attraction was strongest in water (KD = 1.1 × 10–7 M). In the presence of PBS salts at physiological pH, binding association was slightly weakened (KD = 3.2 × 10–5 M). Acidification of the PBS eliminated any binding between BMP-2 and KTN.
Fig 9.
Rates of release of proteins out of bulk KTN gels.
A) Globular proteins diffused out of gels in the following order: negatively-charged albumin, neutral hemoglobin, and positively-charged lysozyme. At the 24-hr time point, the amount of lysozyme released was significantly lower (*p < 0.01) compared to that of albumin and hemoglobin. B) BMP-2 had a slow release profile in PBS (pH 7.4) medium, suggesting tight electrostatic association with the KTN matrix and BMP-2 intermolecular aggregation due to the salting-out effect. KTN bulk degradation was faster relative to the BMP-2 release.