Structure-property relationships of meta-kerateine biomaterials derived from human hair.

The structure-property relationships of kerateine materials were studied by separating crude hair extracts into two protein sub-fractions, referred to as α- and γ-kerateines, followed by their de novo recombination into meta-kerateine hydrogels, sponges and films. The kerateine fractions were characterized using electrophoresis and mass spectrometry, which revealed that the α-fraction contained complexes of type I and type II keratins and that the γ-fraction was primarily protein fragments of the α-fraction along with three proteins of the KAP-1 family. Meta-kerateine materials with increased amounts of γ-kerateines showed diminished physical, mechanical and biological characteristics. Most notably, materials with higher γ-content formed less elastic and less solid-like hydrogels and sponges that were less hydrolytically stable. In addition, a model biological assay showed that meta-kerateine films with greater amounts of γ-kerateines were less supportive of hepatocyte attachment. Investigation into the mechanism of attachment revealed that hepatocyte adhesion to meta-kerateines is not mediated by the ß1 integrin subunit, despite the presence of LDV binding motifs within the type I α-keratins. This work to define the role of protein composition on biomaterial function is essential for the optimization of keratin biomaterials for biomedical applications.

Mechanical and biological properties of keratose biomaterials.

The oxidized form of extractable human hair keratin proteins, commonly referred to as keratose, is gaining interest as a biomaterial for multiple tissue engineering studies including those directed toward peripheral nerve, spinal cord, skin, and bone regeneration. Unlike its disulfide cross-linked counterpart, kerateine, keratose does not possess a covalently cross-linked network structure and consequently displays substantially different characteristics. In order to understand its mode(s) of action and potential for clinical translatability, detailed characterization of the composition, physical properties, and biological responses of keratose biomaterials are needed. Keratose was obtained from end-cut human hair fibers by peracetic acid treatment, followed by base extraction, and subsequent dialysis. Analysis of lyophilized keratose powder determined that it contains 99% proteins by mass with amino acid content similar to human hair cortex. Metallic elements were also found in minute quantities. Protein oxidation led to disulfide bond cleavage and drastic reduction of free thiols due to conversion of sulfhydryl to sulfonic acid, chain fragmentation, and amino acid modifications. Mass spectrometry identified the major protein constituents as a heterogeneous mixture of 15 hair keratins (type I: K31-35 and K37-39, and type II: K81-86) with small amounts of epithelial keratins which exist in monomeric, dimeric, multimeric, and even degraded forms. Re-hydration with PBS enabled molecular assembly into an elastic solid-like hydrogel. Highly-porous scaffolds formed by lyophilization of the gel had the compression behavior of a cellular foam material and reverted back to gel upon wetting. Cytotoxicity assays showed that the EC50 for various cell lines were attained at 8-10 mg/mL keratose, indicating the non-toxic nature of the material. Implantation in mouse subcutaneous tissue pockets demonstrated that keratose resorption follows a rectangular hyperbolic regression with 92% degradation by an 8-week time point. Keratose was shown to integrate with the host tissue as evidenced by infiltration of leukocytes and fibroblasts, bulk material angiogenesis, and minimal fibrous encapsulation. Tissue response benchmarks were superior in keratose compared to the control PLGA 90:10 mesh. Finally, the degraded keratose was observed to remodel with the natural collagen extracellular matrix, verifying the benefit of using keratose as a temporary matrix for regenerative medicine applications.

Saturation labeling with cysteine-reactive cyanine fluorescent dyes provides increased sensitivity for protein expression profiling of laser-microdissected clinical specimens.

Laser capture microdissection (LCM) provides the capability to isolate and analyze small numbers of cells from a specific area of a histologic section. LCM has particular value for analysis of early stage tumors, which are often small and intermixed with non-tumor tissue. It has previously been shown that a new generation of cysteine-reactive cyanine dyes can, in principle, provide increased sensitivity for two-dimensional fluorescence difference gel electrophoresis (2-D DIGE) profiling when sample quantitities are limiting. However, the comparative advantage of the new dyes in a clinical setting has not been established. Here, we report that cysteine-reactive dyes allowed the identification of more features than established, lysine-reactive dyes with a given number of cells. This was true both with extracts prepared from human papillomavirus E6 and E7-transduced human keratinocytes, a model for early-stage cervical cancer, and with LCM samples. In an experiment comparing LCM clinical samples of gastric adenocarcinoma versus precancerous, spasmolytic polypeptide expressing metaplasia (SPEM) from the same patient, cysteine labeling allowed the identification of more than 1000 discrete protein spots in samples containing 5000 cells. This is a 5- to 50-fold smaller sample than used in previous studies. Both labeling methods had a comparable success rate for protein identification by mass spectrometry (MS). The proteins associated with more than 40 differentially abundant spots in the clinical samples were identified by MS. In this exploratory analysis, changes in expression levels of cytoskeletal proteins, molecular chaperones, and cell-signaling proteins were seen. The identification of a number of proteins that are potentially relevant to tumor progression suggests that the method holds promise for biomarker discovery.