A keratin-based microparticle for cell delivery

Abstract

Keratin-based biomaterials represent an attractive opportunity in the fields of wound healing and tissue regeneration, not only for their chemical and physical properties, but also for their ability to act as a delivery system for a variety of payloads. Importantly, keratins are the only natural biomaterial that is not targeted by specific tissue turnover-related enzymes, giving it potential stability advantages and greater control over degradation after implantation. However, in-situ polymerization chemistry in some keratin systems are not compatible with cells, and incorporation within constructs such as hydrogels may lead to hypoxia and cell death. To address these challenges, we envisioned a pre-formed keratin microparticle on which cells could be seeded, while other payloads (e.g. drugs, growth factors or other biologic compounds) could be contained within, although studies investigating the potential partitioning between phases during emulsion polymerization would need to be conducted. This study employs well-established water-in-oil emulsion procedures as well as a suspension culture method to load keratin-based microparticles with bone marrow-derived mesenchymal stem cells. Fabricated microparticles were characterized for size, porosity and surface structure and further analyzed to investigate their ability to form gels upon hydration. The suspension culture technique was validated based on the ability for loaded cells to maintain their viability and express actin and vinculin proteins, which are key indicators of cell attachment and growth. Maintenance of expression of markers associated with cell plasticity was also investigated. As a comparative model, a collagen-coated microparticle (Sigma) of similar size was used. Results showed that an oxidized form of keratin (“keratose” or “KOS”) formed unique microparticle structures of various size that appeared to contain a fibrous sub-structure. Cell adhesion and viability was greater on keratin microparticles compared to collagen-coated microparticles, while marker expression was retained on both.

Keywords

Keratin microparticle stem cell tissue engineering cell plasticity

Introduction

Current tissue engineering approaches often rely on the use of naturally and synthetically derived materials as platforms for regenerating tissue.¹ Incomplete or insufficient regrowth of the native tissue, a lack of autologous tissue for transplantation, or a lack of endogenous cells and factors to stimulate regeneration are all documented as impeding tissue regeneration in traumatic injury.² Naturally derived biomaterials have shown several advantages including inherent cell and tissue compatibility, complimentary mechanical properties and biological cues that support cellular adhesion and growth.³

Often, natural tissue regeneration relies on the presence of undifferentiated stem cells with a high proliferative potential and ability to replace damaged and terminally differentiated cells.⁴ Characterized by the ability to self-renew and differentiate into a variety of specialized cell types, many cellular strategies for tissue regeneration employ stem cells, such as mesenchymal stromal cells (MSC). MSC are culture-adherent, multipotent cells capable of differentiating into muscle, bone, cartilage, fat, tendon and nerve.⁵ Isolated from various sources including bone marrow, adipose tissue, muscle tissue, amniotic fluid, human placenta, periosteum, cord blood and even peripheral blood,⁶ the survival of delivered MSC and their contribution to tissue regeneration has been demonstrated, but can depend on the method of delivery.⁷

Microparticles, 1 micrometers to 1000 micrometers (µm) in diameter, have been touted as a multimodal delivery vehicle, capable of serving as a building block for 3-Dimensional (3D) constructs to fill tissue voids while simultaneously delivering soluble factors and cells.⁸ The application of microparticles provides a larger surface area for exterior cell loading while avoiding concerns about clearance, commonly viewed as a potential negative side effect.⁹ A benefit of a micro-scale protein construct may ultimately lie in its capacity for non-invasive injectable delivery, while at the same time agglomerating to form a nearly contiguous hydrogel.¹⁰

Prior research suggests keratin and keratin derivatives from human hair exhibit several mechanical and biological properties that support their use as scaffolds for cells that can incorporate various payloads including drugs and growth factors.^11–13 Keratin hydrogels mechanically perform as a flexible, viscoelastic material. However, most reports of keratin biomaterials in tissue engineering utilize cell-free constructs, despite keratin’s intrinsic cell and tissue compatibility and apparent support of cell health and function.¹⁴ This may be a consequence of the formation of keratin hydrogels being incompatible with cells, particularly in the kerateine (KTN; reduced form of keratin) system where gelation is best driven by protein self-assembly at basic pH.¹⁵ The keratose (KOS) hydrogel system is more amenable to the inclusion of cells as these gels can form at physiologic pH, but the resulting constructs can be fragile and degrade rapidly. Crosslinking can be used to mitigate this, but some crosslinking chemistries may compromise cell viability.¹⁶ To address this, a system in which the keratin microparticles are formed before cell incorporation may have utility, provided cells can be subsequently adhered effectively to the outer surface.

This study employs a KOS microparticle, synthesized via a water-in-oil emulsion synthesis procedure,¹⁷ as a potential delivery vehicle for bone marrow-derived MSC (BM-MSC) and compares them to a similar collagen-coated microparticle that is commercially available.¹⁸ A keratin microparticle technology has specific advantages over the more typically used keratin hydrogel system, which cannot always be formed in the presence of cells due to solubility limitations and the need for exogenous crosslinkers that may be cytotoxic. A microparticle approach allows for all processing steps to be completed before introducing cells. In the present study, we demonstrate the synthesis and characterization of one such keratin-based system, and demonstrate that a cell useful in tissue regeneration applications, the BM-MSC, can readily attach and remain viable.

Materials and methods

Human hair was obtained from a commercial vendor and used as received; peracetic acid (PAA), tris(hydroxymethyl)-aminomethane (Tris) base, and sodium hydroxide were obtained from Fisher Scientific and used as received. Mem-α cell culture media, phosphate buffered saline (PBS), and penicillin-streptomycin antibiotic were obtained from Gibco™. Butane-diol diglycidyl ether (BDDE, MW:202 Da), ethylene glycol diglycidyl ether (EGDE, MW: 174 Da), poly(propylene) glycol diglycidyl ether (PGDE, MW: 380 Da) as well as Triton-x 100 solution were purchased from Millipore Sigma. Light mineral oil and Hexanes were purchased from Fisher Scientific. Lyophilized elastase was obtained from Worthington Laboratories. Primary antibodies were obtained to target CD90 and CD34 (Santa Cruz Biotechnologies) and CD73 (BD Pharminogen).

Keratose extraction

Keratin proteins were extracted as previously described.¹¹ Briefly, human hair fibers were oxidized using a 2% peracetic acid (PAA) solution (pH 2) for 18 hrs using a 20:1 ratio of PAA to hair mass (20 mL PAA: 1 g hair). Oxidized fibers were rinsed with deionized (DI) water to remove residual oxidant and Tris base (100 mM) was used to solubilize keratin proteins from the fiber cortex. The keratose (KOS) protein extract was centrifuged at 25,000 rpm and a dimeric complex of type I and type II oxidized keratins were purified using a custom tangential flow filtration system employing proprietary buffers.¹⁹ The KOS solution was concentrated, frozen and lyophilized until further use.

Water-in-oil emulsion microparticle (MP) synthesis

KOS was weighed out and dissolved in phosphate buffered saline (PBS) to form a 60 mg/mL solution; crosslinker (three different crosslinkers were investigated, BDDE, EGDE and PGDE) was added to form a 48% w/w solution relative to the KOS mass; the sample was then vortexed and centrifuged at 1000 rpm for 5 min to remove air bubbles. The solution was allowed to stand overnight at room temperature (RT). KOS solutions were added to light mineral oil at a 1:20 volume ratio and stirred using a Caframo® tabletop mixer (Georgian Bluffs, Ontario) at 2000 rpm for 24 hours. After allowing the MP to settle, the light mineral oil was aspirated off, removing as few MP as possible. Samples were subsequently washed with chilled hexanes at approximately –25 °C using a lab bench shaker (Boekel Scientific) set to a medium speed setting for 5 min. Residual hexane was aspirated off and this step was repeated 2 more times to further remove residual oil. Finally, any residual hexane was aspirated and MP were lyophilized prior to further use.

Cell culture and loading procedure

Primary BM-MSCs (iXCells, 10RA-023) isolated from the femurs of male rats were cultured in T-75 culture flasks (Corning) with growth medium consisting of MEM-α media supplemented with 10% heat-inactivated Fetal bovine serum (Gibco) and 1% antibiotic Penicillin-streptomycin (1 IU/mL Penicillin: 1 µg/mL Streptomycin, Gibco). Cultures were stored in a sterile MCO-20AIC series Panasonic™ incubator at 37ᵒC with 5% CO₂.

To integrate cells with MP, a sterile P100 petri dish was pre-wetted with growth medium.

60 mg of MP, previously sterilized using a 100 Gy dose of x-ray irradiation in an RS 2000 x-ray irradiation system (RAD Source, USA), were added to the wetted petri dish to allow for media absorption and swelling. Five million (5x10⁶) cultured BM-MSC were added to the MP/media mixture. A magnetic stir bar was added and the suspension was gently stirred at 200 rpm for 3 hrs in the incubator. As a comparative control for cell-to-material surface interactions and cell survivability, BM-MSC were taken through the same loading procedure with commercially procured polystyrene MP that were dip-coated with Type-1 collagen (ca. 200 µm diameter; Sigma Aldrich).

Scanning electron microscopy

Scanning electron microscopy (SEM) was employed to observe MP size, shape and surface topography. MP were fixed on a metallic chip using adhesive carbon tape and sputter coated with a 10 nm thick layer of platinum/palladium using a Leica ACE600 sputter coater prior to imaging. Samples were imaged using a LEO 1550 (Zeiss) field-emission scanning electron microscope. The resulting SEM images were used largely for descriptive purposes but were also utilized for dimensional measurements.

Size and structure analysis

The average size of the synthesized MP was determined using two methods. In the first method, the previously mentioned SEM images were processed using the ImageJ²⁰ image processing software. Diameter measurements of several spherical MP (n = 20) were taken based on a length to pixel ratio determined via the SEM and averaged to determine the mean and standard deviation.

The second method involved analyzing samples using a Horiba LA-950 particle size analyzer (Kyoto, Japan). To determine the average diameter of MP after synthesis, MP samples were suspended in ethanol to prevent ambient water absorption, swelling and agglomeration of MP. Samples were agitated using a combination of stirring and sonication to prevent further MP agglomeration. MP were analyzed in triplicate via laser diffraction using refractive indices of 1.54 and 1.36 for the keratin MP and ethanol respectively.²¹ MP would later be analyzed under water hydrated conditions to help understand the sizes produced in physiological scenarios.

To analyze MP shape and internal structure we investigated the average porosity and relative pore connectivity using confocal microscopy. MP were incubated with a 4′,6-Diamidino-2-phenylindole dihydrochloride (DAPI) cellular stain (0.1 µg/mL) that was readily absorbed by the keratin protein. MP were incubated at room temperature in the dark for 5 min, washed with PBS three times and visualized using a Zeiss confocal microscope. Z-stacks measuring between 4 and 5 µm thickness were taken and compiled to create a 3-dimensional construct of an individual particle’s internal structure. ImageJ image processing software, specifically the BoneJ plugin,²² was used to quantify the pore fraction (defined as the volume of pore space divided by the total volume of a particle), average pore size and pore connectivity (defined as the fraction of interconnected pores per MP) in three different batches of MP (n = 50).

Hydrogel properties

The swelling behavior of KOS MP was tested to determine water absorption and estimate the anticipated size of a hydrogel particle in vivo. The diameter of dehydrated MP was measured using light microscopy (Supplemental Material). Individual MP were hydrated using one type of a gradient combination of water and ethanol (0%, 25%, 50%, 75%, 100% ethanol). MP were allowed to hydrate and swell for 5 min, after which the size of the particle was measured again. Individual diameters were measured and quantified by correlating the before and after light microscopy image to the imaged length of a standard hemocytometer as a reference, in the same imaging plane.

The capacity for water uptake and persistent water binding (thus forming the hydrogel)²³ was confirmed via a simplistic water uptake experiment. 20 mg of MP were saturated with DI water and were allowed to swell for 1, 12, 24 hours at room temperature or were immediately removed from DI water, as determined by preassigned time periods. After the respective time point residual water was removed and the particle mass was immediately measured; after each samples respective time point an additional measurement was taken after a 24-hour period of air drying at ambient temperature to measure bound water.

Surface characterization

The topography of MP was observed using a Bruker Multimode atomic force microscope (AFM). MP were fixed using adhesive carbon tape and sealed within an air tight chamber prior to testing. Inlet and outlet tubing allowed for the chamber to be filled with DI water via syringe through the inlet while the outlet tube was closed off via a plastic clasp. Particles were allowed to hydrate for 10 min prior to testing. After the unit was calibrated, 10 µm x 10 µm surfaces were analyzed under scanning mode with a scan speed of 3.9 µm/s using a Bruker non-conductive silicon nitride probe tip (k = 0.12 N/m) (n = 20).

Elastic behavior of microparticles

AFM was applied again using the same apparatus mentioned previously to measure the elastic properties of individual hydrated MP as a determinant of the mechanical stiffness experienced by loaded cells.²⁴ MP were fixed to an AFM chip using adhesive carbon tape and sealed within an airtight chamber. Inlet and outlet tubing allowed for a chamber to be filled with DI water via syringe; MP were hydrated for 10 min prior to testing. The Veeco Multimode AFM unit was calibrated and used a Bruker non-conductive silicon nitride probe tip (k = 0.6 N/m) in tapping mode to measure deflection upon surface contact with a forward velocity of 5.75 µm/s and a scan rate of 1 Hz.

Imaging cell-loaded microparticles by SEM

Prior to imaging, cells upon loaded MP were fixed with a 2% paraformaldehyde solution for transportation. MP were then suspended in a 1% osmium tetroxide (OSO₄) solution in 0.1 M sodium cacodylate buffer as a secondary fixative. MP were taken through a dehydration ethanol gradient (15%, 30%, 50%, 70%, 95%, 100%) to remove trapped water. Finally, MP were critical point dried, gold sputtered with a 10 nm coating thickness and imaged via SEM.

Quantifying loaded cell counts

Upon loading MP with BM-MSC, the number of cells was quantified using 4,6-Diamidino-2-phenylindole (DAPI) nucleic stain and confocal microscopy. MP were processed for imaging at 24, 48 and 72 h after culture. MP were DAPI stained for 5 min and washed with PBS three times, after which they were suspended in PBS and imaged within 30 minutes after washing. This process was performed for both keratin and collagen MP.

Validating cell adhesion

24 h after the initial suspension culture, loaded MP were immunofluorescently stained using the FAK100 Actin Cytoskeleton and Focal Adhesion Staining Kit (Millipore Sigma) to identify key cellular components in structural development, motility and adhesion. Specifically, cytoskeletal actin filaments and membrane-cytoskeletal vinculin proteins, which are documented as being critical in cellular development and survival.²⁵ Cells loaded onto MP were fixed for 20 minutes with a 4% paraformaldehyde solution and washed using buffer. A 0.1% Triton-X 100 solution was applied to permeabilize cell membranes, which was also removed with wash buffer. Cells were incubated initially with primary (anti-vinculin) antibody diluted in blocking buffer, followed by a secondary (Goat x mouse, FITC-conjugated) antibody, both at room temperature for 1 hour. Post washing, cells were kept in PBS and imaged within 30 minutes of the finished staining procedure. The same staining procedure was used on cells loaded onto collagen MP. Cells were imaged using a Zeiss confocal microscope.

Maintenance of cell plasticity

Prior to the integration of BM-MSC onto KOS MP, tests were performed to verify initial surface marker expression. Cells were immunofluorescently stained for markers C90, CD73, and CD34, as these have been previously shown to indicate cell plasticity.²⁶ Briefly, 100 µL of cell suspension (∼1x10⁶ cells) was used for each sample. ∼1 µg of fluorochrome-conjugated antibodies was added to their respective tubes. Tubes were vortexed and incubated on ice for 20 minutes. To wash off excess antibody following staining, 1 mL of 1X PBS was added to each tube, which was subsequently centrifuged for 5 minutes at 2000 rpm. The supernatant was aspirated and the pellets were resuspended in 500 µL of 2% paraformaldehyde prior to flow analysis, which occurred within 2 hrs after the process was completed. Controls for flow analyses included blank (unstained) cells and cells stained with individual fluorophores (i.e. no secondary stains).

Cells loaded onto MP were cultured for 72 hours under the same conditions mentioned previously. Post-integration, cells were detached by degrading the keratin protein in a 0.5 U/mL elastase solution for 5 minutes and gently vortexing the suspension to promote the break-up of the KOS and detachment of the cells. After pelleting by centrifugation, cells were immunofluorescently stained as previously described, with the full spectrum of stains or used as controls to compare to unstained but loaded cells or cells stained with individual fluorophores that underwent the same loading conditions.

Statistical analysis

Data is presented using descriptive statistics (mean and standard deviation), where appropriate. To compare experimental groups, analysis of variance (ANOVA) and/or linear regression analysis with a Tukey’s post hoc test was performed in all experiments. To determine significant variances, for all tests, a p-value <0.05 was considered statistically significant. Sample sizes were determined using a power analysis where a power (1-β) of 0.9 and an alpha (α) of 0.05 was employed. All statistical analyses were performed using JMP Pro 11.

Results

Water-in-oil emulsion microparticle synthesis

SEM images of water-in-oil emulsion MP and a cell body atop the keratin surface can be seen in Figure 1. MP appear three dimensional and spherical with a rough surface structure (Figure 1(a)). The keratin solution phase readily formed microdroplets during synthesis, but these were not self-sustaining. No surfactant or other emulsion-stabilizing agent was added, and the keratin itself did not act as a surfactant (despite its polyanionic nature), so the emulsion destabilized if stirring was discontinued. Based on the random distribution of particle size, there appeared to be no equilibrium micelle diameter under the conditions used, and therefore, no uniform sized MP resulted.

MP integrated with cells were hydrated for cell seeding and dehydrated as part of the processing for SEM imaging. As a result, rapid changes in particle size and shape likely made it difficult for cells to retain their adhesion. Consequently, SEM imaging of cells on MP was challenging without prior fixation, although some instances did occur (1B, indicated by yellow arrows). Also of note, MP imaged after hydration are visibly more porous than dehydrated MP. This is attributed to loosely bound keratin being removed upon particle hydration and agitation as keratin particulates were visible after the culture process. Imaged cells appear spread, which was later confirmed by confocal microscopy. Adhesions formed along the periphery of the cell appear to show attachment along protruding ridges of the fibrous structure (1C, indicated by red arrows), although these adhesions cannot be solely verified using SEM.

Size and structure analysis

The size analysis of individual MP using ImageJ was translated by scale bars provided in the SEM software and suggests average particle diameters of 236 ± 80.4 µm (data not shown). Based on laser diffraction analyses, mean particle diameters were determined to be 350 ±71.5 µm when placed in an ethanol suspension (Figure 2). As laser diffraction was capable of processing the sizes of entire batches (on the order of thousands) of MP, there is more reliance on these values. It should be noted that some samples initially produce MP diameters greatly outside of the normal range with diameters verging on the mm scale; this can be attributed to particle aggregation. Simple post-processing steps such as sifting through screens of specific pore size are believed to circumvent these issues.²⁷

The internal structure of the oil-in-water generated MP can be described as that of a coarsely fibrous particle with high porosity (Figure 2). Quantitative results indicate a protein to pore fraction of 0.59 ±0.04 suggesting that ∼40% of a given particle’s volume is unoccupied space. The pore connectivity of the structure was relatively high with a connectivity of 0.53 ±0.17 µm⁻³ (Figure 2).

Hydrogel properties

The applied gradient of ethanol and water produced a relatively linear trend in particle swelling as seen in Figure 3 (Top). Little to no swelling was seen in MP soaked in 100% ethanol, which was to be expected. MP hydrated with 100% DI water produced the greatest level of swelling. Overall, there appears to be a swelling coefficient of approximately 1.6 under pure water conditions, which would be expected to be similar in vivo.

Hydrogel formation was confirmed using the water uptake experiment seen in Figure 3 (Bottom). In all cases, MP swelled and reached masses over 1000% their original mass at their respective time points of measurement. Stable water-bound MP were achieved after 12 hours as indicated by high (greater than 95%) mass retention after air-drying for 24 hours. It can be inferred from the data that full hydration and subsequent hydrogel formation occurs over a shorter period as ∼75% of the mass is retained after just 1 hr of saturation.

Surface characterization

Based on 10 µm X 10 µm AFM scans (Figure 4), particle surfaces exhibit relatively rough topographies with larger ridges in varying regions. Quantitative results suggest an overall average surface roughness of 0.264 µm with no significant variations in the topographies between MP synthesized in different batches. Surfaces of this roughness magnitude, in combination with the aforementioned cell binding sites, suggest the feasibility of cellular attachment.²⁸

Elastic behavior of microparticles

Elastic moduli of individual hydrated MP, shown in Figure 5, vary between 5 and 15 kPa. Analysis of hydrated MP produced an interesting trend in terms of the relationship between the elastic properties of MP and their respective crosslinkers. PGDE crosslinked MP produce significantly greater elasticities than EGDE MP and BDDE crosslinked MP produce significantly greater elasticity than either PGDE or EGDE MP. Based on this data, it is thought that the crosslinker size or molecular weight may play a role and offer a window of tunability, albeit somewhat narrow. As BDDE crosslinked MP conferred the highest modulus of elasticity, all subsequent tests used this crosslinker solely as it is believed this created the most ideal substrate for cellular attachment and survival.²⁹

Quantifying loaded cell counts

Results show that BM-MSC successfully integrated with KOS MP and are attached at higher density compared to collagen-based MP loaded under the same conditions (Figure 6). Physical differences (size, weight, porosity etc.) prevent direct comparisons of loading efficiency between the two constructs however it should be noted that between the two vehicles no significant decreases in cell survivability were observed in cells under the same loading steps, suggesting that this keratin-based MP can be similarly utilized as an efficient delivery vehicle. High variability can be observed in the keratin-based samples and was determined to be a factor of sparse or incomplete loading observed in randomly selected MP.

Validating cell adhesion

BM-MSC suspension cultured with KOS MP shown in Figure 7 show the formation of distinct actin cytoskeletal filaments after 24 h. Similarly, vinculin membrane proteins can be identified along various cell bodies. Formation of the same protein complexes were observed in the collagen-based model, suggesting that a keratin-based MP can promote the same if not an improved cellular response compared to a commonly used collagen-based MP substrate.

Maintenance of cell plasticity

Side versus forward scatter data from the MSC population used for cell loading experiments showed a homogeneous population (data not shown). Prior to integration, BM-MSC exhibit markers consistent with cell plasticity with greater than 99.0% of analyzed cells testing CD90+/CD73+ and 78.0% of the analyzed population testing CD34- (Figure 8). Singular CD34 plots indicate marker presence with positivity to the right of the vertical line and negativity to the left (horizontal gating bar is y-axis are unused in these plots). Seventy two hours after culture in a suspension with MP, a shift in the population was observed with only 63.1% of the cell population remaining CD90+/CD73+ positive while 74.9% of the stem cell population remained CD34-, indicating potential phenotypic drift. It should be noted that the loaded cells needed to be detached from the MP with a potentially destructive elastase solution prior to flow cytometry experiments, which may have contributed to the apparent marker expression shifts. The effect of elastase on cells actually caused some fragmentation in the cell population, as a result gating was adjusted to isolate healthy cells which perpetuated or potentially led to the shifts observed.

Discussion

The procedure for water-in-oil emulsion was successful in generating KOS MP, although size and size distribution varied considerably. Morphology within the MP appeared fibrous, suggesting that the propensity for keratin to self-assemble was at least partially retained. This created a unique microarchitecture with loosely packed fibers that contribute to the highly porous nature of the construct. Mass transport of nutrients and waste throughout the KOS MP is expected to be high, based on these parameters, and may contribute to cell viability. Moreover, the fibrous structure contributed to a rough surface topography, which may facilitate cell adhesion through presentation of sites for attachment. Analysis of cell-loaded MP by SEM showed that particle surfaces appeared more porous than unloaded MP, suggesting that loosely bound keratin may have been removed during suspension/agitation. This was likely a consequence of the initial purity of the KOS used in these experiments, which has previously been shown to include a small fraction of keratin peptide that would not be expected to contribute to network formation and therefore would be removed during the cell integration process.³⁰

Observations by light microscopy during the development of the water-in-oil emulsion method showed round, smooth micelles (Supplemental Material), so the surface roughness observed by AFM and SEM were likely generated during the lyophilization process. This may be a beneficial consequence of using KOS as many other systems, using both synthetic and natural biomaterials, have been engineered for improved cell adhesion and survival using a variety of approaches, including surface roughness.³¹ In addition, immobilization of growth factors have been shown to be pro-survival for attached cells.³² Other biomaterials used as cell carriers, some showing good cell adhesion and survival, include silk fibroin, chitosan and PLGA.^33,34 Potential advantages of keratin MP over these other biomaterial systems include its cell and tissue compatibility, and resistance to enzymes that target tissue turnover (e.g. collagenase). Mammals do not produce keratinases, so there are no enzymes with specific activity toward keratins, which may offer significant advantages in human medical applications.

Cell loading was accomplished for keratin MP and loading procedures were successful for both keratin a collagen MP. Noted differences between the two MP, specifically size and surface area, exist; although it was thought that disparities in cell counts could be mitigated by the larger number of collagen MP. Substrate micro-features are a well-documented factor in cell loading, survival, and growth.³⁵ Other characteristics, like the curvature of surfaces, may also impact loading efficiency but require further investigation. Of benefit to the current keratin construct is a high porosity that, at minimum, provides a greater surface areas for potential attachment. It may similarly create more concave (inward facing) surfaces for cell attachment which have been shown to be more advantageous compared to convex (outward facing) material counterparts³⁶ but further investigation is needed to verify any true benefit. SEM images of loaded MP show that cells were are able to form attachments through focal adhesions, which allows for more normal physiologic activity and may contribute to cell survival, including during freeze-thaw cycling. However, flow cytometry analysis demonstrated a measurable shift in markers for cell plasticity, suggesting some changes may be elicited by contact with a keratin substrate, although this may also be attributed to the cell removal process (i.e. digestion of the MP with elastase). Spontaneous differentiation of MSC attached to MP may be of concern - depending on the clinical application - in systems where cells are being expanded for further use, or where a cell-seeded MP construct is the therapeutic product.

Numerous commercial “microcarriers” are available for MSC expansion, including those made from dextran, cellulose, polystyrene and gelatin, some of which include coatings of collagen, fibronectin and other proprietary materials.³⁷ Maintenance of the attached cells’ plasticity and differentiation potential of several commercial microcarriers have been evaluated.³⁸ Rafiq et al. conducted extensive flow cytometric characterization and assessments of plasticity both pre- and post-microcarrier attachment/removal. The investigators found that MSC generally retained their plasticity following harvest from the microcarrier-based system, but that there were significant differences in these outcomes for the different microcarriers. The authors recommended that once a system is designed and validated, the microcarrier system should not be changed, even though the cell source may vary. In another paper, Sart et al. reviewed the ability of microcarriers to influence stem cell fate.³⁹ The authors discussed the connection of microcarrier composition, charge, size and porosity to cell spreading, and noted that many studies focused on the osteogenic potential of the cells. However, the focus of this paper was on driving cell differentiation, rather than maintenance of plasticity.

There are currently no MSC-loaded MP therapeutic products approved by the US Food and Drug Administration (FDA). However, several are in preclinical development. These typically utilize a biodegradable MP as opposed to the non-biodegradable forms used in MSC expansion operations. As previously stated, such biodegradable systems have included silk fibroin, chitosan and PLGA.^33,34,40 Studies typically assess cell adhesion, and in some cases application-specific cell markers, but none investigated maintenance of cell plasticity. Morille et al. did investigate the ability of MSC attached to PLGA microcarriers (which also contained TGFβ-3) to stimulate cartilage repair in vitro and in vivo (mice), and found a protective effect (i.e. reduction of cartilage degradation) compared to microcarrier alone. Applications include brain (Parkinson’s treatment), bone, and smooth muscle.^41–43

Many keratin biomaterial-based constructs have been investigated over the past decade.⁴⁴ Most of these investigations have utilized hydrogels, sponge-like scaffolds, or films, although there is a published study on keratin MP made from feather.⁴⁵ Interestingly, these investigators demonstrated the formation of spherical feather keratin MP using an acid precipitation method rather than an emulsion, although size distribution data is not shown. However, it should be noted that the reduced form of keratin (i.e. “kerateine”) was used, not keratose. Another kerateine-based MP system is described that does make use of hair fibers as the source material, but the characteristics of the MP/microspheres are not described.⁴⁶ A third method, sulfitolysis, was used to generate keratin MP that were formed by spray drying, which resulted in “raisin-like” MP with a narrow size distribution (mean of approximately 6 µm).⁴⁷ Similar to other keratin biomaterial studies, the MP demonstrated a lack of cell toxicity.

Keratin, particularly keratose, represents an interesting alternative to many current MP systems. The data in the present study suggest good adhesion of MSC with only a few hours of suspension culture, which appears to mature to focal adhesion-mediated attachment. This is accomplished by virtue of the keratin itself (i.e. its primary amino acid structure) and topography of the MP surface, and without the need for a coating (e.g. fibronectin). Moreover, our data suggests that a significant amount of plasticity is retained, although further studies are necessary to probe differentiation potential and function of implanted cells in a tissue engineering application. The KOS MP system may also be amenable to the expansion of MSC for cell therapy, and perhaps as a substrate, more generally, for the growth of mammalian cells and subsequent production of biopharmaceuticals.

Conclusions

The present study demonstrated the successful synthesis of an oxidized keratin or keratose (KOS) MP system. MP were variable in size, but had an interesting surface topography and fibrous nature throughout the bulk of the MP. This feature, which may have been due to residual self-assembly characteristics, may provide unique advantages to a keratin-based MP system. Surface roughness likely contributed to rapid cell adhesion during co-suspension culture with MSC, but staining/microscopy demonstrated that seeded cells formed mature focal adhesions. This may have led to improved resistance to cell stress as viability after a single freeze-thaw cycle was improved for KOS MP-seeded cells compare to cells seeded on conventional collagen coated MP.

Supplemental Material

sj-pdf-1-jba-10.1177_0885328220951892 - Supplemental material for A keratin-based microparticle for cell delivery

Supplemental material, sj-pdf-1-jba-10.1177_0885328220951892 for A keratin-based microparticle for cell delivery by Marc Thompson, Aaron Giuffre, Claire McClenny and Mark Van Dyke in Journal of Biomaterials Applications

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Virginia Polytechnic Institute and State University Cunningham Fellowship Award.

ORCID iD

Mark Van Dyke

Supplementary material

Supplementary material is available for this article online.

References

Kanbe

Yang

Wei

, et al. Pericellular matrilins regulate activation of chondrocytes by cyclic load-induced matrix deformation. J Bone Miner Res 2007; 22: 318–328.

Naderi

Matin

Bahrami

AR.

Review paper: critical issues in tissue engineering: biomaterials, cell sources, angiogenesis, and drug delivery systems.

J Biomater Appl 2011; 26: 383–417.

Wang

Vorwald

Dreher

, et al. Evaluating 3D-printed biomaterials as scaffolds for vascularized bone tissue engineering. Adv Mater 2015; 27: 138–144.

Man

Wang

Guo

, et al. Angiogenic and osteogenic potential of platelet-rich plasma and adipose-derived stem cell laden alginate microspheres. Biomaterials 2012; 33: 8802–8811.

, et al. Regulating myogenic differentiation of mesenchymal stem cells using thermosensitive hydrogels. Acta Biomater 2015; 26: 23–33.

Prigozhina

Khitrin

Elkin

, et al. Mesenchymal stromal cells lose their immunosuppressive potential after allotransplantation. Exp Hematol 2008; 36: 1370–1376.

Oliveira

Mano

JF.

Polymer-based microparticles in tissue engineering and regenerative medicine.

Biotechnol Progr 2011; 27: 897–912.

Qazi

Mooney

Pumberger

, et al. Biomaterials based strategies for skeletal muscle tissue engineering: existing technologies and future trends. Biomaterials 2015; 53: 502–521.

Behrens

Sikorski

, et al. Blood-aggregating hydrogel particles for use as a hemostatic agent. Acta Biomater 2014; 10: 701–708.

10.

Woo

Park

Lee

KY.

Shear reversible cell/microsphere aggregate as an injectable for tissue regeneration.

Macromol Biosci 2014; 14: 740–748.

11.

Ledford

Simmons

Chen

, et al. Keratose hydrogels promote vascular smooth muscle differentiation from C-kit-positive human cardiac stem cells. Stem Cells Dev 2017; 26: 888–900.

12.

Saul

Ellenburg

de Guzman

, et al. Keratin hydrogels support the sustained release of bioactive ciprofloxacin. J Biomed Mater Res Part A 2011; 98: 544–553.

13.

Kowalczewski

Tombyln

Wasnick

, et al. Reduction of ectopic bone growth in critically-sized rat mandible defects by delivery of rhBMP-2 from kerateine biomaterials. Biomaterials 2014; 35: 3220–3228.

14.

Konop

Czuwara

Klodzinska

, et al. Development of a novel keratin dressing which accelerates full-thickness skin wound healing in diabetic mice: in vitro and in vivo studies. J Biomater Appl 2018; 33: 527–540.

15.

Hill

Brantley

Van Dyke

Some properties of keratin biomaterials: kerateines.

Biomaterials 2010; 31: 585–593.

16.

Tanabe

Okitsu

Yamauchi

Fabrication and characterization of chemically crosslinked keratin films. Mat Sci Eng C-Bio S 2004; 24: 441–446.

17.

Viswanathan

Thomas

Pandit

, et al. Preparation of non-porous microspheres with high entrapment efficiency of proteins by a (water-in-oil)-in-oil emulsion technique. J Control Release 1999; 58: 9–20.

18.

Batorsky

Liao

Lund

, et al. Encapsulation of adult human mesenchymal stem cells within collagen-agarose microenvironments. Biotechnol Bioeng 2005; 92: 492–500.

19.

Van Dyke

Rahmany

Keratin nanomaterials and methods of production. Google Patents, 2017. Patent Number: US20170051027A1.

20.

Schneider

Rasband

Eliceiri

KW.

NIH Image to ImageJ: 25 years of image analysis.

Nat Methods 2012; 9: 671–675.

21.

Leertouwer

Wilts

Stavenga

DG.

Refractive index and dispersion of butterfly chitin and bird keratin measured by polarizing interference microscopy.

Opt Exp 2011; 19: 24061–24066.

22.

Doube

Kłosowski

Arganda-Carreras

, et al. BoneJ: free and extensible bone image analysis in ImageJ. Bone 2010; 47: 1076–1079.

23.

Ahmed

EM.

Hydrogel: Preparation, characterization, and applications: a review.

J Adv Res 2015; 6: 105–121.

24.

Wiedemair

Serpe

Kim

, et al. In-situ AFM studies of the phase-transition behavior of single thermoresponsive hydrogel particles. Langmuir 2007; 23: 130–137.

25.

Wang

Taraballi

Tan

, et al. Human keratin hydrogels support fibroblast attachment and proliferation in vitro. Cell Tissue Res 2012; 347: 795–802.

26.

Pittenger

Mackay

Beck

, et al. Multilineage potential of adult human mesenchymal stem cells. Science 1999; 284: 143–147.

27.

Gauthier

Bouler

Weiss

, et al. Short-term effects of mineral particle sizes on cellular degradation activity after implantation of injectable calcium phosphate biomaterials and the consequences for bone substitution. Bone 1999; 25: 71s–74s.

28.

Goldmann

WH.

Role of vinculin in cellular mechanotransduction.

Cell Biol Int 2016; 40: 241–256.

29.

Ansari

Chen

Hasani-Sadrabadi

, et al. Hydrogel elasticity and microarchitecture regulate dental-derived mesenchymal stem cell-host immune system cross-talk. Acta Biomater 2017; 60: 181–189.

30.

Potter

Van Dyke

Effects of differing purification methods on properties of keratose biomaterials. ACS Biomater Sci Eng 2018; 4: 1316–1323.

31.

Neto

Oliveira

Mano

JF.

Microparticles in contact with cells: from carriers to multifunctional tissue modulators. Trends Biotechnol 2019; 37: 1011–1028.

32.

Aday

Zoldan

Besnier

, et al. Synthetic microparticles conjugated with VEGF 165 improve the survival of endothelial progenitor cells via microRNA-17 inhibition. Nature Commun 2017; 8: 747.

33.

Perucca Orfei

Talò

Viganò

, et al. Silk/fibroin microcarriers for mesenchymal stem cell delivery: optimization of cell seeding by the design of experiment. Pharmaceutics 2018; 10: 200.

34.

Malagón-Romero

Clavijo

Moreno

DJ.

Production of Chitosan microcarriers using electrospray equipment. Chem Eng 2018; 64: 361–366.

35.

Tamjid

Simchi

Dunlop

JWC

, et al. Tissue growth into three-dimensional composite scaffolds with controlled micro-features and nanotopographical surfaces. J Biomed Mater Res Part A 2013; 101: 2796–2807.

36.

Werner

Blanquer

Haimi

, et al. Surface curvature differentially regulates stem cell migration and differentiation via altered attachment morphology and nuclear deformation. Adv Sci 2017; 4: 1600347.

37.

Jossen

van den Bos

Eibl

, et al. Manufacturing human mesenchymal stem cells at clinical scale: process and regulatory challenges. Appl Microbiol Biotechnol 2018; 102: 3981–3994.

38.

Rafiq

Coopman

Nienow

, et al. Systematic microcarrier screening and agitated culture conditions improves human mesenchymal stem cell yield in bioreactors. Biotechnol J 2016; 11: 473–486.

39.

Sart

Agathos

YJBp.

Engineering stem cell fate with biochemical and biomechanical properties of microcarriers.

Biotechnol Progr 2013; 29: 1354–1366.

40.

Morille

Toupet

Montero-Menei

, et al. PLGA-based microcarriers induce mesenchymal stem cell chondrogenesis and stimulate cartilage repair in osteoarthritis. Biomaterials 2016; 88: 60–69.

41.

Moradian

Keshvari

Fasehee

, et al. Combining NT3-overexpressing MSCs and PLGA microcarriers for brain tissue engineering: a potential tool for treatment of Parkinson’s disease. Mater Sci Eng C 2017; 76: 934–943.

42.

Shekaran

Lam

Sim

, et al. Biodegradable ECM-coated PCL microcarriers support scalable human early MSC expansion and in vivo bone formation. Cytotherapy 2016; 18: 1332–1344.

43.

Parmar

Ahmadi

Day

RM.

A novel method for differentiation of human mesenchymal stem cells into smooth muscle-like cells on clinically deliverable thermally induced phase separation microspheres.

Tissue Eng Part C Methods 2014; 21: 404–412.

44.

Shavandi

Silva

Bekhit

, et al. Keratin: dissolution, extraction and biomedical application. Biomater Sci 2017; 5: 1699–1735.

45.

Sharma

Gupta

Chik

SMST

, et al. Characterization of keratin microparticles from feather biomass with potent antioxidant and anticancer activities. Int J Biol Macromol 2017; 104: 189–196.

46.

Barrows

TH.

Biomaterials made from human hair. Google Patents, 2015. Patent Number: US9072818B2.

47.

Cilurzo

Selmin

Aluigi

, et al. Regenerated keratin proteins as potential biomaterial for drug delivery. Polym Advan Technol 2013; 24: 1025–1028.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.12 MB