Collagen nanofibers have excellent potential for use as scaffolding material for bone tissue engineering. This study evaluated nanofiber sheets formed by electrospinning of type I collagen. Rat bone marrow stromal cells (BMSCs) were cultured, expanded and seeded onto the nanofiber sheets, tissue culture plastic or collagen coated surfaces. Unseeded nanofiber sheets served as controls. The BMSCs were cultured for up to five weeks and their attachment, viability and differentiation into osteogenic lineage were characterized using Fluorescent Microscopy (live/dead assay), histology (Von Kossa), and Scanning Electron Microscopy (SEM). The results demonstrated that the BMSCs attached and spread well on all collagen surfaces confirming biocompatibility. After five weeks of culture without differentiation medium, BMSCs differentiated into osteoblasts and produced calcium minerals on the collagen surfaces and within the nano-scaffolds but not on the tissue culture plastic. The mineralization was confirmed by Genesis X-Ray microanalysis (EDAX). The results of this work demonstrated that extracellular protein collagen type I can serve as a single stimulus for BMSCs differentiation. Importantly, this study showed that collagen nanofibers can be used as scaffolds for osteogenic differentiation of BMSCs in vitro without the addition of differentiation stimuli.
Keywords: Collagen; Nanofibers; Mineralization; Rat Bone Marrow; Stromal cells; Osteogenic Differentiation
Tissue engineered bone materials are attractive alternative to synthetic grafts since they are designed to degrade after the appropriate cells start making their own matrix. Collagen Type -1 is one of the most abundant proteins and a critical structural component of bones. Collagen meets all the criteria for an applicable biomaterial as it is extensively involved in the process of adhesion and proliferation of many cell types. As a bioactive material; the physicochemical properties of collagen can be modified by crosslinking with many reagents since it has amino, carboxyl and hydroxyl groups that can serve as crosslinking sites . Many studies have reported that the tensile properties of collagen fibers improve after crosslinking [2-4]. Rat bone marrow mesenchymal stem cells (BMMSCs) are multipotent cells and are characterized in vitro by their ability to differentiate into osteogenic, chondrogenic, and adipogenic phenotypes [5- 7]. After isolation the rat BMMSCs retains their self-renewal and differentiation capacity. The combination of collagen nanofibers and BMSCs make them suitable for the engineering of various tissues including bone. It was reported that bone marrow mesenchymal stem cells can be differentiated into osteoblast on several, polymeric materials [8-17]. The materials used in these studies was shown to be a good substrate for the osteogenic differentiation of bone marrow mesenchymal stem cells (BMMSCs) but the process involved the addition of differentiation factors to the culture medium, such as β- glycerolphosphate, dexamethasone, ascorbic acid and tetracycline.
The mechanisms for differentiation are not fully understood, but some work on human BMMSCs indicates that some extracellular
matrix protein may play an important role in their differentiation. The work done by Salasznyk et al showed that vitronectin and collagen I promote the osteogenic differentiation of human mesenchymal stem cells (hMSC), and that signaling through the collagen 1 receptor α1β1 integrin and the VN receptor αVβ3 integrin played the most significant role in promoting osteogenesis . The same group further showed that laminin-5, expressed by hMSC, stimulates osteogenic gene expression in these cells in the absence of differentiation medium .
Beside the composition of the extracellular matrix, surface topography plays an important role in promoting the differentiation of BMMSCs. George and Miyata proved that rat bone marrow MSCs differentiated into osteoblasts on a honeycomb structure of collagen without the addition of other differentiation factors . Yet, it is known that collagen fiber alignment significantly contributes to the mechanical strength of bone. It was found that if the predominant fiber direction is about the osteon’s long axis, there is increase in the ultimate tensile strength and strain of bone [21, 22]. Therefore, it would be expected that the use of aligned nanofibers would improve the mechanical properties of the whole construct.
In this work, we tested the BMSCs differentiation capability on collagen nanofibers without the use of any differentiation factors. We expect that the contact of BMSCs with the fiber structure of collagen is sufficient to induce differentiation in these cells. The results of this study are promising and the mineralized nanofiber sheets can be used for guided bone regeneration in the areas of low load, as well as these sheets can be combined to form stronger material for load bearing areas.
Materials and Methods
Dulbecco’s Modified Eagles Medium (DMEM), Fetal Bovine Serum (FBS), live/dead viability/cytotoxocity kit, and trypsin- EDTA solution were purchased from Invitrogen. 1-ethyl- 3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC) was purchased from Pierce, and phosphate buffer saline (PBS) from Gibco. 1, 1, 1, 3, 3, 3-Hexafluoro-2-propanol (HFP), Silver nitrate and sodium thiosulfate were purchased from Sigma- Aldrich.
Collagen nanofibers extrusion
Collagen from rat tail tendon was isolated by the method of Elsdale  and used to form collagen nanofibers. The lyophilized dry collagen at a concentration of 12 % w/v was dissolved in 2.45ml of HFP and 0.05ml of deionized water. The protocol for electrospinning was based on the work of Matthews and Bowlin , using a custom apparatus modified with two charged aluminum plates designed to introduce a secondary electrical field  This field, along with the rotational velocity of the grounded collector resulted in more aligned nanofibers. Scanning electron microscopy (SEM) was used to characterize the morphology of the nanofibers.
Collagen Coated Plates Preparation
Tissue culture plates with 35mm in diameter were used; one ml of collagen solution (1mg/ml) dissolved in 10mM hydrochloric acid was placed in each culture plate, to create a thin coating. The collagen was allowed to dry for 24 hours, and then the surfaces were sterilized in 80% ethanol for 24 hours followed by extensive washing with sterile PBS. Ready plates were kept in sterile PBS until the time of culture.
Crosslinking of collagen nanofibers
Nanofibers were crosslinked using 10mM EDC dissolved in 90% acetone. Fiber sheets were submerged in this solution for 24 hours then transferred to fresh 90% acetone for another 24 hours. Crosslinked fibers were consecutively washed in 50% acetone for four hours followed by two hours washing in 0.1M Na2HPO4. Subsequently, they were rinsed extensively with dH2O, and then dried at room temperature . Crosslinked nanofibers were sterilized in 70% ethanol for 24 hours, then washed with sterile distilled water four time; sterilized samples were dried at room temperature under sterile condition.
Bone marrow stromal cells isolation and culture
Stromal cells were isolated from bone marrow of adult female Sprague-Dawley rats immediately after sacrifice . All animal work was performed in compliance with institutional committee guidelines for animal welfare. The femurs were dissected out and the ends of the bone were cut open. The shafts of the bones were flushed with sterile phosphate buffered saline (PBS) using an 18-gauge needle, the cell suspension was passed through a 100-μm nylon cell strainer. The cells were centrifuged at 1 500 rpm for ten minutes and washed twice with sterile PBS. The cells were suspended in DMEM containing 10% FBS supplemented with penicillin 20 mU/ml and streptomycin 20 μg/ml, and the cell suspension plated in 25- cm2 tissue culture flasks. The first medium change was done after five days and then every other day. Adherent cells were grown to 70% confluency, then trypsinized and subcultured at a ratio of 1:3. The cells used in this experiment were from passage four.
Crosslinked fiber sheets were cut into squares with the dimensions of 2x2 cm with a thickness of 1mm. Each dry sheet was set on a coverslip and placed in 35mm Petri dish, and then each was directly seeded at a density of 1x 106 cells per cm 2 in 100μl of medium, unseeded sheets served as control. Other controls used in this experiment were collagen coated plates and tissue culture plastic; those were seeded at a density of 2x 103 cells per cm2 all conditions were incubated at 37°C and 5% CO2 for one hour to allow cell attachment. After the incubation period fresh medium without differentiation factors was added, and consequent medium change was done every two days.
Fluorescent Microscopy Analysis
Cultured cells were observed daily under a Phase contrast microscope. Cell viability was assessed using live dead assay at two and four weeks. Each sample was incubated for 15 min at 37 °C with 3 mL of DMEM medium that contains 4 μM calcein- AM and 2 μM ethidium homodimer. After the incubation period scaffolds were washed twice with warm PBS and then viewed using epifluorescence microscopy (Zeiss) coupled with a digital camera (AxioCam). Fluorescent images were taken using the software AxioVision.
Scanning Electron Microscopy and EDAX Analysis
Dry nanofibers were gold coated and assessed for alignment before seeding using SEM (JSM-6510 LV-LGs from JEOL). SEM analysis of seeded nanofibers performed at the end of the culture period, nanofibers were fixed in 10% buffered formalin for 24 hours then washed with distilled water for three times followed by air drying for 24 hours. Seeded fibers were mounted on carbon tape and evaluated for mineral content using Genesis X-Ray microanalysis (EDAX), after which the samples were sputter coated with gold and the morphological characteristics observed using SEM.
Von Kossa Staining
The cells in the plastic control and the collagen coated plates from weeks two and four were fixed in 3ml of 10% phosphate- buffered formalin for 20 minutes. After the fixation step the plates were washed twice with distilled water then 2% silver nitrate solution added and the plates were exposed UV light for two hours. Samples were rinsed several times with distilled water to remove the unreacted silver followed by the addition of 5% sodium thiosulfate for five minutes then washed again with distilled water several times. After the washing steps cells were counterstained with 2% neutral red for three minutes then rinsed with distilled water. The seeded and the control unseeded nanofiber sheets from weeks two and five were fixed in 10ml of 10% phosphate-buffered formalin for 24 hours, and then they were stained similarly. Staining visualized under bright light microscopy (Zeiss).
Results and Discussion
BMSCs Viability on Collagen Nanofibers
BMSCs from passage four were seeded on crosslinked collagen nanofibers. Cultured cells attached on the nanofibers and maintained their viability throughout the culture period. Although quantification of proliferation was not performed, live dead assay performed on days 14 and 28 showed very high viability as seen through fluorescence (Figure 1-A and B). Fiber alignment and cell attachment were further demonstrated by SEM. Figure 2-A is showing the nanofibrous structure before cell seeding and figure 2-B showing complete coverage of the material by well attached and spread cells after 5 weeks of culture (Figure 2-A and B). The desirable cellular response to the material indicates that collagen nanofibers are biocompatible material suitable for BMSCs culture.
Figure 1. Live/ dead assay performed on day 14 (A) and day 28 (B) of the cultured MSCs on the nanofibers. The fiber structure is completely covered with cells. Live cells stain green and dead cells stain red, images taken using 20x objective.
Figure 2. (A) SEM image showing the morphology of collagen nanofibers before seeding. (B) SEM image of seeded nanofibers after five weeks of culture showing complete covering of the material by differentiating cells.
Osteogenic Differentiation of BMSCs on Collagen Coated
Surfaces and on Nanofiber Sheets
The classic method for differentiation of MSCs to osteoblasts in vitro involves expanding the MSCs for one week then culturing them with osteogenic media for additional two to three weeks. Additional stimulating factors were also used for MSCs differentiation in three dimensional nanofibrous materials . In this study the ability of BMSCs to differentiate into osteoblasts without the use of osteogenic medium or other induction factors was examined on tissue culture plastic, collagen coated culture plates and on collagen nanofiber sheets. Salaznick et al in 2004 showed that hMSCs could differentiate into osteoblasts and produce minerals on purified bovine collagen I coated plates as early as 16 days of culture. In our experiment, at two weeks of culture BMSCs formed nodules on rat tail collagen I coated surfaces but mineral deposition was not detected by Von Kossa staining (Figure 3-A). This delay in mineralization observed on our culture at two weeks could be due to species variation. On the same surfaces at four weeks of culture the BMSCs differentiated, formed lager, denser nodules and secreted their own matrix that was stained positively by von Kossa (Figure 3-B), while cells on the tissue culture plastic showed negative staining (Figure 3-C).
Figure 3. Von kossa staining of BMSCs cultured on collagen coated plates for two weeks (A) and four weeks. (B). Similar staining at four weeks for the cells cultured on the plastic control (C).
Figure 4. Von Kossa staining of whole scaffolds shows light deposition of minerals at two weeks of culture (A) scale bar 20μm, dense deposition at five weeks of culture (B) and negative staining of the control not seeded scaffolds bar 50 μm(C).
Figure 5. Von Kossa staining of paraffin embedded section after 5 weeks of culture showing homogeneous distribution of the minerals within the scaffold ( black aggregates), and the surrounding cells (pink) as viewed at low (A) and high (B) magnification.
Additional stimulating factors were also used for osteogenic differentiation in three dimensional materials , patterned collagen surfaces  and nanofibers [9, 14]. With the advancement in understanding the role of extracellular matrix some investigators started to differentiate stem cells on three dimensional ECM scaffolds without the use of differentiation factors, George et al used porous honeycomb scaffolds prepared from bovine dermal atelocollagen to differentiate rat bone marrow MSCs into osteoblasts . They were able to show an increase in alkaline phosphatase activity as early as on day 14 of their culture. They related their positive results to the specific morphology of the honeycomb porous structure. Similarly nanofibers provide large surface area for cell–ECM interaction which may promote earlier differentiation.
After two weeks of incubation whole nanofibers sheets were fixed in buffered formalin, and stained with Von Kossa, the scaffolds turned dark gray and when viewed under the light microscopy, we observed homogenously distributed small mineral deposits within the whole material indicating that the BMSCs are already differentiated and they are producing their own matrix (Figure 4-A). The culture was run for five weeks and the same staining was performed at that time point, which showed an increase of the mineral salts within the material (Figure 4-B). Control nano sheets cultured without cells negatively stained with Von Kossa (Figure 4-C). In order to show the distribution of the mineralization within the material, paraffin embedded sections were stained using the same method. These stained sections showed homogenous distribution of the calcium salts within the material (Figure 5-A, and B). The mineral deposition by the differentiated cells was confirmed on the surface of the scaffolds by SEM (Figure 6-A and B) and the integrity of the porous nanofibers was maintained during the culture period as shown in Figure 6-C; EDAX analysis indicated that these minerals are composed of calcium phosphate (Figure 7).
Figure 6. SEM images of nano-scaffolds after five weeks of culture with BMSCs showing different morphology of the mineral deposition on the surface of the material, (A and B) and a cross section view showing the porous structure of the material (C).
Figure 7. X-ray microanalysis (EDAX) of the deposited material identifies this material as calcium phosphate.
In this work we demonstrated that collagen type I in the form of two dimensional (coated plates) or three dimensional microstructure (nanofiber sheets), is an appropriate ECM component for differentiation of rat bone marrow stromal cells into osteogenic lineage. Furthermore this work indicates that BMSCs can be used as a cell source for bone tissue engineering as they were able to differentiate into osteogenic lineage on the collagen nanofiber scaffolds. This work provides insight into the possibility of using this biomaterial/cell combination for clinical applications such as for guided bone regeneration or as bone replacement material for bone filling defects. Future studies will focus on mechanical characterization of the crosslinked collagen nanofiber scaffolds before and after mineralization by the differentiated BMSCs, and in-vivo testing of the mineralized scaffolds in animal model.
This project was partially funded by DOD Award Number OR090654.