Evaluation of castor oil-based polyurethane membranes in rat bone-marrow cell culture

PuRPOsE: To evaluate three methods to isolate rats MSCs and to analyze the potential of a castor oil polyurethane base membrane as a scaffold for MSCs. METHOds: Four male Wistar rats, aged 20-30 days were used. Bone marrow aspirates from femur and tibia were harvested using DMEM high glucose and heparin. The cell culture was performed in three different ways: direct culture and two types of density gradients. After 15 days, was made the 1st passage and analyzed cell viability with markers Hoerscht 33342 and propidium iodide. The MSCs were characterized by surface markers with the aid of flow cytometry. After this, three types of castor oil polyurethane membranes associated with the MSCs were kept on the 6-well plate for 5 days and were analyzed by optical microscopy to confirm cell aggregation and growth. REsuLTs: Separation procedures 1 and 2 allowed adequate isolation of MSCs and favored cell growth with the passage being carried out at 70% confluence after 15 days in culture. The cells could not be isolated using procedure 3. When the 3 castor oil polyurethane membrane types were compared it was possible to observe that the growth of MSCs was around 80% in membrane type 3, 20% in type 2, and 10% in type 1. CONCLusION: Both Ficoll-Hypaque densities allow isolation of rat MSCs, and especially castor oil-based membrane type 3 may be used as a scaffold for MSCs.


Introduction
Bone marrow contains a type of stem cell denominated mesenchymal stem cell (MSC) that possesses a high degree of self-renewal and is capable of differentiation into several lineages of connective tissue 1,2 .MSCs are present in the bone marrow of newborns as well as that of adults, but there is evidence that these cells present an age-related decline in frequency 1 .Under defined culture conditions the MSCs may originate a spectrum of specialized mesenchymal tissues including bone, cartilage, muscle, marrow stroma, tendon, ligament, fat and a variety of other connective tissues [2][3][4] .
In this context, tissue engineering has used a combination of scaffolds, of natural or synthetic origin, and cells or a combination of both to produce a three-dimensional tissue that should be structurally and mechanically similar to or better than the tissue that it was designated to replace 3,5 .
Among the several different materials already tested may be cited polyurethane derived from the castor oil plant 6 ; which is a polymer formed by the urethane reaction of two components, namely polyol and pre-polymer 7 .The product allows versatility of formulations 8 , with the commercial presentations being especially moldable as biomass and pre-molded blocks 7 .This polymer has been studied to evaluate toxicity, biocompatibility, osteointegration, capacity to induce bone neoformation, among other properties 6,[9][10][11] .Some properties are still controversial such as, the ability of polyurethane to perform as an osteoconductive agent or just as a space filler 9,11 .
A prior study 6 evaluated polyurethane derived from the castor oil plant in cell culture with three different chemical compositions: pure, calcium carbonate addition or calcium phosphate addition.Bone marrow cells were cultivated under conditions that allowed osteoblastic differentiation.The results showed that addition of calcium carbonate and, even better, the addition of calcium phosphate facilitated the events that promoted matrix mineralization.Moreover, another study 12 cultivating MSCs from bone marrow with granules of castor oil polyurethane for seven days found that, besides initiating matrix deposition, the biomaterial allowed growth of MSCs that adhered and multiplied.
Since the chemical composition, consistency and format of the polymer may influence cell growth and adhesion to biomaterial, the aim of this study was to evaluate three methods to cultivate MSCs of rats and to analyze the potential of castor oil polyurethane in membrane format as a scaffold for MSCs.

Isolation of rat MSCs
This study followed the Guide for Care and Use of Laboratory Animals and was approved by the Ethics Committee of our Veterinary School (protocol nº 32/2009-CEEA).Four male Wistar rats, aged 20-30 days were used.Animals were euthanized by using halothane.Bone marrow aspirates from femur and tibia were harvested using Dulbecco's Modified Eagle Medium (DMEM) high glucose (DMEM + GlutaMAX -Gibco 10569-010) and heparin at a final concentration of 5U/mL.
The cell culture was performed in three different ways.
In the first one, the bone marrow cell suspension was diluted with 4mL of DMEM high glucose, and transferred slowly onto Ficoll-Paque aqueous medium 1.077g/mL density, according to Boyum 13 , and the material was centrifuged at 1500rpm for 40 minutes.After this, the mononuclear cell ring was collected and washed with 2mL of DMEM high glucose once at 1500rpm for 10 minutes.The cells were then resuspended in 5mL DMEM high glucose with 20% fetal calf serum (FCS), 100UI/ml of penicillin, 100µ/ml of streptomycin and 3µ/ml of amphotericin B and plated in 25 cm 2 culture bottles.
The second method was the same as the first, except that the density gradient utilized Ficoll-Paque aqueous medium at the density of 1.084g/ml.In the third method, bone marrow cell suspension was directly transferred to 25 cm 2 culture bottles in 5mL complete DMEM high glucose medium containing 20% FCS, 100UI/ml of penicillin, 100µ/ml of streptomycin, and 3µ/ml of amphotericin B.
The medium was changed every 5 days.Once the cells achieved 70% sub-confluence, they were resuspended, and the cell viability analysis was performed.Hoerscht 33342 (Sigma Chemical Co) staining was applied for 10-15 minutes; then the culture was washed with Phosphate buffered saline (PBS), and propidium iodide (Sigma Chemical Co) was used for 5 minutes.

Characterization of MSCs
To confirm the mesenchymal stem cell lineage, CD34 and CD44-specific surface antibodies (AbD Serotec, Oxford, UK) were used to mark mesenchymal cells.After the first passage, the isolated cell populations were prepared according to the antibody manufacturer protocols.The CD34 antibody FITC (Fluorescein Isothiocyanate) -conjugated was negative at direct immunofluorescence staining for flow cytometry (FACS Calibur-BD).The CD44 antibody associated with RPE secondary antibody (R. Phycoerythrin) was positive at indirect immunofluorescence staining for flow cytometry.

Evaluation of membrane surface
For the evaluation of the biomaterial's surface, a 1cm 2 sample of each type of membrane was submitted to scanning electron microscopy in JEOL JSM5310, with a tungsten filament at 10kV, using secondary electron detectors.For this, the membrane samples were mounted onto a metal stub with double-sided carbon tape, and liquid silver coating was used to increase the conductivity of the material.The samples were then coated with gold using a Denton Vacuum Desk II sputter.

Biomaterials in cell culture
Each type of membrane was dampened with FCS and each well was filled with 5ml of a medium containing DMEM low glucose, 20% FCS, 100UI/ml of penicillin, 100µ/ml of streptomycin, and 3µ/ml of amphotericin B before MSCs were added.The membranes associated with the MSCs were kept on the 6-well plate for 5 days and were analyzed by optical microscopy to confirm cell aggregation and growth.

Results
Scanning electron microscopy of the membrane surfaces (from x200 to x7500) showed morphological and topographical differences depending on the percentage of castor oil (Figure 1).Defects, pores of non-uniformity diameter, castor oil fibers, bubbles, sulcus and reentrances varied in size and amount according to the percentage of castor oil in the membranes.
Membrane type 3 had a rough irregular surface and presence of more defects, pores, bubbles, sulcus, reentrances, and large amount of castor oil fibers.Membrane type 2 had an aspect similar to membrane type 3, but the bubbles, pores and castor oil fibers were fewer than in membrane type 3. Membrane type 1 showed microscopic geometric irregularities on the surface characterizing a smoother pattern comparing to the others.In addition, the membrane type 1 had fewer bubbles, pores and castor oil fibers than the other types.Separation methods 1 and 2 allowed adequate isolation of MSCs and favored cell growth with the passage being carried out at 70% confluence after 15 days in culture.Cell viability was the same in both methods with an average of 93% of cells intact and 7% non-viable (Figure 2).The cells could not be isolated using method 3.
Flow cytometry showed 83% CD 44-positive/34negative (mesenchymal) and 17% CD 34-positive/44-negative (hematopoietic stem) markers.By optical microscopy, it was observed that the cell growth around the membranes was similar in the three types of membranes.However, the growth of MSCs on the membrane tops was approximately 80% in membrane type 3, 20% in type 2 and 10% above type 1 (Figures 3, 4 and 5).

discussion
MSCs are a minor fraction of the total nucleated cell population in bone marrow and thus they require isolation and expansion in culture to obtain sufficient numbers for clinical or experimental use 2,3 .In general, the sample of whole bone marrow is submitted to fractionation in solution based on density gradient, after which the cells are plated at densities from 1 x 10 4 cells/ cm 2 to 0.4 x 10 6 cells/cm 2 2 .The present study revealed that both 1.084 g/mL and 1.077 g/mL density gradients were good to obtain cell isolation, which was evidenced by the high percentage of CD 44-positive cells.The separation of erythrocytes and granulocytes present in the sample was facilitated by the fact that after dilution the bone marrow aspirate had a lower density than the reagents 14 .
On the other hand, the direct culture was demonstrated to be inadequate.As cited by other authors, purification of MSCs only by their adherence to plastic may not be sufficient 2 .
Scaffolds for MSCs should allow cell adhesion, be porous, permit bioactive molecules to have contact with cells, integrate in the neotissue, supply cellular cueing and be mechanically adequate for the site, among other properties 3 .Scaffolds composed of various classes of biomaterials have been tested, each one with their advantages and disadvantages.However, they should provide mechanical and structural properties similar to the tissue to be replaced 3,15 .
The castor oil polyurethane in membrane format could be used, for example, in promoting guided bone regeneration.A barrier membrane serves the function of excluding unnecessary cell lines and allowing the growth of required tissues 16,17 .For this purpose, factors such as composition, morphology, surface texture, presence of pores, and reabsorption capacity are determinate for membrane function 17,18 .In the present study the differences in castor oil concentration influenced the presence of pores, bubbles, sulcus, reentrances, and castor oil fibers among the membranes.
Concerning the MSC behavior on polyurethane membranes, it was possible to prove the growth and adhesion of cells by optical microscopy, a similar finding was observed in two other studies 6,12 .In the first study 6 the castor oil polyurethane was co-cultured with rat bone marrow cells, but under conditions that allowed osteoblastic differentiation, while the second one 12 used dog's MSCs.Both studies used compact polyurethanes discs 6 or granules 12 that may have contributed to cell adhesion.Additionally, the granules had calcium carbonate in their composition, what, according to the author, contributed to the initiation of matrix deposition by MSCs, as observed by electron microscopy 12 .In contrast, cell attachment and ALP activity were not affected by the chemical composition of the discs, but the events that promoted matrix mineralization were favored by the addition of calcium carbonate or calcium phosphate, both of which are considered more osteoconductive than pure polyurethane 6 .
In the present study the membranes with a higher percentage of castor oil showed the greatest growth of MSCs.
Theses membranes (types 2 and 3) had more surface pores than membrane type 1 as observed by scanning electron microscopy.These pores are formed by incorporation of air bubbles during the mixture of the basic polyurethane components 7,19 and were associated with higher castor-oil concentrations.Pore size and surface characteristics of a biomaterial are important for cell attachment and growth 20 .Therefore, in addition to porousness, the presence of the bubbles, sulcus, reentrances and castor oil fibers in membranes types 2 and 3 may have contributed to increasing the contact surface for the cells, allowing adhesion, growth and multiplication of the MSCs.

Conclusion
Both Ficoll-Hypaque densities allow isolation of rat MSCs; and especially castor oil-based membrane type 3 may be used as a scaffold for MSCs.
Finally the culture was washed with PBS to maintain cell integrity.Intact cells were stained blue by Hoerscht staining, and non-viable cells were stained red by propidium iodide.Fluorescence microscopy was used to analyze 5 different fields at 200X magnification.

FIGuRE 3 -
FIGuRE 3 -Optical microscope showing the cells growth (blue arrow) and the apoptosis cells (yellow arrow) on the membrane 1 (x200).

FIGuRE 4 -
FIGuRE 4 -Optical microscope showing the cells growth (blue arrow) and the apoptosis cells (yellow arrow) on the membrane 2 (x200).

FIGuRE 5 -
FIGuRE 5 -Optical microscope showing the cells growth (blue arrow) and the apoptosis cells (yellow arrow) on the membrane 3 (x200).