Collagen Based 3D Model for In Vitro Study of Osteogenesis.
Musculoskeletal diseases and trauma are responsible for mortality (1 million annually in the US  and considerable morbidity (30 million / severe disabilities annually, WHO) and can seriously impact the quality of life. The cost of health care in this arena is over $700 billion annually in the US, and is a significant national economic burden. Most of the nontraumatic, chronic musculoskeletal diseases are age related  and the increasing population over 65 years of age is projected to reach over 2.8 billion by 2100 (UN statistics). All these factors contributed to the designation of 2001-2011 as the "Bone and Joint Decade" by the WHO, highlighting the importance and the need for musculoskeletal research.
The human skeleton is a multifunctional system that provides mechanical support and protects soft internal tissues, stores physiologically essential calcium and phosphate ions and is the hematopoietic and immune system stem cell niche . Bone is a dynamic tissue in which three cell types, osteoclasts, osteoblasts and osteocytes, regulate its homeostasis by constant resorption and deposition of the mineral components and structural realignment and remodeling of its matrix in response to biochemical and mechanical signals. The primary component of bone matrix is collagen type I, which is converted to a hard porous scaffold by the process of ossification--deposition of hydroxyapatite (HA, a complex calcium phosphate salt) [4,5]. A 3-D morphology is required for the initiation and completion of ossification [6,7]. When cultured as a monolayer, osteoblasts initiate mineralization only after formation of nodules, a 2-D approximation of a 3-D scaffold. The expression of proteins characteristic of osteogenic differentiation is much more pronounced in 3-D cultures than in cells cultured as monolayers [8,9]. Thus in vitro, when the cells populate a 3-D matrix, the mineralization process begins earlier and proceeds faster. In vivo, osteoblasts repair a fracture by invading and mineralizing the collagenous matrix in a random manner , and as the mineral content increases, osteoblasts terminally differentiate into osteocytes, the most abundant cell type in bone. Thus the course of ossification in 2-D cultures differs markedly from that which takes place in 3-D, and supports superiority of 3-D models as in vivo like tools for bone biology research.
An important aspect of the 3-D structure is that it allows the bone to respond to mechanical forces by undergoing deformation to optimize its load bearing capacity . This mechanosensing capability and the ensuing remodeling process is an integral component of the bone function in vivo. Thus understanding of bone tissue biology would be incomplete without the 3-D environment that facilitates studies involving mechanotransduction and remodeling. Majority of the current 3-D models used as research tools lack mechanical integrity and stability because they are soft hydrogels . Application of compression or tension/torsion forces to these models, causes compaction of the matrix followed by expulsion of water, and invariably results in cell death. Studies that examine the responses to applied forces use cells cultured as monolayers on elastic membranes that can be stretched continuously or cyclically. This approach may be appropriate for studying the effects of tensile forces (vascular, tendon and skeletal muscle cells) but is inappropriate for studying responses to forces acting on bone (compressive and torsional forces). Thus there is a great need for a stable 3-D bone model that is populated with the appropriate osteogenic cells and is responsive to factors that regulate bone functions in vivo.
In the last decade 291 studies related to fracture healing were carried out in animal models . These studies included the use of small animals (mice, rats and rabbits) and large animals (dogs, sheep, pigs and goats). An appropriate 3-D in vitro model would allow execution of a variety of preliminary (proof of concept) studies at a greatly reduced cost and minimize ethical considerations that regulate the use of experimental animals.
A composite that best serves as a model for hard tissue biology research in 3-D relevant mechanoresponsive fashion, would also be of great benefit in tissue engineering of bone grafts. It will assist performance of fundamental studies in an in vivo relevant fashion and bridge the gap between the cell monolayer and animal models. Such a 3-D model should be easy to assemble, versatile, convenient to handle, and simple to evaluate using traditional and novel analytical approaches. We here present initial studies that are directed towards the design and assembly of such a 3-D model.
Materials and Methods
Porous collagen scaffold--Avitene Ultrafoam[R] was a kind gift from Davol Inc. (Warwick, RI, USA). Ultrafoam[R] is a native collagen type I sponge (US Patent 6454787, Davol Inc), which is completely bioabsorbed within six months of implantation without any residue (information from Davol). In all experiments the Ultrafoam[R] scaffold was divided into cubes (approximately 5 x 5 x 5 mm).
Collagen Type I Gel
Cold (4[degrees]C) porcine collagen type I-A solution (Cellmatrix[R], Nitta Gelatin, Wako Chemicals, Richmond, VA, USA, 3 mg/mL) (8 parts v/v) was mixed with cold (4[degrees]C) solution of MEM-a (serum free, 10X, non-buffered) (1 part, v/v) and then with cold (4[degrees]C) neutralization buffer (1 part v/v, sodium hydroxide 0.5 N, sodium bicarbonate 22 g / L HEPES free acid 47.7 g / L, to pH 7.4)) with thorough and careful mixing after each addition to obtain a homogeneous air bubble free viscous solution (CS). When incubated at 37[degrees]C under 5% C[O.sub.2] the CS gelled within 30 min. and this impregnation method was used in all experiments.
Characterization of Collagen Type I Scaffold
Cubical pieces (5 x 5 x 5 mm) of dry Ultrafoam[R] were weighed and then immersed in water or CS (from above) for 24 hours in a humidified incubator (37[degrees]C, 5% C[O.sub.2]). Once the uptake of fluids was complete the Ultrafoam[R] pieces were weighed and the ratio of (wet weight - dry weight) / dry weight was determined. The time taken for the collagen scaffold to become saturated after immersion in water was noted, and used to estimate the amount of cross-linking in Ultrafoam[R] .
Cy5[TM] labeling of Collagen
CS (10 mL) was reacted with bis-functional dye Cy5[TM] (0.2 mg, Amersham Biosciences, UK) dissolved in DMSO (200 [micro]L), stirring continuously for 24 hours at 4[degrees]C) and the labeled collagen solution was used as described.
GFP-tFbs: Human dermal fibroblasts immortalized by ectopic expression of hTERT were a kind gift from Dr. J. W. Shay (Department of Cell Biology, UT Southwestern Medical School, Dallas Texas). These cells were kindly transduced in the laboratories of Dr. Victor Gonzales (Howard Hughes Medical Center, Dallas Texas) using enhanced GFP and isolation, by cell sorting of the stable transfectants. 
GFP expressing hTERT dermal fibroblasts (GFP-tFbs) were cultured in DMEM (GIBCO, Invitrogen, Carlsbad, CA) containing fetal bovine serum (FBS, 10%, Atlanta Biologicals, Lawrenceville, GA) in a humidified incubator (37[degrees]C, 5% C[O.sub.2]) with medium changes every second day.
hMSC: Bone marrow derived hMSC (Lonza, Switzerland) were cultured in MEMa (GIBCO, Invitrogen, Carlsbad, CA, USA) containing FBS (10%, Atlanta Biologicals, Lawrenceville, GA), 1% Penicillin-streptomycin (GIBCO, Invitrogen, Carlsbad, CA, USA) and FGF (10 nM, R&D Systems, Minneapolis, MN, USA), with medium changes every second day. hMSC were passaged at 80-90% confluence using trypsin (0.05%) with 0.53 mM EDTA in HBSS (GIBCO, Invitrogen, Carlsbad, CA, USA) and plating at 3000 cells/[cm.sup.2]. Cells were counted using heamocytometer and Trypan blue (Sigma-Aldrich, St. Loius, MO, USA) dye exclusion viability stain.
hOST: Primary human osteoblasts (hOST, Lonza, Switzerland) were cultured in osteoblast growth medium (OGM, Lonza, Switzerland) which maintains the proliferative phenotype. hOST were expanded as described above for hMSC. Osteogenic differentiation of hMSC was induced using MEMa (GIBCO, Invitrogen, Carlsbad, CA, USA) containing FBS (10%, Atlanta Biologicals, Lawrenceville, GA, USA), ascorbate-2-phosphate (200 mM, Sigma-Aldrich, St. Loius, MO), dexamethasone (10 nM, Sigma-Aldrich, St. Loius, MO, USA), and [beta]-glycerol phosphate (10 mM, Sigma-Aldrich, St. Loius, MO, USA). This medium is referred to as osteogenic differentiation medium (ODM) and was changed every other day.
Impregnation of the Collagen Scaffold
After preparation, dispensing accuracy of CS, was determined by pipetting 50 mL into plastic weigh-boat weighing (OHAUS GA110 analytical balance, Pine Brook, NJ, USA). The collagen type I scaffold - Ultrafoam[R] cubes were immersed in medium, the excess medium blotted on filter paper and the cell suspension in CS (105 cells/mL, 250 mL) seeded on to the scaffold. The impregnated Ultrafoam[R] cubes were incubated at 37[degrees]C to allow collagen solution to gel and will be thereafter referred to as collagen gel impregnated collagen scaffold (CGCS).
Retrieval of Cells from CGCS
To release the cells, collagen type I foam scaffold (Ultrafoam[R]) or CGCS were treated with collagenase P (1 mL / 5 x 5 x 5 mm cube, Roche, Germany) for one hour at 37[degrees]C. Collagenase P was reconstituted as directed by Roche in Hanks Buffered Salt Solution to a concentration of 2 mg/ml.
The distribution of the cells in the CGCS was studied using scanning laser confocal microscopy (Zeiss, LSM-410) by analyzing the cellular content of the model at various optical planes throughout the specimens. Images were obtained using a 10 x obj ective, while keeping the pinhole and other parameters constant. GFP expression in GFP-tFbs was observed using 488 nm and Cytotracker [TM] Orange (Invitrogen, Carlsbad, CA) labeled hOST cells were observed using 568 nm excitation laser.
Neutral Red Staining
Neutral Red (3-Amino-7-dimethylamino-2-methylphenazine hydrochloride, Sigma-Aldrich, St. Louis, MO, USA) was dissolved in distilled water (1 gm in 100 mL) CGCS with GFPtFbs was maintained in culture for 30 days and then placed in serum free DMEM with 10% neutral red and incubated for 2 hours at 37[degrees]C. At the end of the incubation period CGF was washed with phosphate buffered saline (0.256 g/L Na[H.sub.2]P[O.sub.4] [H.sub.2]O, 1.19 g/L [Na.sub.2]HP[O.sub.4], 8.76 g/L NaCl, pH 7.4, in distilled water) (PBS) and examined by confocal microscopy.
The CGCS 3-D constructs, placed in wells of a 24-well plate, were washed with PBS (1 mL/well), and incubated with 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, (MTT, 0.2 mg / mL of PBS - 500 mL /well, Sigma-Aldrich, St. Louis, MO, USA) for two hours at 37[degrees]C. The formazan from metabolized MTT was dissolved in solubilization solution (250 mL, MTT assay kit) by rocking for 90 minutes, and 100 mL aliquots were transferred to wells of a 96-well plate (duplicate wells/specimen). The 96-well plate was examined using the plate reader (Molecular Devices Spectramax 340 PC, MDS Analyticals, Toronto, Canada) and the absorbance at 570 nm recorded. Using a monolayer of hMSC (seeding density 7.5 x [10.sup.3] - 2.5 x [10.sup.5] cells per well in a 24 well plate) the standard curve for the conversion of MTT into formazan (a straight line with Spearmans correlation statistic of 1, and p < 0.05) was used to calculate the cell number from the formazan absorbance. The standard curve was used to determine the cells lost through "escaped/overflow" during CGCS impregnation and subsequently the true number of cells residing in CGCS.
Alkaline Phosphatase Assay (ALP)
The CGCS 3-D model containing hMSC (5 x 5 x 5 mm cubes) were cultured in 24-well plate with medium (ODM) changed every other day. Samples were removed from culture weekly during the period of four weeks. The specimens washed with PBS (1 mL/well) in the wells in which they were cultured and were pulverized (small pestle) in the tissue lysis buffer (250 [micro]L, 1x component of ALP assay kit, Anaspec Enzolyte[TM], San Jose, CA, USA). Aliquots of tissue lysate (50 [micro]L) were transferred to a 96-well plate and each well was treated with p-nitrophenyl phosphate solution supplied in ALP assay kit (50 [micro]L / well, pNPP). After incubation (30 min at 37[degrees]C) the 96-well plate was examined at 405 nm (Molecular Devices Spectramax 340 PC, MDS Analyticals, Toronto, Canada). DNA content of tissue lysates was determined using an Eppendorf Biophotometer (Eppendorf, Westbury, NY, USA) with the built in settings for measuring double stranded DNA in solution.
Samples of CGCS were fixed in buffered formalin (10% formalin/ formaldehyde 10 mL, PBS, Fisher Scientific) at 4[degrees]C for 24 h, dehydrated through a series of ethanols and xylenes and embedded in paraffin. Embedded specimens were sectioned (~30 p thickness), and sections were deparaffinized by incubations in xylenes and ethanols. After rehydration for 30 min in PBS and distilled water washes (3 x 10 min), sections were used for histochemical / immunohistochemical analysis. Mounted specimens were examined on Olympus AX70 fluorescent microscope (Olympus, Center Valley, PA, USA) using SPOT[R] TWAIN software (Microsoft, Issaquah, WA, USA).
Rehydrated tissue sections were washed in Amplifying buffer (1x, 3 mL/slide, IHC Kit, Prohisto, Columbia, SC, USA), 5 minutes rocking at room temperature (RT, amplifying chamber, IHC, Prohisto). Slides were treated with epitope unmasking solution (1x, IHC Kit, Prohisto), covering all slides in the slide rack, and the slide rack heated in boiling water for 20 minutes. After cooling to RT, the slides were transferred to amplifying chamber (IHC Prohisto) in which all the subsequent steps were performed. Slides were washed with rocking (5 min at RT) with amplifying buffer (1x, 3 mL / slide, IHC Kit, Prohisto), blocked with 1% BSA (3 mL per slide, EMD Chemicals, Gibbstown, NJ, USA) in PBS, rocking for 30 minutes at RT. The slides were incubated with rocking for 24 hours at 4[degrees]C with mouse anti-osterix antibody (Novus Biologicals, Littleton, CO, USA) diluted to 1:500 or rabbit anti-osteopontin antibody (Abcam Inc., Cambridge, MA, USA) diluted to 1:800 dilutions in amplifying antibody dilution buffer (1x, 3 mL / slide IHC Kit, Prohisto). The slides were then washed 3 times in amplifying wash buffer (1x, 3 mL / slide, IHC Kit, Prohisto), rocking for 5 minutes at RT and then treated with 3 mL / slide antibody solution for 24 hours at 4[degrees]C on a rocker with Alexa Fluor 594 labeled goat anti-mouse and Alexa Fluor 488 labeled goat antirabbit (both from, Invitrogen, Eugene, OR, USA); secondary antibodies were diluted in PBS, 1:4000. After washing 3 times in amplifying wash buffer (1x, 3 mL per slide, IHC Kit, Prohisto), rocking for 5 minutes, at RT, and 3 distilled water washes, 3 mL/ slide, and rocking for 5 minutes at RT; the slides were covered with glass coverslips and mounted using Prolong Gold Antifade Reagent containing DAPI (Invitrogen, Eugene, OR, USA). After drying overnight the slides were examined by fluorescent microscopy as described above.
Alcian Blue Staining
Rehydrated tissue sections were immersed in Alcian blue stain solution (Alcian blue 8GX, 1 gm / 100 mL 3% acetic acid pH 2.5, Sigma-Aldrich, St. Louis, MO, USA) for 15 minutes, and then rinsed with distilled water and counterstained with neutral red solution (1 g in 100 mL water containing 100 [micro]L glacial acetic acid, Sigma-Aldrich, St. Louis, MO, USA) for one minute. After washing in distilled water the sections were covered with glass coverslips, sealed at the edges using nail polish and immediately examined by microscopy as previously described.
Von Kossa Staining
The staining solution was freshly prepared by mixing: naphthol AS MX-P[O.sub.4] (0.005 g, Sigma-Aldrich, St. Louis, MO, USA), N,N-dimethylformamide, (200 [micro]L DMF, Fisher, Fair Lawn, NJ, USA), Tris-HCl (25 [micro]L, 0.2 M, pH 8.3), distilled water (25 mL), red violet LB salt (0.03 g, Sigma-Aldrich, St. Louis, MO, USA). Filtered staining solution (Whatman's No. 1 filter paper) was applied to cover the slides and the slides incubated at RT for 45 min, rinsed in distilled water 3-4 times and left to stand in distilled water for 1 hour. Finally the slides were counterstained with silver nitrate (2.5% aqueous, Sigma-Aldrich, St. Louis, MO, USA) for 30 min. The excess silver nitrate was aspirated, the slides rinsed with distilled water 3 times and dried overnight, covered with a glass coverslips, sealed at the edges (nail polish) and immediately examined microscopically as previously described.
Graphical presentations of the results were generated using GraphPad Prism 4 (GraphPad Software, LaJolla, CA, USA), and the same software used for statistical analysis. The area of calcification was calculated using Matlab (The Mathworks, Natick, MA, USA). In all the experiments the "n" indicates biological replicates.
Characterization of Collagen Scaffold
The collagen type I foam scaffold (Avitene Ultrafoam[R]) had variable pore size (Fig 1a). Its high porosity was demonstrated by saturation with four times its dry weight of water (Fig. 1b) after 24 hour of immersion (fractional change in weight, mean = 4.0, SD= 0.26, n=3). Significantly, saturation of dry Ultrafoam[R] with 5 times its dry weight of neutralized collagen type I (CS), which was allowed to gel for 24 hours at 37[degrees]C was essential for assembly of CGCS (fractional change in weight, mean = 5.3, SD = 0.56, n=3). The absence of cross-linking of the Ultrafoam[R] was demonstrated by the "wetting time" that exceeded 20 seconds . No residue remained after the treatment of Ultrafoam[R] with collagenase P for one hour. These results support the fact that the Ultrafoam[R] is native collagen type I and has not been extensively cross-linked during the manufacturing process.
The Presence of Collagen Gel in CGCS Increases Cellular Seeding Efficiency and Loading
Pippeting 50 [micro]L of CS, dispenses a mean of 49.2 mg (~49.2 ml) with a standard deviation of 1.4 mg (~1.4 mL) (n=6). This ensures that collagen impregnation of the pores in Ultrafoam[R] scaffold to generate CGCS, can be precisely controlled and that the number of cells delivered to any porous scaffold can be accurately determined. The efficiency of delivering 50K cells suspended in the medium (500 ml) during impregnation of the Ultrafoam[R] scaffold pieces was 80-85% (7,500 to 10,000 cells escaped delivery) (n=2). The efficiency of cell delivery using 50K cells suspended in CS was above 92% (less than 3750 cells escaped delivery) (n=2) however, the penetration of cells into the scaffold was much higher but was difficult to quantify.
Cells suspended in CS saturated the pores throughout the Ultrafoam[R] scaffold, as shown by formazan color formation (deep purple) after MTT staining of the CGCS (Fig 2a). The presence of hMSC suspended in gelled collagen solution was demonstrated by hematoxylene / eosin (H and E) staining of the scaffold sections of the scaffold cut orthogonally to the seeding surface (Fig 2b).
Using confocal microscopy and the models cells (GFP expressing, telomerised human dermal fibroblast, GFP-tFbs), suspended in Cy-5 labeled collagen prepared in our laboratories, it was shown that the cells were distributed throughout the CGCS immediately after seeding (Fig 3).
Collagen supports proliferation and long-term survival
hMSC survive and proliferate for up to two months when cultured in CGCS (Fig. 4a). GFP-tFbs, cultured in the CGCS, showed normal morphology and viability at one month (neutral red staining - Fig. 4b). When GFP-tFbs were maintained in this 3-D model they were viable for a period of four months (data not shown).
Cell Proliferation in the CGCS
The presence of collagen gel in CGCS contributed significantly to cell proliferation and viability during one month of culture when compared with hMSC cultured in Ultrafoam[R] alone (Fig 5a, 2 way ANOVA). When the primary human osteoblasts (hOST) were seeded and cultured directly on the Ultrafoam[R], confocal microscopy showed CytotrackerTM Orange labeled cells to be predominantly on or close to the surface of the scaffold (Fig 5b panels 1i, 1ii). In contrast, when these cells were seeded as a component of CGCS, they were "delivered" to the interior of the CGCS (e.g. 100 m above the non-seeded surface (Fig. 5b. panels 2i, 2ii). After 14 days in culture hOST seeded on Ultrafoam[R], did not remain in the deeper regions of the scaffold (Fig 5b. panels 1iii, 1iv) but hOST seeded as a component of CGCS were observed in the interior of the 3-D model, although more cells were present at the surface (Fig. 5b panels 2iii, 2iv).
Migration of the Cells that Populate CGCS
The cell migration model was set up (as shown in Fig 6a) by placing CGCS without cells close to the edge of a well in 12-well tissue culture plate and the well was filled with neutralized collagen type I solution (3 mL) and allowed to gel. A small cavity (5 mm diameter) was created in the center of the well containing acellular, gelled collagen type I. A suspension of cells in neutralized collagen type I solution was added to the central cavity and the model was cultured and cell migration was observed over 14 days.
GFP-tFbs migrated from the central cavity and colonized / invaded the CGCS (located at the edge of the well) (Fig 6b, panel 1). When collagen type I impregnated CGCS containing GFP-tFbs was placed into the central cavity in acellular collagen gel, GFP-tFbs exited the CGCS and invaded the collagen gel (Fig 6b, panel 2). Repeating these experiments with Cytotracker [TM] Orange labeled hOST, showed similar migration of hOST into and out of the CGCS (Fig 6c).
CGCS is Mechanoresponsive
GFP-tFbs seeded on CGCS were randomly oriented after seeding (Fig 7a, panel 1). When the CGSS was stretched by hand for ten seconds and then studied after overnight incubation in the medium, the cells had realigned with their long axes parallel to the direction of the stretch (Fig 7a, panel 2). This change was observed throughout the scaffold as shown on gallery of images obtained by confocal microscopy (Fig 7b).
Osteogenic Differentiation of hMSC in CGCS
The expression of alkaline phosphatase (ALP) normalized to DNA increased to day 14 and then declined at day 28 (Fig 8a). Alcian blue staining demonstrated proteoglycan deposition around hMSC in CGCS after 28 days in culture under osteogenic differentiation conditions (Fig 8b).
Immunohistochemistry was performed to track expression of markers of osteogenic differentiation (Fig 9a and 9b). hMSC did not express osterix (transcription factor) or osteopontin (extracellular matrix protein) at the start of the experiment. However, hMSC expressed both markers by the 28th day of osteogenic differentiation in CGCS and when cultured in Ultrafoam[R].
Collagen type I Dependence of Mineralization in CGCS
Calcium deposits were detected using a modified von-Kossa staining protocol. No foci of calcification were observed in case of hMSC seeded on CGCS and maintained under nondifferentiation conditions (no dexamethasone, no ascorbate) (Fig 10a, panels 1i and 1 ii). Very few foci were observed when hMSC were seeded directly on Ultrafoam[R] and maintained under differentiation conditions (Fig 10a, panels 2i and 2ii) for 28 days, but multiple foci of calcification were observed after 28 days of differentiation, when hMSC were the cellular component of CGCS (Fig 10A, panels 3i and 3ii and Fig 10B).
Analysis of von Kossa stain positive areas showed that hMSC seeded on CGCS, and cultured under differentiation conditions for 14 days, 0.28% (mean, SD= 0.03%, n=3) of the image area was covered by calcium deposits. This area increased more than 10 fold (3.53%, mean, SD= 0.63%, n=3) when culture period was extended to 28 days (unpaired t-test p <0.001). The area of the image occupied by Ultrafoam[R] was 17.43% (mean, SD= 2.01%, n=3) and by gelled collagen was 60.34% (mean, SD= 3.67%, n=3). The remaining blank area of the image resulted from fixation, dehydration and paraffin embedding.
The ECM of bone contains 70 to 90% of collagen type I, the most abundant structural protein in higher living organisms [4,5]. To produce the characteristic hard bone tissue and yet retain some measure of flexibility, microcrystaline hydroxyapatite (complex calcium phosphate) is deposited on the network of collagen fibrils that are further organized into bundles. Fulfilling its multifunctional physiological role the bone provides a compartment for the stem cells that maintain blood and immune system cell populations as well those that comprise the skeletal system . The cellular constituents of bone include osteoblasts, osteoclasts and osteocytes. Osteoclasts and osteoblasts are responsible for bone homeostasis (cycles of deposition and resorption of mineralized ECM) , bone repair (regeneration), and structural remodeling in response to hormonal and mechanical signals. The terminally differentiated osteocytes are the "mechanosensors" and the major cell population in bone.
The bone formation process during development is well documented but bone repair and regeneration is not as well understood . While 3-D morphology is essential for mineralization of collagen type I bone matrix [6,7], differentiation of MSC to osteoblasts takes place in 2-D [17,18] and 3-D cultures [6,19]. Following an injury (fracture) activated osteoblasts invade and mineralize and become embedded in the collagenous microenvironment and differentiate into osteocytes. In contrast 2-D osteoblasts cultures generate multilayered cell sheets and form nodules of calcification [6,7] without terminal differentiation into osteocytes. The expression levels of osteogenic biomarkers are higher in 3-D cultures than in monolayer cultures [8,9] and mineralization begins earlier and proceeds faster in a 3-D environment [8,9,19]. Thus in a 3-D environment the process of differentiation and mineralization represents a more in vivo relevant ossification process . Although the course of osteoblast differentiation in 2-D is relatively uniform, the differentiation in 3-D is dependent on the model used because not all the models are truly three-dimensional. Cells seeded on various porous scaffolds form a monolayer on the pore surfaces, and although these surfaces are tortuous and convoluted, the osteogenesis forms cell multilayers that synthesize collagen and form nodules of mineralization, a progression that is characteristic of 2-D cultures.
Our efforts in engineering of human tissue have lead to the development of three-dimensional (3-D) models (also known as tissue equivalents or organotypic cultures) of the skin, cornea and conjunctiva . In these 3-D models of living human tissue, appropriate connective tissue cells (usually fibroblasts) populate the acid solubilized non-denatured collagen type I gels (the extracellular matrix, ECM). In contrasts to similar tissue constructs that contract spontaneously , our patented connective tissue models are dimensionally stable. A benefit of the non-contractile matrix is the translucency of our 3-D models that allows morphological observation of the cellular behavior by non-intrusive microscopic methods. Furthermore, our constructs allow free exchange of nutrients and waste products, resulting in long-term (months) stability and cell viability, and support cell migration . Expanding our collagen-based tissue engineering strategy, we here report assembly of an in vitro model of bone tissue that allows studies of poorly understood aspects of human osteogenesis in an in vivo relevant manner. We have examined the suitability of porous non-denatured commercially available collagen type I foam (Ultrafoam[R]), a scaffold stiffer than the collagen gel alone, yet cytocompatible, easy to handle and amenable to physical deformation. This approach is justified by the increasing interest in hydrogels as critical components that provide significant biological advantage to composite constructs and grafts .
Most of the current 3-D models are hydrogels that lack adequate mechanical strength . Human mesenchymal stem cells (hMSC, ) and osteoblasts (hOST, ) rapidly contract collagen hydrogels that they populate
to one third of the initial size, which decreases cell viability and survival. In order to prevent compaction and allow long-term cell viability we selected collagen foam (Ultrafoam[R]) as a stiff scaffold that will reinforce our patented collagen hydrogel . Furthermore, stiffer scaffolds have also been reported to favor osteogenic differentiation [10, 24]. Ultrafoam[R] is highly porous and accommodates five times its own weight of collagen type I gel (Figs. 1A and 1B). It is made from native (non-denatured) collagen type I that is not cross-linked as demonstrated by the long wetting time and complete, residue free, degradation after collagenase treatment. This is important because aldehyde based collagen cross-linking is one of the factors responsible for pathological calcification (foreign body reaction) that leads to structural deterioration and is a major reasons for failure of modified collagen implants . Ultrafoam[R] also provides a surface favorable for cell adhesion, is mechanically stronger, stable over long-term culture periods and easy to manipulate.
In order to generate consistent and uniform osseous structure and facilitate proper function a 3-D model requires homogeneous cell distribution. Penetration of cells into the interior of a scaffold at depths greater than 500mm has been a challenge , and has been approached by utilizing very high seeding densities . Thus introducing cells, suspended in a solution of neutralized collagen type I, into Ultrafoam[R], and allowing the collagen to gel at 37[degrees]C, formed collagen hydrogel containing collagen scaffold (CGCS). This approach ensured that even at low seeding density, cells could penetrate into and be distributed throughout the scaffold. Thus MTT staining, that generates formazan dark blue color  (Fig 2A), hematoxylin-eosin staining of paraffin embedded sections of CGCS (Fig 2B), and confocal microscopy (Fig 3) demonstrated distribution of cells throughout the CGCS, although the cell density was higher at the surface than in the interior of the model. This picture was further clarified by using fluorescently labeled "model cells" (a human dermal fibroblast cell line characterized by ectopic expression of telomerase and GFP expression, GFP-tFbs) suspended in Cy5[TM] labeled collagen into CGCS as shown in Fig 3. After 12 hrs. in culture, the seeding efficiency in CGCS was increased from 85%, when cells were seeded on Ultrafoam[R] without collagen gel, to more than 92% in the presence of collagen gel. We therefore increased by a factor of more than two, the reported seeding efficiency (40%) reported with collagen diluted with culture medium as the cell delivery vehicle .
It has been shown that lower seeding density does not affect tissue formation , and that very high density inhibits proliferation in 3-D scaffolds . We used cell-seeding density of 100,000 cells/ mL with 50,000 cells per scaffold for all our experiments. Lack of proliferation on collagen scaffolds, after one week in culture, was attributed to differentiation,  but since even early osteoblasts are proliferative, the arrested proliferation we observed is more likely due to compromised nutrition supply and overcrowding. In our CGCS model hMSC proliferated for a period of 8 weeks under differentiation conditions (Fig 4A), a much longer survival and viability than has so far been reported. This demonstrates the suitability of our model for long-term studies (Fig 4B), and four-month viability was also observed but not recorded.
The presence of collagen did not suppress hMSC proliferation even after 4 weeks of culture under differentiation conditions [see growth curves for hMSC cultured in Ultrafoam[R] and in CGCS (Fig 5A)]. Osteoblasts cultured in Ultrafoam[R] for 4 weeks, migrated to the surface and very few cells were present in the Ultrafoam[R] interior (Fig 5B), probably due to a more favorable nutrient supply at the scaffold surface. In contrast, a greater number of osteoblasts were present in CGCS interior (Fig 5B) most likely because presence of collagen gel stabilizes the growth factor access and availability, and maintains equally favorable nutritional conditions in the interior of the scaffold as at its surface. Regardless of the cell type (fibroblast, hMSC, osteoblast) the use of collagen gel impregnation facilitates total penetration into the scaffold, a more uniform distribution of cells, higher seeding efficiency and improved cell survival and proliferation.
Osteoblasts can up-regulate matrix metalloproteinases (MMP) secretion to degrade and remodel the surrounding matrix . MMP activity is important for differentiation of osteoblasts into osteocytes, cell survival by creating cell migration pathways, production of matrix fragments that serve as signaling molecules, and the release of matrix bound latent growth factors [10,24]. Consequently the presence of non-degradable (stable) crosslinks interferes with progression of ossification . Since the Ultrafoam[R] and the collagen hydrogel are native collagen type I (Fig 3), both are substrates for collagenase which was shown to completely dissolves CGCS. Thus cells populating the CGCS model are able to manipulate the collagen content as needed, demonstrating the potential for degradation / remodeling in vitro and if used as a graft in vivo. Cells recovered after collagenase digestion of CGCS attach and proliferate when placed in culture.
To produce a seamless interface when serving as a graft, our 3-D model (CGCS) has to support free movement of cells so that the graft-host interactions are facilitated. The schematic in fig 6A shows the design of the preliminary migration experiment, which we used to demonstrate qualitatively cell motility. The cells (fibroblasts and osteoblasts) were shown to migrate out of CGCS into surrounding acellular collagen gel (Fig 6B and C, panel 2) indicating the potential use of CGCS to deliver cells to the site of injury. Successful penetration by cells from collagen hydrogel into CGCS (Fig 6B and C, panel 1) indicates its ability to accommodate invasion by resident host cells during integration of graft with host tissue.
During the remodeling bone responds to mechanical forces by changing its structure and shape to optimize its load bearing capacity . The effects of force at the cellular level are currently studied using cell monolayers cultured on elastic membranes and subjecting the membranes to a variety of stretching forces. When a tensile force is applied to fibroblasts populating our 3D model--CGCS, they aligned parallel to the direction of application of force within 24 hours (Fig 7A). The alignment occurs throughout the specimen (Fig 7B), validating the use of CGCS for in vitro studies of mechanotransduction in 3-D. When the experiment was repeated with osteoblasts the alignment was not observed. This is likely to be due to the fact that the mechanosensing cells in bone are osteocytes, which instruct osteoclasts and osteoblast to remodel a specific area of need.
The exposure of hMSC populating CGCS to osteogenic differentiation conditions showed that alkaline phosphatase (ALP) activity peaked at 14 days and decreased by day 28 (Fig 8A). This ALP activity profile is similar to that reported to take place during differentiation of hMSC in a monolayer culture and confirms that the ALP activity is an early differentiation event  in our 3-D model also. Synthesis and pericellular deposition of glycosaminoglycans that occurs during chondrogenesis, also takes place during endochondral ossification and was observed by day 28 in Alcian Blue stained paraffin sections of CGCS (Fig 8B). Although demonstration of osteogenic differentiation monitors the changes in mRNA levels of several biomarkers, we favor demonstration of translation and presence of the corresponding proteins, a more challenging process in 3-D models. While in histochemical analysis uniform distribution of cells translates into fewer numbers of cells within a field of view (particularly at higher magnification), on Western blot analysis, the presence of collagen overwhelms the low concentrations of cellular proteins. Nevertheless, transcription factor osterix (Fig. 9A) and the extracellular matrix protein osteopontin (Fig 9B), were observed after 28 days of differentiation in Ultrafoam[R] without collagen gel as well as in CGCS. These results support our conclusion that osteogenesis is taking place in our in vitro model.
hMSC cultured in collagen gel led to formation of bone like tissue, expressed differentiation markers at levels similar to those in bone and showed enhanced mineralization . Similar behavior was not exhibited by hMSC cultured in fibrin, alginate, or pluronic F127 hydrogels. Differentiation of hMSC on thin collagen films  and culture of osteoblasts in collagen gels  deposited calcium phosphate on collagen fibrils. Osteoblasts cultured in a compressed collagen hydrogel, survived predominantly on the surface of the thin dense matrix and rapidly deposited calcium hydroxyapatite on the collagen fibrils . These studies support our hypothesis that the presence of collagen type I is essential for in vitro ossification. Previous in vitro studies with cross-linked collagen scaffolds did not differentiate between physiological (cell-mediated) calcification, and the crystallization due to spontaneous degradation of b-glycerol phosphate  (media component) in the monolayer culture controls. In our study one of the controls was the culture of hMSC in Ultrafoam[R] under differentiation conditions for 28 days (Fig 10A, panels 1i and 1ii). The second control was culture of hMSC in CGCS under non-differentiation conditions (no dexamethasone, no ascorbate) for 28 days (Fig 10A, panels 2i and 2ii). No calcification was observed in either of these controls supporting the fact that the calcification we observed in cultures of CGCS populated with hMSC under differentiation conditions for 28 days ((Fig 10A, panels 3i and 3ii), was the physiological mineralization. Higher magnification micrograph (Fig 10B) clearly shows mineral deposition in association with collagen hydrogel in the pores of CGCS. Quantification of areas rich in calcium deposits shows ten fold increase from day 14 to day 28 (Fig 10C) in corresponding cultures.
Data presented in this study demonstrated several findings critical to ossification. Our model is "cytocompatible" and supports the transition of our strategy from successful engineering of 3-D models of living "soft" human tissue to modeling bone morphogenesis. CGCS is not cross-linked, is biodegradable, and has sufficient mechanical integrity to facilitate easy handling and remains intact during prolonged culture periods. It retains its original shape and size without contraction, and supports invasion, proliferation, metabolic homeostasis, mechanotransduction and osteogenic differentiation of hMSC including calcification in a physiological manner. We further demonstrated that the presence of collagen type I is an essential cohort for effective cell delivery and long-term cell viability in the impregnated scaffolds and is particularly important for the osteogenic mineralization. In our 3-D models we have used several cells types and differentiation along a single lineage, but the data supports the proposal that CGCS can serve as an excellent in vitro 3-D model to study variety of cells in an in vivo relevant manner. CGCS is particularly appropriate as an in vitro 3-D model for studies of osteogenesis.
The first author received a graduate research stipend from the University of North Texas Health Science Center for the duration of the work.
[1.] Lidgren L. The bone and joint decade 2000-2010. Bull World Health Organ.; 81(9):629. 2003
[2.] Brooks PM. The burden of musculoskeletal disease--a global perspective. Clin Rheumatol.; 25(6):778-81; Nov 2006
[3.] Hing KA. Bone repair in the twenty-first century: Biology, chemistry or engineering? Philos Transact A Math Phys Eng Sci.; 362(1825):2821-50; Dec 2004.
[4.] ROBINSON RA, WATSON ML. Collagen-crystal relationships in bone as seen in the electron microscope. Anat Rec.; 114(3):383-409; Nov 1952.
[5.] Kielty CM, Hopkinson I, Grant ME. Collagen- the family: structure, assembly and organization in the extracellular matrix In: Connective Tissue and its Heritable Disorders. Wiley-Liss; p. 103-147; 1993.
[6.] Rattner A, Sabido O, Le J, Vico L, Massoubre C, Frey J, Chamson A. Mineralization and alkaline phosphatase activity in collagen lattices populated by human osteoblasts. Calcif Tissue Int.; 66(1):35-42; Jan 2000.
[7.] Schantz JT, Teoh SH, Lim TC, Endres M, Lam CX, Hutmacher DW. Repair of calvarial defects with customized tissue-engineered bone grafts I. evaluation of osteogenesis in a three-dimensional culture system. Tissue Eng.; 9 Suppl 1:S113-26; 2003.
[8.] Kale S, Biermann S, Edwards C, Tarnowski C, Morris M, Long MW. Three-dimensional cellular development is essential for ex vivo formation of human bone. Nat Biotechnol.; 18(9):954-8; Sep 2000.
[9.] Niemeyer P, Krause U, Fellenberg J, Kasten P, Seckinger A, Ho AD, Simank HG. Evaluation of mineralized collagen and alpha-tricalcium phosphate as scaffolds for tissue engineering of bone using human mesenchymal stem cells. Cells Tissues Organs.; 177(2):68-78; 2004.
[10.] Buxton PG, Bitar M, Gellynck K, Parkar M, Brown RA, Young AM, Knowles JC, Nazhat SN. Dense collagen matrix accelerates osteogenic differentiation and rescues the apoptotic response to MMP inhibition. Bone.; 43(2):377-85; Aug 2008.
[11.] Mizuno M, Fujisawa R, Kuboki Y. Type I collagen-induced osteoblastic differentiation of bone-marrow cells mediated by collagen-alpha2beta1 integrin interaction. J Cell Physiol.; 184(2):207-13; Aug 2000.
[12.] Hing KA, Best SM, Tanner KE, Bonfield W, Revell PA. Mediation of bone ingrowth in porous hydroxyapatite bone graft substitutes. J Biomed Mater Res A.; 68(1):187-200; Jan 2004.
[13.] O'Loughlin PF, Morr S, Bogunovic L, Kim AD, Park B, Lane JM. Selection and development of preclinical models in fracture-healing research. J Bone Joint Surg Am.; 90 Suppl 1:79-84; Feb 2008.
[14.] Levy RJ, Schoen FJ, Sherman FS, Nichols J, Hawley MA, Lund SA. Calcification of subcutaneously implanted type I collagen sponges. effects of formaldehyde and glutaraldehyde pretreatments. Am J Pathol.; 122(1):71-82; Jan 1986.
[15.] Bodnar AG, Ouellette M, Frolkis M, Holt SE, Chiu CP, Morin GB, Harley CB, Shay JW, Lichtsteiner S, Wright WE. Extension of life-span by introduction of telomerase into normal human cells. Science.; 279(5349):349-52; Jan 1998.
[16.] Gerstenfeld LC, Cullinane DM, Barnes GL, Graves DT, Einhorn TA. Fracture healing as a post-natal developmental process: Molecular, spatial, and temporal aspects of its regulation. J Cell Biochem. 88(5):873-84; Apr 2003.
[17.] Caplan AI. Mesenchymal stem cells. J Orthop Res. 9(5):641-50; Sep 1991.
[18.] Jaiswal N, Haynesworth SE, Caplan AI, Bruder SP. Osteogenic differentiation of purified, culture-expanded human mesenchymal stem cells in vitro. J Cell Biochem. 64(2):295-312; Feb 1997.
[19.] Boehrs J, Zaharias RS, Laffoon J, Ko YJ, Schneider GB. Three-dimensional culture environments enhance osteoblast differentiation. J Prosthodont. 17(7):517-21; Oct 2008.
[20.] Dimitrijevich SD, Gracy RW, inventors Non-contracting tissue equivalent. 6471958. Oct 2002.
[21.] Nishiyama T, Tominaga N, Nakajima K, Hayashi T. Quantitative evaluation of the factors affecting the process of fibroblast-mediated collagen gel contraction by separating the process into three phases. Coll Relat Res. 8(3):259-73; May 1988.
[22.] Cornell CN. Osteoconductive materials and their role as substitutes for autogenous bone grafts. Orthop Clin North Am. 30(4):591-8; Oct 1999.
[23.] Sumanasinghe RD, Osborne JA, Loboa EG. Mesenchymal stem cell-seeded collagen matrices for bone repair: Effects of cyclic tensile strain, cell density, and media conditions on matrix contraction in vitro. J Biomed Mater Res A. 88(3):778-86; Mar 2009.
[24.] Engler AJ, Sen S, Sweeney HL, Discher DE. Matrix elasticity directs stem cell lineage specification. Cell. 126(4):677-89; Aug 2006.
[25.] Lode A, Bernhardt A, Gelinsky M. Cultivation of human bone marrow stromal cells on three-dimensional scaffolds of mineralized collagen: Influence of seeding density on colonization, proliferation and osteogenic differentiation. J Tissue Eng Regen Med. 2(7):400-7; Oct 2008.
[26.] Morgan DML. Tetrazolium (MTT) Assay for Cellular Viability and Activity In: Polyamine Protocols. Humana Press; p. 179-184; 1998.
[27.] Eslaminejad MB, Mirzadeh H, Nickmahzar A, Mohamadi Y, Mivehchi H. Type I collagen gel in seeding medium improves murine mesencymal stem cell loading onto the scaffold, increases their subsequent proliferation, and enhances culture mineralization. J Biomed Mater Res B Appl Biomater. 90(2):659-67; Aug 2009.
[28.] Holy CE, Shoichet MS, Davies JE. Engineering three-dimensional bone tissue in vitro using biodegradable scaffolds: Investigating initial cell-seeding density and culture period. J Biomed Mater Res. 51(3):376-82; Sep 2000.
[29.] Lutolf MP, Lauer-Fields JL, Schmoekel HG, Metters AT, Weber FE, Fields GB, Hubbell JA. Synthetic matrix metalloproteinase-sensitive hydrogels for the conduction of tissue regeneration: Engineering cell-invasion characteristics. Proc Natl Acad Sci USA. 100(9):5413-8; Apr 2003.
[30.] Gentleman E, Swain RJ, Evans ND, Boonrungsiman S, Jell G, Ball MD, Shean TA, Oyen ML, Porter A, Stevens MM. Comparative materials differences revealed in engineered bone as a function of cell-specific differentiation. Nat Mater. Jul 2009.
[31.] Weinand C, Pomerantseva I, Neville CM, Gupta R, Weinberg E, Madisch I, Shapiro F, Abukawa H, Troulis MJ, Vacanti JP. Hydrogel-beta-TCP scaffolds and stem cells for tissue engineering bone. Bone. 38(4):555-63; Apr 2006.
[32.] Mauney JR, Kirker-Head C, Abrahamson L, Gronowicz G, Volloch V, Kaplan DL. Matrix-mediated retention of in vitro osteogenic differentiation potential and in vivo bone-forming capacity by human adult bone marrow-derived mesenchymal stem cells during ex vivo expansion. J Biomed Mater Res A. 79(3):464-75; Dec 2006.
[33.] George J, Kuboki Y, Miyata T. Differentiation of mesenchymal stem cells into osteoblasts on honeycomb collagen scaffolds. Biotechnol Bioeng. 95(3):404-11; Oct 2006.
Anupam Ashutosh Sule, Arvind Nana, S. Dan Dimitrijevich
University of North Texas Health Science Center, Fort Worth, TX, USA
Received 27 November 2017; Accepted 30 December 2017; Published online 31 December 2017
* Coresponding author: Dr. Anupam Ashutosh Sule; E-mail: email@example.com
Caption: Figure 1: Characterization of collagen foam. (a)--Phase contrast light micrograph of collagen type I foam (Ultrafoam[R]) showing large irregular pores. (b)--Retention capacity after impregnation. This was determined by measuring weight of water, or gelled neutralized collagen type I solution, retained by dry Ultrafoam[R] after 24 hour immersion, and expressed as the gain in weight. (Mean and SEM, n=3)
Caption: Figure 2: Distribution of bone marrow derived hMSC in collagen type I impregnated Ultrafoam[R] (CGCS). 2A. Uniform distribution of hMSCs populating CGCS is shown after MTT treatment; formazan stains the entire specimen (dark purple). 2B. H&E stained, paraffin cross section (10m) of CGCS showing Ultrafoam[R] (dark purple), impregnated with collagen gel (light pink), and hMSC (dark blue nuclei marked by black arrows). The specimen was sectioned orthogonally to the seeding surface.
Caption: Figure 3: Confocal microscopy of 3-D model assembled with Cy5 labeled collagen type I, and GFP-tFbs. Optical slice, 100 mm superior to the basal surface shows uniform penetration and distribution of cells and collagen type I in the CGCS scaffold (Ultrafoam[R]). Unlabeled scaffold - Ultrafoam[R] (dark areas) in which pores contain Cy-5 labeled collagen type I gel (red), and GFP-tFbs cells
Caption: Figure 4: Proliferation and viability of GFP-tFbs and hMSC in CGCS determined by MTT Assay. (a)--Formazan absorbance (570 nm) provides a quantitative measure of proliferation, the number of viable hMSC residing in CGCS (56 days in culture). (Mean and SEM, n=3). (b)--Our 3-D model also supports long-term viability (30 days in culture) of cell types other than hMSC as shown by neutral red stained CGCS populated with GFP-tFbs. Light microscopy shows normal cell morphology (fluorescence in image 1), and high cell viability (image 2) in the same optical section.
Caption: Figure 5: The effect of the collagen type I presence on distribution, survival and proliferation of hMSC and hOST cells in CGCS. (a)--Cells in CGSC. Cultures of hMSC populating CGCS (1 x [10.sup.6] /mL), result in significantly higher cell number (seeding efficiency and proliferation) when compared to cultures on Ultrafoam[R] alone (2 way ANOVA, mean, SEM graphed. n=3, p < 0.05). (b)--Cytotracker[TM] Orange labeled hOST seeded on Ultrafoam[R] (row 1) and on CGCS (row 2). Day 1 images show inadequate delivery of cells in the absence of collagen solution; more cells at the surface (panel 1i) than 100 m below the surface (panel 1ii) of Ultrafoam[R]. Day 14 images show cell migration from the interior to the Ultrafoam[R] after prolonged culture period in the absence of collagen gel; more cells on the surface (panel 1iii) with no cells detected at 100 [micro]m below the surface (1iv). The overall increase in proliferation indicates cell death did not take place in the interior of the Ultrafoam[R]. (c)--Cells in CGCS. Day 1 confocal images at the surface (panel 2i) and 100 [micro]m below the surface (panel 2ii) CGCS respectively when compared with confocal images on day 14, demonstrate the presence of cells on the surface (panel 2iii) as well as in the interior (100 [micro]m) below the surface (panel 2iv).
Caption: Figure 6: Cell migration studies in 3-D using CGCS. (a) --Cell Migration in 3-D. Acellualar CGCS, placed at the edge of the well (12-well plate), was totally immersed in neutralized collagen type I solution and allowed to gel. GFP-tFbs model cells, suspended in neutralized collagen type I solution, were added into a small cavity created in the center of the well and the cell migration was observed over a 14-day period. (b)--Migration of GFP-tFbs. After 14 days in culture, cell migration from the central cavity was observed from top left to bottom right as shown in image 1. Cell migration from CGCS populated with GFP-tFbs into surrounding acellular collagen was also demonstrated from top left to bottom right as shown in image 2. The scale bars are 200 [micro]m. (c)--Migration of hOST Cells. Using the same methodology CytotrackerTM Orange labeled hOST migrated from central cavity (from bottom right) into the acellular CGCS (to top left). Image 1 on day 1 and image 2 is on day 14. The scale bars are 200 [micro]m.
Caption: Figure 7: The response of cells in CGCS to mechanical force. (a)--A Randomly orientated GFP-tFbs populating CGCS are shown in image 1. (b)--After brief stretch (10 seconds) in the direction of the arrow, followed by 24 hours resting period, the cells realigned with the long axes in the direction of the force (image 2) as shown by confocal microscopy with optical sectioning and assembly of the gallery of images ranging from 50 [micro], to 150 [micro], below the surface
Caption: Figure 8: Differentiation of hMSC in CGCS. (a)--Osteogenic differentiation was demonstrated colorimetric assay of alkaline phosphatase (ALP) activity, in which pNPP is converted to pNP (normalized to cellular DNA to compensate for hMSC proliferation), The maximum ALP activity observed on day 14 was not sustained and decreased by day 28. (Mean and SEM plotted, n=3). (b)--Deposition of extracellular glycosaminoglycan another marker of osteogenic differentiation of hMSC, is shown by Alcian blue staining (marked by black arrow) in CGCS at day 28.
Caption: Figure 9: Differentiation of hMSC in Ultrafoam[R] and CGCS. (a)--Immunohistochemical analysis shows that Osterix, a transcription factor associated with late stages of differentiation process (red fluorescence), is not expressed by hMSC on day 1 but is present after 28 days in culture under differentiation conditions in Ultrafoam[R] (panel 1) or CGCS (panel 2). Panel 3 is higher magnification of panel 2. Yellow arrowheads mark the positions of the cells as determined by nuclear staining with DAPI. (b)--Immunohistochemical analysis shows that Osteopontin (green fluorescence), an extracellular protein deposited by osteoblasts late in the differentiation process, is not observed on day 1 but is present after 28 days of culture under differentiation conditions (panel 2--Ultrafoam[R] and panel 2--CGCS). Yellow arrowheads mark the positions of the cells as determined by DAPI staining.
Caption: Figure 10: Collagen type I is required for in vitro mineralization in CGCS model (a) Histochemical Analysis under proliferation and differentiation conditions. Van Kossa stained paraffin cross sections (10 m) of CGCS populated with hMSC and maintained under proliferation condition; day 1 panel 1i (day 1) and day 28 panel 1ii. Van Kossa stained paraffin cross sections (10 m) of Ultrafoam[R] (without collagen impregnation) populated with hMSC and maintained under differentiation conditions; day 1 panels 2i and day 28 2ii. Van Cosaa stained paraffin cross sections (10 m) of CGCS populated with hMSC and maintained under differentiation conditions; day 1 panel, 3i and day 28 panel 3ii. Positive Von Kossa staining (black staining) observed in panel 3ii only, indicated calcification--active ossification under differentiation culture conditions. (b) Shows higher magnification of panel 3ii. (c) A significant increase in mineralization was observed after 28 days of in differentiation in CGCS (cross section images analyzed using Matlab) (Mean and SEM n=3, unpaired t-test p < 0.001).
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Original Article|
|Author:||Sule, Anupam Ashutosh; Nana, Arvind; Dimitrijevich, S. Dan|
|Publication:||Trends in Biomaterials and Artificial Organs|
|Date:||Oct 1, 2017|
|Previous Article:||[beta]-Dicalcium Silicate Cement Modified with [beta]-tricalcium Phosphate: In vitro Bioactivity and Mechanical Strength.|
|Next Article:||Diamond-Like Carbon Coating for Corrosion Protection of Metallic Implants.|