pISSN 1738-2696 · eISSN 2212-5469 http://dx.doi.org/10.1007/s13770-014-9945-6

ORIGINAL ARTICLE

Sequential Differentiation of Human Bone Marrow Stromal Cells for Bone Regeneration Eva Johanna Huebner1*, Nestor Torio Padron2, David Kubosch1, Guenter Finkenzeller2, Norbert P. Suedkamp1, Philipp Niemeyer1 Department of Orthopedic Surgery and Traumatology, Freiburg University Hospital, Freiburg, Germany Department of Plastic and Hand Surgery, Freiburg University Hospital, Freiburg, Germany

1 2

In this study we hypothesized that as a simulation of endochondral bone formation, bone marrow stromal cell (BMSC) provide a sequential chondro-osteogenic differentiation potential. A chondrogenic priming of BMSC leads to a spontaneous three-dimensional cell formation. BMSC were chondrogenically differentiated prior to an osteogenic stimulation. Duration of cell culture was 28 days, whereas in group A BMSC were chondrogenically differentiated for 1 day, followed by an osteogenic differentiation for 27 days. In group B BMSC were chondrogenically differentiated for 14 days prior to an osteogenic differentiation of 14 days and group C BMSC were differentiated chondrogenically for 28 days serving as a chondrogenic control group. Chondrogenic priming induced a spontaneous three-dimensional cell formation. To survey the stability of the osteogenic phenotype in the absence of an osteogenic stimulus, investigations were performed in vivo in a specially adapted chorioallantoic membrane model of fertilized White Leghorn eggs. Histology and real time polymerase chain reaction revealed a higher amount of osteogenic extracellular matrix synthesis and significant higher expressions of osteogenic marker genes in group B after 14 days of chondrogenic and 14 days of osteogenic stimulation. Matrix calcification in vivo in the absence of an osteogenic stimulus could be demonstrated. The results of the present study support the theory of a sequential differentiation potential of BMSC. A chondrogenic priming of BMSC stimulated into the osteogenic lineage result in a stable osteogenic phenotype in a scaffold-free, three-dimensional tissue engineering application. Tissue Eng Regen Med 2015;12(5):331-342 Key Words: Tissue engineering; Bone marrow stromal cells; Bone regeneration; Bone defect; Pellet culture

INTRODUCTION Treatment of large bone defects caused by trauma, degenerative or congenital diseases and cancer is one of the most challenging subjects in current orthopaedic and traumatological research. A number of artificial bone materials have been tested in recent years. Artificial bone grafts, most of them based on calcium and phosphate, which are components of natural bone, usually only provide osteoconductive properties, which allow adjacent osteoblasts to migrate [1]. Allogenous transplants may potentially be infectious [2,3], and the sterilization processes prior to transplantation cause a loss of osteoinductive properties [4,5]. Although the amount of autogenous cancellous bone available is limited, and graft harvesting, for example from the iliac crest leads to significant morbidity [6], autogenous bone Received: June 2, 2014 Revised: April 6, 2015 Accepted: April 20, 2015 *Corresponding author: Eva Johanna Huebner, Department for Orthopedic Surgery and Traumatology, Freiburg University Hospital, Hugstetter Str. 55, D-79106 Freiburg, Germany. Tel: 49-761-270-24010, Fax: 49-761-270-25200 E-mail: [email protected]

grafts, combining osteogenic, osteoconductive, and osteoinductive properties must be accepted as the gold standard for bone replacement for now. The known disadvantages of the grafting procedure underline the high expectations directed towards cell-based bone tissue engineering. Bone marrow stromal cell (BMSC) represent a promising approach for cell-based tissue engineering. BMSC can be harvested from human bone marrow (BM) aspirates, e.g., from the iliac crest, from the anterior tibial metaphysis or from the femoral bone and be easily expanded in vitro [7-9]. BMSC can differentiate into the osteogenic, chondrogenic and the adipogenic lineage both in vitro and in vivo and, in contrast to embryonic stem cells, they are not under ethical scrutiny [10,11]. A major approach in generating bone substitutes from BMSC is an in vitro stimulation towards the osteogenic lineage aimed at better osteogenic potential prior to transplantation. In contrast to a spontaneous three-dimensional (3-D) alignment of BMSC that are stimulated into the chondrogenic lineage [12], an osteogenic stimulation does not cause any 3-D cell embedding. BMSC cultured under osteogenic conditions remain in monolayer forming detached nodules of cells secreting osteo© The Korean Tissue Engineering and Regenerative Medicine Society

331

Huebner et al. Chondro-Osteogenic Differentiation of BMSC for Bone Regeneration

specific matrix-proteins [13]. A three-dimensional alignment of osteogenic pre-stimulated cells can currently only be obtained by seeding cells into a suitable matrix or by cultivation on a scaffold before in vivo transplantation. Up to now the amount, homogeneity and stability of the bone formed by tissue engineering approaches are still insufficient for clinical application. Neither in vitro manipulated BMSC prior to transplantation in vivo nor undifferentiated BMSC inserted into a bony defect situation achieved results comparable to autologous cancellous bone grafting, which is deemed the current gold standard. In vivo investigations were performed to examine the stability of the osteogenic phenotype of the chondro-osteogenic induced BMSC in the absence of an osteogenic stimulus in a specially adapted chorioallantoic membrane (CAM) model of fertilized White Leghorn eggs [14]. Steffens et al. [15] were able to demonstrate a complex 3-D network of perfused human neovessels in a spheroidal coculture system consisting of human primary endothelial cells and human primary osteoblasts in the CAM model and identify it as a promising approach for improving vascularization in bone tissue engineering. The CAM model of the chick embryo is one of the most widely used in vivo assays for the evaluation of normal angiogenic processes, of malignant invasion as aspects of tumor biology or investigations of ocular, mucosal or vascular toxicity of applied test substances [16,17]. Advantages of the CAM model include easy access to the CAM vascular network, lack of immunocompetence and low costs. In vivo investigations were performed to establish the CAM model as an alternative to animal experiments as a semi-in vivo model for bioengineered tissue applications and as a possible model for further investigation of neovascularization in bone tissue engineering [15,18]. Former investigations confirmed an improvement of osteospecific matrix synthesis in 3-D embedded BMSC seeded on mineralized collagen sponges. A 3-D cell alignment caused significantly higher secretion of an extracellular matrix rich in calcium and phosphate, as well as the expression of osteogenic genes like alkaline phosphatase, bone morphogenic protein 2 (BMP 2), osteocalcin and bone sialoprotein compared to BMSC stimulated into the osteogenic lineage in monolayer culture [19]. Looking for reasons for the inferior quality of bioengineered tissue applications, a pheno-typical instability of stimulated BMSC has to be taken into con-sideration [20]. An osteogenic stimulation induced by β-glyce-rolphosphate, ascorbic acid and dexamethason may lead to the secretion of osteo-specific matrix proteins while 3-D cell embedding is lacking [21]. Based on these findings the goal of this study was to use the spontaneous 3-D alignment of BMSC that are stimulated into

332

Tissue Eng Regen Med 2015;12(5):331-342

the chondrogenic lineage [12] as a scaffold-free osteogenic tissue application imitating the process of endochondral ossification to investigate the osteogenic potential of BMSC after sequential chondro-osteogenic stimulation. This study aimed to compare different study protocols with regard to the optimal duration of chondrogenic priming. Based on the physiological procedures of osteogenesis by endochondral ossification we hypothesized that BMSC provide a sequential chondro-osteogenic differentiation potential. By a chondrogenic priming of BMSC a spontaneous 3-D cell formation is achieved. We hypothesized that a sequential chondro-osteogenic stimulation of BMSC towards the endochondral ossification process may combine a 3-D cell alignment and may result in a stable osteogenic tissue engineered construct.

MATERIALS AND METHODS Cell samples

All procedures followed were in accordance with the ethical standards and were approved by the local ethics board (042/ 2000, 251/2002, “Tissue bank for research in the field of tissue engineering” GTE-2002). After informed consent from each patient included in the study, BM aspirates (10–30 mL) were obtained from hematologically healthy donors (n=4) during routine orthopaedic surgical procedures involving exposure of the iliac crest. BM was immediately transferred into plastic tubes containing 10000 IU heparin. BMSC were isolated as published elsewhere, with minor variations [22]. Briefly, BM mononuclear cells were purified by Biocoll density gradient centrifugation (d=1.077 g/L; Biochrom, Berlin, Germany). Mononuclear cells were plated in expansion medium at a density of 100000 cells/cm2 in tissue culture flasks (Nunc, Wiesbaden, Germany). The expansion medium used consisted of 60% low glucose DMEM (Biowhittaker, Apen, Germany), 38% MCDB 201 (Sigma, Taufkirchen, Germany), 2% fetal calf serum (FCS) supplemented with 2 mM l-glutamin, 100 U/mL penicillin/streptomycin, 1x insulin transferrin selene [24], 1x linoleic acid, 20 nM dexamethasone, 0.1 mM l-ascorbic acid 2-phosphate, platelet-derived growth factor and epidermal growth factor (10 ng/mL each; Peprotech, New York, USA). The medium was changed twice a week. Cells were expanded for three to five passages.

BMSC characterization

To confirm their BMSC character according to the criteria of the Stem Cell Committee [23], cells were analyzed by flow cytometry with a panel of cell surface markers (Table 1). FACS analysis revealed a similar cluster of differentiation (CD) pattern as published elsewhere [24]. The ability to differentiate into os-

teogenic, chondrogenic and adipogenic lineages was confirmed in monolayer culture or in pellet culture at passages 3 or 4.

Sequential chondro-osteogenic differentiation

Trypsinized BMSC at passages 3 or 4 were resuspended in chondrogenic differentiation medium at a final concentration of 1×10 cells/mL, and 1 mL of the cell suspension was placed into individual wells of a 48-well plate. Cells derived from BM aspirates isolated from 4 different patients were used. Experiments were performed in quadruplicate. The chondrogenic differentiation medium used consisted of high glucose DMEM (Gibco), supplemented with 2 mM l-glutamin (Biowhittaker), 100 U/mL penicillin/streptomycin (PAA Laboratories), 1.25x insulin transferrin selene, 1x linoleic acid, 100 nM dexamethasone, 0.2 mM l-ascorbic acid 2-phosphate, TGF-β3 10 ng/mL (all from Sigma). A chondrogenic priming of BMSC leads to a spontaneous 3-D cell formation. Cells were incubated at 37°C under humidified conditions and 5% carbon dioxide (CO2) in chondrogenic differentiation medium for 24 hours and afterwards for 27 days in osteogenic medium (group A). Cells of group B were cultured under chondrogenic conditions for 14 days, subsequently they were exposed to osteogenic differentiation medium [90% DMEM low glucose (Biowhittaker)], 10% FCS (Gibco), 100 U/mL penicillin/streptomycin (PAA Laboratories), 0.2 mM l-ascorbic acid 2-phosphate, 100 nM dexamethasone, 10 mM β-glycerophosphate (all from Sigma) supplemented with 2 mM l-glutamin (Biowhittaker) for another 14 days. Cells of group C were stimulated to the chondrogenic lineage for 28 days.

Histology

For paraffin sections cell pellets were dehydrated, embedded in paraffin and sections were cut dry (5 and 7 μm) on a Leica RM 2165 microtome. Slices were stained with hematoxylin eosine (HE) (Merck and Shandon Eosin Y, Runcorn, UK), with von Kossa and Safranin O stains (0.2 g Safranin T Fluka, 100 mL Aqua dest., 1 mL ethanoic acid; counterstained with Fast Green: 0.04 g Chroma 1A 304, 100 mL Aqua dest., 0.2 mL ethanoic acid). Histological investigations were performed at day 14, 21, and 28 of culture.

In situ detection of cell death

Cell death was detected by TdT-mediated dUTP nick-end labeling (Fluorescein FragELTM DNA Fragmentation Detection Kit, QIA 39, Calbiochem, Germany). Staining was performed according to the manufacturers’ guidelines and analyzed by means of a flourescence microscope. Positive controls were generated by DNAse I-digestion (1 μg/μL DNAse I in 1x TBS/1 mM MgSO4) prior to the DNA labeling. Flourescence micros-

copy was performed using a filter for DAPI, 330–380 nm, to visualize the total cell population and a standard flourescein filter, 465–495 nm to visualize labeled nuclei.

RT-PCR

For quantitative assessment of matrix properties detected histologically, real-time polymerase chain reaction (RT-PCR) analysis was performed. RNA samples were taken at day 28, n=4 in every group, transcribed into cDNA, and RT-PCR for gene expression was carried out as described elsewhere [24]. Total RNA was prepared using the RNeasy Mini extraction kit according to the manufacturers’ instructions (Qiagen GmbH, Hilden, Germany). Random-primed cDNA synthesis was performed using 1 μg total RNA and the Omniscript Reverse Transkriptase Kit (Qiagen, Hilden, Germany) according to the manufacturers’ instructions. PCR primers are shown in Table 2. Quantitative PCRs were performed in a LightCycler instrument (Roche Diagnostics, Mannheim, Germany). After an initial denaturation step (95°C for 10 min), amplification was performed for 40 cycles (95°C for 1 sec, at a primer specific annealing temperature for 7 sec, and 72°C for 14 sec). Data were analyzed using the relative standard curve method, with each sample being normalized to glyceraldehyde-3-phosphate dehydrogenase. Relative expression was calculated using averages of duplicates of four samples for each group. Comparison specimens of BMSC stimulated into the osteogenic lineage in monolayer culture for 28 days were investigated as mentioned above without establishing the expression of Col2a1 and cartilage oligomeric matrix protein (COMP) genes. Data are expressed as arbitrary units. The quantitative RT-PCR was performed in cooperation with the clinic for hematology and oncology, University of Heidelberg, Germany.

In vivo CAM model

To evaluate the stability of the osteogenic phenotype in the absence of an osteogenic stimulus, investigations were perTable 1. Cell surface markers

Positive

Negative

CD13 CD44 CD73 CD90 CD105

CD34 CD45 HLA-DR (MHC-II) CD40 CD40-Ligand CD80 CD86 (B7-2)

BMSC-characteristic expression pattern of cell surface antigens in flow cytometric analysis. BMSC are positive for mesenchymal markers (+) and negative (-) for haematopoietic markers [23]. BMSC: bone marrow stromal cell www.term.or.kr 333

Huebner et al. Chondro-Osteogenic Differentiation of BMSC for Bone Regeneration

formed in vivo in a specially adapted chorioallantoic membrane model [14]. Cell pellets of group B consisting of about 1×10 cells at a time, after in vitro chondro-osteogenic priming for 14 days each, were transplanted onto the chorioallantoic membrane of fertilized eggs. Cells derived from BM aspirates isolated from 3 different patients were used in triplicate test preparation. Fertilized White Leghorn chick eggs were incubated at 37.8°C and 60–65% relative humidity in a horizontal position in an incubator (BSS 160, Grumbach, Asslar, Germany). On day three of incubation, a round window corresponding to the cylinder diameter was cut through the egg shell with a small circular saw after removal of 1–2 mL of albumen so as to detach the developing CAM from the shell.

Cylinder implantation and explantation

On day 8 of incubation, a plastic cylinder was placed onto the CAM by pushing it through the window until the CAM became attached to its edges by cohesive forces. In vitro primed cell pellets were put onto the CAM, cylinders were covered with matching cylinder caps and eggs were returned to the incubator. The experiment was determined at day 16 (corresponding to day 8 after transplantation) and the cylinder, including cell pellet and adherent CAM, was removed. The cell-matrix composite, including the adherent CAM was detached from the cylinder and processed for further histological and immunohistochemical examination.

Histology and immunohistochemistry

Cell-CAM-constructs were embedded into paraffin, cross-

sections, vertical to the CAM; they were drawn and stainings of HE, von Kossa and Safranin O were performed as described above. The number of viable cells after 28 days of in vitro high density pellet culture under chondro-osteogenic differentiation conditions and 8 days of in vivo culture without an osteogenic stimulus, was determined using the dead alive assay Fluorescein FragELTM DNA Fragmentation Detection Kit (QIA 39, Calbiochem, Germany) as mentioned earlier. For detection of human BMSC after in vitro and in vivo investigations, antivimentin immunohistochemistry was performed with a human-specific antibody according to the manufacturers’ instructions. Controls were performed using human skin as a positive control. Slices of the chorioallantoic membrane without transplanted human cells served as negative controls. Osteogenic differentiation was assessed by immunohistochemical detection of osteocalcin in cell-CAM-constructs at the endpoint of investigations. Analogue to other stainings, samples were incubated with the primary antibodies against osteocalcin (Biogenex, AM386-5M, GX Duiven, the Netherlands). Negative controls were performed using the goat anti mouse secondary antibodies and AEC chromogen substrate to exclude an unspecific bonding. All procedures were performed according to the manufacturers’ protocols.

Statistical analysis

SPSS for windows version 16.0 (SPSS Inc., Chicago, IL, USA) was used for the statistical analysis to work up the data ascertained in this study. After verification of normality, the paired Student t-test was used to evaluate significant differences be-

Table 2. Primers used for RT-PCR analysis

Gene name

Primer

Forward 5’-TGG GCT ACA CTG AGC ACC AG-3’ Reverse 5’-CAG CGT CAA AGG TGG AGG AG-3’ Forward 5’-ATG CCC TGG AGC TTC AGA AG-3’ Alkaline phosphatase Reverse 5’-TGG TGG AGC TGA CCC TTG AG-3’ Forward 5’-GGC AGC GAG GTA GTG AAG AGA C-3’ Osteocalcin Reverse 5’-GGC AAG GGG AAG AGG AAA GAA G-3’ Forward 5’-AAC ACT GTG CGC AGC TTC C-3’ BMP 2 Reverse 5’-CTC CGG GGT GGT TTC CCA C-3’ Forward 5’-ATG TTC AGC TTT GTG GAC CTC-3’ Col-1 Reverse 5’-AGT TTG AAG CAC AGC ACT CG-3’ Forward 5’-GAG ACA GCA TGA CGC CGA G-3’ Col2a1 Reverse 5’-GCG GAT GCT CTC AAT CTG GT-3’ Forward 5’-CAA TGA ACA GCG ACC CAG G-3’ COMP Reverse 5’-TCA CAT GGA ACG TGC CCT C-3’ Forward 5’-TCC TTG AAC TTG GTT CAT GGA GT-3’ Col10a1 Reverse 5’-ACT GTG TCT TGG TGT TGG GTA GTG-3’ GAPDH: glyceraldehyde-3-phosphate dehydrogenase, BMP 2: bone morphogenic protein type 2, COMP: cartilage oligomeric matrix protein, RT-PCR: real-time polymerase chain reaction GAPDH

334

Tissue Eng Regen Med 2015;12(5):331-342

tween gene expression in relation to the duration of chondrogenic priming of BMSC. p-values<0.05 (indicated by “*”) were considered statistically significant, those at <0.01 (indicated by “†”) were considered strongly significant.

RESULTS BMSC characterization

To confirm their BMSC character according to the criteria of the Stem Cell Committee [23], cells were analyzed by flow cytometry with a panel of cell surface markers (Table 1). FACS analysis revealed a similar CD pattern as published elsewhere [23].

A  

B

C

D

E

F

Figure 1. Cell pellets of group B: (A, B, and C) Von Kossa staining, (D, E, and F) Safranin O staining (counterstained with Fast Green). (A and D) BMSC after 14 days of chondrogenic induction, LM 20×. (B and E) BMSC after 14 days of chondrogenic and 7 days of osteogenic induction, LM 10×. (C and F) BMSC after 14 days of chondrogenic and 14 days of osteogenic induction, LM 10×. (A, B, and C) Show an increase in osteogenic matrix properties, (D, E, and F) an antidromic decrease of chondrogenic proteoglycans. BMSC: bone marrow stromal cell, LM: light microscope.

A

B

Figure 2. BMSC after 14 days of chondrogenic and 7 days of osteogenic induction, LM 20×. (A) Von Kossa staining with hypertrophic chondrocytes in the central part of the cell pellet. (B) In Safranin staining in the central part of the pellet proves chondrogenic proteoglycans. BMSC: bone marrow stromal cell, LM: light microscope. www.term.or.kr 335

Huebner et al. Chondro-Osteogenic Differentiation of BMSC for Bone Regeneration

Chondro-osteogenic differentiation in BMSC pellet culture

vested and processed intact for histology. BMSC pellets cultured under chondrogenic conditions for merely 24 hours prior to an osteogenic subculture for further 27 days (group A) showed a less densely packed structure and by day 14, 21, and

BMSC cultured in the presence of TGF-β3 generated a solid 3-D, non plastic-adherent tissue structure that could be harCol-1

30 25

30 25

*

20

*

BMP 2 *

30 25

*

20 15

15

10

10

10

5

5

5

0

A

B

C

Osteocalcin

30 25

*

20

0

B

A

B

*

15

15

10

10

10

5

5

5

0

0

A

B

C

E

A

B

*

B

*

C

COMP *

20

15

D  

A

25

*

20

0

C 30

†

25

*

C

Col2a1

30

*

*

20

15

A  

Alkaline phosphatase

C

0

F  

A

B

C

Col10a1

30 25

*

20 15 10 5 0

G

A

B

C

Figure 3. Quantitative gene expression after 28 days in BMSC pellet culture. (A) Quantitative gene expression of Collagen 1 (Col 1) after 28 days in BMSC pellet culture. (B) Quantitative gene expression of bone morphogenic protein type 2 (BMP 2) after 28 days in BMSC pellet culture. (C) Quantitative gene expression of Alkaline Phosphatase (AP) after 28 days in BMSC pellet culture. (D) Quantitative gene expression of Osteocalcin (OC) after 28 days in BMSC pellet culture. (E) Quantitative gene expression of Collagen 2a1 (Col2a1) after 28 days in BMSC pellet culture. (F) Quantitative gene expression of cartilage oligomeric matrix protein (COMP) after 28 days in BMSC pellet culture. (G) Quantitative gene expression of Collagen 10a1 (Col10a1) after 28 days in BMSC pellet culture. Group A: 1 day chondrogenic and subsequently 27 days of osteogenic induction, group B: 14 days of chondrogenic and osteogenic induction, respectively, group C: 28 days of chondrogenic induction (x-axis, arbitrary units). Multiple of mRNA-expression of undifferentiated BMSC cultured in expansion medium at day 1 (y-axis). Significances are given in the results section. pvalues<0.05 (indicated by “*”) were considered statistically significant, those at <0.01 (indicated by “†”) were considered strongly significant. BMSC: bone marrow stromal cell.

A

B

C

D

Figure 4. (A-D) Dead alive assay, BMSC after 14 days of chondrogenic and 14 days of osteogenic induction (group B), fluorescence microscopy 10×. (A and C) Dapi filter 330–380 nm, viable cells are detected as a broad peripheral collar. (B and D) Fluorescein filter 465–495 nm, apoptotic cells are detected in particular in central regions of the tissue bead. BMSC: bone marrow stromal cell.

336

Tissue Eng Regen Med 2015;12(5):331-342

28 a significant smaller amount of calcified extracellular matrix in von Kossa stainings compared to group B as well as a still larger amount of chondrogenic proteoglycans proved by Safranin O stainings. BMSC cultured in chondrogenic and subsequently in osteogenic medium each for 14 days (group B) performed antidromically with regard to the increase of mineralized osteogenic matrix and decrease of chondrogenic pro30

Col 1 BMP 2 AP OC

20

10

14 days chondrogenic and 14 days of osteogenic differentiation in pellet

28 days of osteogenic differentiation in monolayer culture

Figure 5. Quantitative gene expression after 28 days in BMSC pellet culture and after 28 days in monolayer culture, respectively (x-axis, arbitrary units). Y-axis: relative mRNA expression (arbitrary units 0–30). BMP 2: bone morphogenic protein 2, AP: alkaline phosphatase, OC: osteocalcin, BMSC: bone marrow stromal cell.

teoglycans identified by examination of cell pellets at days 14, 21, and 28. After 14 days of chondrogenic priming, Safraninpositive hyaline cartilage could be demonstrated. By day 21 pellets of group B showed a wider peripheral zone negative for chondrogenic proteoglycans, encasing a central core of histologically well-defined, Safranin-positive hyaline cartilage. By day 28 no chondrogenic proteoglycans were detected in the pellets of group B, but rather a homogeneously distributed, bone-like calcified matrix was found. By day 21, in von Kossa stainings calcium/calcium salt was detected even in central parts of the pellets (Fig. 1). Besides morphological evaluation of cells in the central cartilaginous region of the tissue construct our findings of type X collagen expression by PCR indicated progression to hypertrophy of chondrocytes (Figs. 2 and 3). Neither a dual chondro-osteogenic differentiation of BMSC nor evidence for hypertrophy of chondrocytes was found in the control samples of group C. Evaluation of dead alive assay showed a broad peripheral collar of viable cells in pellet tissue constructs. Apoptotic cells were detected in particular in central regions of the tissue bead in all groups investigated (Fig. 4).

RT-PCR

Histological results were validated by RT-PCR and matrix properties were evaluated quantitatively. The expression of car-

A

B

C

D

E

F

Figure 6. Work steps of the CAM model. (A) Opening of the egg shell with a circular saw. (B) Preparation of a round opening for the plastic cylinder. (C) Area vasculosa on day three of incubation. (D) Implantation of the plastic cylinder and the in vitro primed cell pellets. (E) Closed cylinder and re-incubation. (F) Cell-matrix composites are placed onto a cell filter for further histological and immunohistochemical examination. CAM: chorioallantoic membrane. www.term.or.kr 337

Huebner et al. Chondro-Osteogenic Differentiation of BMSC for Bone Regeneration

tilage and bone matrix proteins and the expression of signal transduction proteins involved in the process of endochondral bone formation were investigated by RT-PCR at day 28. RTPCR analysis was performed for mRNA expression of alkaline phosphatase (AP), osteocalcin, collagen type I, BMP 2, Col10a1, Col2a1, and COMP. Cell constructs of group B compared to those of group A showed a significantly higher expression of osteogenic marker genes such as BMP 2 (p=0.02), AP (p=0.02), osteocalcin (OC) (p=0.02) and higher expressions of chondrogenic marker genes (Col2a1: p=0.02, COMP: p=0.04) as well as significant elevation of Col10a1 indicating hypertrophic chondrocytes (p=0.04). Group B samples showed significantly higher gene expressions of characteristic bone matrix proteins than the control samples of group C stimulated into the chondrogenic lineage for 28 days such as Col-1 (p=0.04), BMP 2 (p= 0.003), AP (p=0.02), and OC (p=0.01). Expression of BMP 2 and OC mRNA was significantly higher in group A than in group C (p=0.04 resp. p=0.01), expression of mRNA of chondrogenic genes was significantly higher in group C than in group A (Col2a1: p=0.0007, COMP: p=0.003) (Fig. 4). Gene expression in group B was at least comparably high as in cells differentiat-

ed into the osteogenic lineage in monolayer culture (Fig. 5).

Chorioallantoic membrane model

To survey the stability of the osteogenic phenotype in the absence of an osteogenic stimulus, cell pellets of group B after in vitro chondro-osteogenic priming were placed onto the CAM on day 8 of incubation (Fig. 6). After 8 more days of incubation pellets were adherent to and overgrown by the CAM. HE stainings revealed vessel ingrowth in peripheral areas of the cell pellet (Fig. 7), whereas dead alive assays showed viable cells especially around ingrown vessels in a peripheral collar (data not shown). Extensive matrix calcification including more than 50% of the pellet was assessed by von Kossa stainings and the absence of any chondrogenic proteoglycans in Safranin O stainings (Fig. 7). Although there was inter-donor variability in this study cell pellets showed distinct histological and immunohistochemical results. Immunohistochemical evidence of osteocalcin secretion was consistent with calcium-positive areas in von Kossa stainings and occurred in areas facing towards the CAM (Fig. 7).

A

B

C

D

E

F

G

H

I

J

K

L

Figure 7. BMSC after 14 days of chondrogenic and osteogenic differentiation respectively, in vitro, subsequent cultivation onto the CAM for a further 8 days without any osteogenic stimulus. (A-D) HE staining. Vessel ingrowth was detected in HE stainings in peripheral parts of the cell pellets. (E-H) Von Kossa staining. High amounts of calcium/calcium salt are detectable in von Kossa stainings in different cell pellets in the CAM facing parts. (I and J) Safranin-O-staining (counterstained with Fast Green). Osteogenic matrix properties are detectable in the absence of chondrogenic proteoglycans in each pellet, here shown paradigmatically for 2 different pellets. (K and L) Osteocalcin staining. Osteocalcin is detectable in phosphate positive, the CAM facing areas of the cell pellets. BMSC: bone marrow stromal cell, CAM: chorioallantoic membrane.

338

Tissue Eng Regen Med 2015;12(5):331-342

Negative controls were performed using the goat anti mouse secondary antibodies and AEC chromogen substrate to exclude an unspecific bonding (data not shown). Human vimentin was detected in all samples spread homogeneously over the cell pellets (data not shown).

DISCUSSION In this article, the possibility of a sequential chondro-osteogenic differentiation of BMSC was assessed as an imitation of endochondral ossification according to different study protocols to investigate the osteogenic potential of BMSC for bone tissue engineering approaches. Investigations therefore focused on the spatial and temporal patterns of chondro- and osteogenesis in an in vitro high density pellet culture that mimics the process of the endochondral ossification seen in the growth plate during longitudinal bone growth. As phenotypic variability is observed over time in clonal cell strains in culture [20] and expression of an in vitro phenotype reminiscent of osteogenic cells does not necessarily predict the ability of a given cell strain to generate bone upon in vivo transplantation [21,25], we supposed that stable three-dimensional cell embedding by chondrogenic priming of BMSC might lead to a stable osteogenic phenotype even in an ectopic in vivo setting. The importance of a stable 3-D cell configuration was proven by our results as pellets sequentially exposed to chondrogenic and osteogenic medium for 14 days respectively, showed significantly higher gene expression of osteogenic matrix proteins and histologically higher amounts of matrix mineralization than cells that were exposed to chondrogenic conditions for merely 24 hours before a further osteogenic induction. In fact, chondrogenic priming of merely 24 hours did not improve the osteogenic potential of BMSC. After this short period of chondrogenic stimulation, cells do not seem to be able to form in an effectual amount or establish a stable extracellular matrix. Gene expression levels of osteogenic matrix proteins did not achieve the same results as in group B nor as for cells stimulated into the osteogenic lineage in monolayer culture as we had performed earlier. As emphasized in the last years the role of extracellular matrix is far beyond being a mechanical scaffold because it is involved in the complex system of cell-matrix interactions, hence involved in processes of growth and differentiation [26], likewise dysfunction of cell-matrix interactions might cause diseases [27]. Therefore, it can be assumed that the osteogenic monolayer culture was just a first step towards the osteogenic differentiation of BMSC in vitro but it cannot reflect physiological processes. Our pellet culture system as a model of endochondral ossification provides a relatively simple and rea-

sonable in vitro assay for a better understanding of osteogenesis and the osteogenic capacity of BMSC. Nevertheless, our current protocols require improvement not only in differentiation efficiency but also in nutrient limitations in larger constructs. Chondrogenic primed BMSC, subsequently stimulated into the osteogenic lineage showed better results in cell viability than cells directly differentiated to the osteogenic lineage, but stable 3-D cell embedding takes time. Our data suggest a higher osteogenic capacity of BMSC that are exposed to chondrogenic medium for 14 days in comparison to those primed chondrogenically for merely 1 day. Although we had to deal with the issue of core degradation, arising from lack of nutrient delivery to and waste removal from the center of tissue-engineered constructs–a major concern in the field of tissue engineering, cell pellets of group B showed comparable amounts of osteogenic gene expression as BMSC in osteogenic monolayer culture. Histological results revealed hypertrophic chondrocytes in the center of the cell pellets after 14 days of chondrogenic priming, followed by 7 days of osteogenic differentiation which prove the metabolic activity of chondrocytes involved in a process of chondrocyte maturation. An increase in the rate of apoptotic cells might therefore be related to an increase of matrix calcification caused by glycerolphosphate which would be in line with published data by Farrell et al. [28] suggesting a related release of vascular endothelial growth factor (VEGF) by the amount of collagen type II and X expression as well as the release of matrix metalloproteinases (MMP). Matrix mineralization inhibiting the expression of collagen type II and X would therefore lead to a decreasing amount of VEGF and a lack of vascularization [28]. However, matrix mineralization by adding a phosphate donor leads to significantly higher rates of cell apoptosis probably caused by biochemical stress [29] and Kuznetsov et al. [30] showed that long time exposure to FCS might decrease the capacity of BMSC to form bone. Interestingly, we detected a broad collar of viable cells in the peripheral proportion of cell pellets by dead alive assays questioning the theory about cell apoptosis caused by biochemical stress as the outer cells stay in direct contact with the differentiation medium. On the other hand, paths of diffusion are shorter and oxygenic maintenance should be better in peripheral than in central regions of the densely packed cell constructs. The difficulties of a sufficient nutrient delivery to central parts of tissue-engineered constructs have yet to be solved and physiological processes will only be understood by improving vascularization of cell constructs. Cartilage, as a bradytrophic tissue, can survive in hypoxic environments [31] and produces anabolic and catabolic factors that are important for the conversion of avascular tissue to vascularized tissue. Farrell et al. [28] demonstrated that BMSC www.term.or.kr 339

Huebner et al. Chondro-Osteogenic Differentiation of BMSC for Bone Regeneration

stimulated into the chondrogenic lineage show a higher expression of VEGF and MMP release than cells switched to a culture medium containing glycerolphosphate. MMPs play an important role in the process of endochondral ossification such as matrix degradation, vascularization and matrix remodelling when it comes to bone formation. Against the background of the ongoing discussion about hypertrophic chondrocytes’ fate [32] the internal remodelling of cartilage into bone in our system might reflect either a direct phenotypic conversion of chondrogenic cells to an osteogenic phenotype or an osteogenic cell population being replaced by a chondrogenic one. Former investigations in vitro and in vivo, suggesting an eventual osteogenic fate of hypertrophic chondrocytes [14,33] are consistent with our observations concerning a morphologic evaluation of the chondro-osteogenic cell pellets showing hypertrophic chondrocytes, as well as the evaluation of gene expression revealing significantly higher expressions of Col10a1 in group B than in group A. The stimulation of BMSC towards the chondrogenic lineage can be performed to different stages of differentiation and maturation, namely until a cartilaginous extracellular matrix has been formed or when the extracellular matrix is mineralized, which is often seen as something to be avoided in cartilage tissue engineering [34]. Therefore, more recently, the potential benefit of the endochondral ossification approach to bone tissue engineering by chondrogenic priming in vitro has been realized [35]. A modified chorioallantoic membrane model was used as a model for scrutinizing the phenotypic stability of sequentially chondro-osteogenic differentiated cells in the absence of an osteogenic stimulus in vivo. Cell pellets sequentially exposed to chondrogenic and osteogenic medium for 14 days respectively showed increasing matrix mineralization as proven by von Kossa stainings and evidence of osteocalcin secretion in immunohistochemical stainings after 8 days of incubation in vivo onto the CAM. Anti-vimentin stainings revealed human cells after more than 5 weeks of in vitro and in vivo investigations. Avian vessel ingrowth from the CAM was detected microscopically in HE stainings, and dead-alive assays showed viable cells in their surroundings. Although there was an inter-donor variability in this study cell pellets showed matrix calcification including up to more than 50% of the slices’ surface. The results provide proof of the phenotypic stability of chondrogenic primed BMSC stimulated towards the osteogenic lineage in an ectopic in vivo setting. Furthermore, the cylinder CAM model might constitute an interesting approach for investigations of vascularization in bone tissue engineering. The CAM model of the chick embryo is one of the most widely used in vivo assays for the evaluation of normal angiogenic processes, of malignant invasion as aspects of tumor biology or investigations

340

Tissue Eng Regen Med 2015;12(5):331-342

of ocular, mucosal or vascular toxicity of applied test substances [16,17]. Advantages include easy access to the CAM vascular network, lack of immunocompetence and low costs. We assume that the endochondral ossification approach for bone tissue engineering using chondrogenic primed BMSC in vitro might be advantageous. Chondrocytes, which are less oxygen sensitive, might withstand the initial hypoxic conditions in high density pellet culture, might serve as a scaffold for osteogenic cells and matrix properties to ensure a stable threedimensional cell embedding and support in vivo vascularization at the target site. Nevertheless, this study makes no claim to be complete. It aimed at proof of principle concerning a sequential chondroosteogenic differentiation potential of BMSC. In regard to published data investigations were designed to prove the importance of a stable 3-D cell embedding of BMSC for osteogenic differentiation. Therefore, only two different groups of chondrogenic priming were established. Although the study groups were small, significant differences between the study groups and control groups were detected. Histological results were unequivocal so that we refrained from further quantitative analysis, though results were confirmed by quantitative RT-PCR. Although we are aware of the limitations of using an immunocompromised in vivo model, we felt the CAM model might be advantageous, especially with regard to the extensive microvascular network and its easy approach for further investigations of vascularization in bioengineered tissue constructs. Our findings strongly support the theory of a chondroosteogenic differentiation potential of BMSC even after varied duration of chondrogenic priming. Chondrogenic pre-induction may lead to a stable 3-D cell embedding which seems to be crucial for the formation of a stable osteogenic phenotype in an ectopic in vivo setting without an osteogenic stimulus. This study aimed to assess the possible advantages of using an endochondral ossification model of bone formation for use in tissue engineering approaches. This approach might provide a basis for a scaffold free tissue engineering application combined with a high osteogenic potential of BMSC. Further investigations will be necessary to determine an optimal point in time of transplantation dependent on the stage of differentiation and/or matrix mineralization of sequentially chondro-osteogenic differentiated BMSC. Furthermore no biomechanical testing of the bioengineered constructs has been performed yet. Further investigations of the mechanical properties of bioengineered tissue constructs will be required. In addition further investigations will be necessary to prove the differentiation potential of BMSC into osteoblast-lineage cells after chondrogenic priming in vivo.

Acknowledgements

We thank the team of the laboratory Prof. Dr. med. B. Stark, Beate vom Hoevel, Anu Houtari, Thomas Schoenberger for excellent technical assistance. The study was funded by the Department of Education and Research Germany.

Conflicts of Interest

The authors have no financial conflicts of interest.

Ethical Statement

All procedures followed were in accordance with the ethical standards and were approved by the local ethics board (042/ 2000, 251/2002, “Tissue bank for research in the field of tissue engineering” GTE-2002). REFERENCES 1. Bauer TW, Muschler GF. Bone graft materials. An overview of the basic science. Clin Orthop Relat Res 2000;(371):10-27. 2. Conrad EU, Gretch DR, Obermeyer KR, Moogk MS, Sayers M, Wilson JJ, et al. Transmission of the hepatitis-C virus by tissue transplantation. J Bone Joint Surg Am 1995;77:214-224. 3. Li CM, Ho YR, Liu YC. Transmission of human immunodeficiency virus through bone transplantation: a case report. J Formos Med Assoc 2001;100:350-351. 4. Hallfeldt KK, Stützle H, Puhlmann M, Kessler S, Schweiberer L. Sterilization of partially demineralized bone matrix: the effects of different sterilization techniques on osteogenetic properties. J Surg Res 1995; 59:614-620. 5. Knaepler H, Haas H, Püschel HU. [Biomechanical properties of heat and irradiation treated spongiosa]. Unfallchirurgie 1991;17:194-199. 6. Banwart JC, Asher MA, Hassanein RS. Iliac crest bone graft harvest donor site morbidity. A statistical evaluation. Spine (Phila Pa 1976) 1995; 20:1055-1060. 7. Kasten P, Vogel J, Beyen I, Weiss S, Niemeyer P, Leo A, et al. Effect of platelet-rich plasma on the in vitro proliferation and osteogenic differentiation of human mesenchymal stem cells on distinct calcium phosphate scaffolds: the specific surface area makes a difference. J Biomater Appl 2008;23:169-188. 8. Kasten P, Beyen I, Niemeyer P, Luginbühl R, Bohner M, Richter W. Porosity and pore size of beta-tricalcium phosphate scaffold can influence protein production and osteogenic differentiation of human mesenchymal stem cells: an in vitro and in vivo study. Acta Biomater 2008;4:1904-1915. 9. Niemeyer P, Szalay K, Luginbühl R, Südkamp NP, Kasten P. Transplantation of human mesenchymal stem cells in a non-autogenous setting for bone regeneration in a rabbit critical-size defect model. Acta Biomater 2010;6:900-908. 10. Ashton BA, Allen TD, Howlett CR, Eaglesom CC, Hattori A, Owen M. Formation of bone and cartilage by marrow stromal cells in diffusion chambers in vivo. Clin Orthop Relat Res 1980;(151):294-307. 11. Friedenstein AJ, Petrakova KV, Kurolesova AI, Frolova GP. Heterotopic of bone marrow. Analysis of precursor cells for osteogenic and hematopoietic tissues. Transplantation 1968;6:230-247. 12. Johnstone B, Hering TM, Caplan AI, Goldberg VM, Yoo JU. In vitro chondrogenesis of bone marrow-derived mesenchymal progenitor cells. Exp Cell Res 1998;238:265-272. 13. Jaiswal N, Haynesworth SE, Caplan AI, Bruder SP. Osteogenic differentiation of purified, culture-expanded human mesenchymal stem cells in vitro. J Cell Biochem 1997;64:295-312.

14. Galotto M, Campanile G, Robino G, Cancedda FD, Bianco P, Cancedda R. Hypertrophic chondrocytes undergo further differentiation to osteoblast-like cells and participate in the initial bone formation in developing chick embryo. J Bone Miner Res 1994;9:1239-1249. 15. Steffens L, Wenger A, Stark GB, Finkenzeller G. In vivo engineering of a human vasculature for bone tissue engineering applications. J Cell Mol Med 2009;13:3380-3386. 16. Rizzo V, DeFouw DO. Macromolecular selectivity of chick chorioallantoic membrane microvessels during normal angiogenesis and endothelial differentiation. Tissue Cell 1993;25:847-856. 17. Rosenbruch M. [Early stages of the incubated chicken egg as a model in experimental biology and medicine]. ALTEX 1994;11:199-206. 18. Borges J, Tegtmeier FT, Padron NT, Mueller MC, Lang EM, Stark GB. Chorioallantoic membrane angiogenesis model for tissue engineering: a new twist on a classic model. Tissue Eng 2003;9:441-450. 19. Niemeyer P, Krause U, Fellenberg J, Kasten P, Seckinger A, Ho AD, et al. Evaluation of mineralized collagen and alpha-tricalcium phosphate as scaffolds for tissue engineering of bone using human mesenchymal stem cells. Cells Tissues Organs 2004;177:68-78. 20. Muraglia A, Cancedda R, Quarto R. Clonal mesenchymal progenitors from human bone marrow differentiate in vitro according to a hierarchical model. J Cell Sci 2000;113(Pt 7):1161-1166. 21. Kuznetsov SA, Krebsbach PH, Satomura K, Kerr J, Riminucci M, Benayahu D, et al. Single-colony derived strains of human marrow stromal fibroblasts form bone after transplantation in vivo. J Bone Miner Res 1997;12:1335-1347. 22. Fankhauser F, Schippinger G, Weber K, Heinz S, Quehenberger F, Boldin C, et al. Cadaveric-biomechanical evaluation of bone-implant construct of proximal humerus fractures (Neer type 3). J Trauma 2003;55:345-349. 23. Dominici M, Le Blanc K, Mueller I, Slaper-Cortenbach I, Marini F, Krause D, et al. Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy 2006;8:315-317. 24. Stahl A, Wenger A, Weber H, Stark GB, Augustin HG, Finkenzeller G. Bi-directional cell contact-dependent regulation of gene expression between endothelial cells and osteoblasts in a three-dimensional spheroidal coculture model. Biochem Biophys Res Commun 2004;322:684-692. 25. Satomura K, Krebsbach P, Bianco P, Gehron Robey P. Osteogenic imprinting upstream of marrow stromal cell differentiation. J Cell Biochem 2000;78:391-403. 26. Adams JC, Watt FM. Regulation of development and differentiation by the extracellular matrix. Development 1993;117:1183-1198. 27. Krieg T, LeRoy EC. Diseases of the extracellular matrix. J Mol Med (Berl) 1998;76:224-225. 28. Farrell E, van der Jagt OP, Koevoet W, Kops N, van Manen CJ, Hellingman CA, et al. Chondrogenic priming of human bone marrow stromal cells: a better route to bone repair? Tissue Eng Part C Methods 2009;15:285-295. 29. Niemeyer P, Kasten P, Simank HG, Fellenberg J, Seckinger A, Kreuz PC, et al. Transplantation of mesenchymal stromal cells on mineralized collagen leads to ectopic matrix synthesis in vivo independently from prior in vitro differentiation. Cytotherapy 2006;8:354-366. 30. Kuznetsov SA, Mankani MH, Robey PG. Effect of serum on human bone marrow stromal cells: ex vivo expansion and in vivo bone formation. Transplantation 2000;70:1780-1787. 31. Pfander D, Gelse K. Hypoxia and osteoarthritis: how chondrocytes survive hypoxic environments. Curr Opin Rheumatol 2007;19:457-462. 32. Wang W, Xu J, Kirsch T. Annexin-mediated Ca2+ influx regulates growth plate chondrocyte maturation and apoptosis. J Biol Chem 2003;278:37623769. 33. Muraglia A, Corsi A, Riminucci M, Mastrogiacomo M, Cancedda R, Bianco P, et al. Formation of a chondro-osseous rudiment in micromass cultures of human bone-marrow stromal cells. J Cell Sci 2003;116(Pt 14):2949-2955.

www.term.or.kr 341

Huebner et al. Chondro-Osteogenic Differentiation of BMSC for Bone Regeneration

34. Pelttari K, Winter A, Steck E, Goetzke K, Hennig T, Ochs BG, et al. Premature induction of hypertrophy during in vitro chondrogenesis of human mesenchymal stem cells correlates with calcification and vascular invasion after ectopic transplantation in SCID mice. Arthritis Rheum

342

Tissue Eng Regen Med 2015;12(5):331-342

2006;54:3254-3266. 35. Jukes JM, Both SK, Leusink A, Sterk LM, van Blitterswijk CA, de Boer J. Endochondral bone tissue engineering using embryonic stem cells. Proc Natl Acad Sci U S A 2008;105:6840-6845.