Effect of Cell Density on Osteoblastic Differentiation and Matrix

Dec 21, 2007 - Medical School, Hampstead Campus, London, NW3 2PF, United Kingdom, ... Division of Biomaterials and Tissue Engineering, UCL Eastman Den...
0 downloads 0 Views 1MB Size
Download PDF
Biomacromolecules 2008, 9, 129–135

129

Effect of Cell Density on Osteoblastic Differentiation and Matrix Degradation of Biomimetic Dense Collagen Scaffolds Malak Bitar,*,†,‡ Robert A. Brown,§ Vehid Salih,† Asmeret G. Kidane,4 Jonathan C. Knowles,† and Showan N. Nazhat*,†,⊥ Division of Biomaterials and Tissue Engineering, UCL Eastman Dental Institute, 256 Gray’s Inn Road, London, WC1X 8LD, United Kingdom, Materials Biology Interactions Group, Swiss Federal Laboratories for Materials Testing and Research (EMPA), Lerchenfeldstrasse 5, CH-9014 St. Gallen, Switzerland, UCL Tissue Repair & Engineering Centre, Institute of Orthopaedics, Stanmore Campus, London, HA7 4LP, United Kingdom, Academic Division of Surgery & Interventional Sciences, Royal Free & University College Medical School, Hampstead Campus, London, NW3 2PF, United Kingdom, and Department of Mining and Materials Engineering, McGill University, 3610 University Street, Montreal, QC, H3A 2B2, Canada Received October 5, 2007; Revised Manuscript Received November 6, 2007

Plastic compression of hyperhydrated collagen gels produces tissue-like scaffolds of enhanced biomechanical properties. By increasing collagen density, these scaffolds could be developed into highly Biomimetic cell-seeded templates. When utilizing three-dimensional (3-D) scaffold systems for tissue repair, and indeed when investigating the cytocompatibility of two-dimensional (2-D) surfaces, the cell seeding density is often overlooked. In this study, we investigated this potentially critical parameter using MG-63 cells seeded in the dense collagen scaffolds. This is conducted within the overall scope of developing these scaffolds for bone repair. Cell proliferation, osteoblastic differentiation, and matrix remodelling capacity in relation to various seeding densities, ranging from 105 to 108 cells/ml compressed collagen, were evaluated in vitro. This was performed using the AlamarBlue assay, quantitative polymerase chain reaction (qPCR), and tensile mechanical analysis respectively. Variations in cell seeding density significantly influenced cell proliferation where lower initial seeding density resulted in higher proliferation rates as a function of time in culture. Gene transcription levels for alkaline phosphatase (ALPL), runt-related transcription factor 2 (RUNX2), and osteonectin (SPARC) were also found to be dependent on the cell density. While ALPL transcription was down-regulated with culturing time for all seeding densities, there was an increase in RUNX2 and SPARC transcription, particularly for scaffolds with cell densities in the range 106-107 cells/ml collagen. Furthermore, this range of seeding density affected cell capacity in conducting collagenous matrix degradation as established by analyzing matrix metalloproteinase 1 (MMP1) transcription and scaffold mechanical properties. This study has shown that the seeded cell population in the three-dimensional dense collagen scaffolds clearly affected the degree of osteoblastic cell proliferation, differentiation, and some aspects of matrix remodelling activity. The seeding density played a major role in influencing the corresponding cell differentiation and cell-matrix interaction.

1. Introduction Recent advances in tissue engineering have highlighted its potential as an approach to repairing clinical defects. Such an approach typically involves the use of autologous cell transplantation in biocompatible, three-dimensional (3-D) matrices as a nucleus within which de novo tissue morphogenesis may be induced. As the main extracellular matrix (ECM) constituent of connective tissue, collagen can be used to produce matrices of ideal biomimetic properties by providing a native-like environment for the seeded cells.1–3 Recently, the “plastic compression” (PC) of highly hydrated collagen gels was developed to produce a novel 3D matrix system.4 This rapid process dramatically reduces the water content of collagen gels * Corresponding authors. E-mail: [email protected] (M.B.); [email protected] (S.N.N.). † Division of Biomaterials and Tissue Engineering, UCL Eastman Dental Institute. ‡ Materials Biology Interactions Group, Swiss Federal Laboratories for Materials Testing and Research (EMPA). § UCL Tissue Repair & Engineering Centre, Institute of Orthopaedics. 4 Academic Division of Surgery & Interventional Sciences, Royal Free & University College Medical School. ⊥ Department of Mining and Materials Engineering, McGill University.

using a combination of capillary action and unconfined compressive load, thus forming tissue-like structures of significantly enhanced mechanical properties.5 Furthermore, the effect on the seeded cell viability and function as a result of the compression process has been shown to be considerably limited. Using this novel and controllable process, rapid fabrication of transplantable matrices for connective tissue repair may soon become a reality.6,7 Previous results using this system have shown, and following plastic compression, an increase in collagen concentration in these matrices up to ∼13 wt %.5 This configuration supported osteoblastic cell survival and a degree of cell-induced matrix remodelling after 10 days in culture.8 Inevitably, cell transplantation in matrices involves an in vitro stage of variable time scale in order to permit cell colonization and initial interaction with the scaffold.9–11 It is recognized, however, that the isolation and in vitro expansion may result in the cell dedifferentiation.12 It is important, therefore, that the in vitro cell-scaffold culture environment is carefully controlled toward the morphogenesis of the appropriate tissue. This is also true when utilizing stromal stem cells as precursors for tissuespecific regeneration.13

10.1021/bm701112w CCC: $40.75  2008 American Chemical Society Published on Web 12/21/2007

130 Biomacromolecules, Vol. 9, No. 1, 2008

Many factors have been shown to influence cell differentiation and subsequent tissue formation such as cell seeding related factors, the use of bioreactor culture environments, and the introduction of several soluble agonists to directly promote functional differentiation.14–21 Among these factors, the cell seeding density may play a major role in influencing the fate of the seeded population.22,23 This factor remains largely uninvestigated, particularly when developing 3-D matrices for bone repair. In this regard, the few studies conducted have demonstrated that cell seeding density influences osteoblastic proliferation, differentiation, and the subsequent ECM interaction.24,25 These studies revealed the seeding density to modulate efficient scaffold mineralization, thereby underlying its importance in bone repair. This study, therefore, investigated whether the seeding density influences cell proliferation, osteoblastic differentiation, and contribution to the mechanical properties of the scaffolds with the aim of exploiting a particular density as a driving force for osteoblastic cell transplantation into dense collagen matrices. It is envisaged that optimizing the cell density may represent a vital step in the development of these systems toward a viable option as a tissue engineering system.

2. Experimental Section 2.1. Cell Culture. MG-63 cells (human osteosarcoma cell line) were chosen for their osteoblastic features. This well-characterized cell line has been shown to produce mineralized tissue in vivo and is frequently used to conduct bioperformance studies in relation to bone tissue regeneration.26,27 The investigated cell densities were 6 × 103, 3 × 104, 3 × 105, and 1 × 106 cells/ml (all precompression values). The precompression cell seeding density of 3 × 105 cells/ml was previously used to assess the impact of the extent of compression on cell viability and function and was shown, additionally, to retain up to 80% cell viability in compressed collagen scaffolds.8 Cell culture was maintained in a humidified atmosphere at 37 °C and 5% CO2. Cells were cultivated in Dulbecco’s modified Eagle medium (D-MEM) containing 10% (v/ v) fetal calf serum, 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin (all from Invitrogen Ltd., Paisley, UK). When conducting qPCR experiments (see below), RNA yields were undetectable at 24 h in culture for the lowest seeding density. Accordingly, the first time point for investigating gene expression and cell proliferation was set at day 2 in culture. 2.2. Preparation of Cellular Scaffolds. Collagen gels were produced by adding 0.2 ml of X10 D-MEM to 1.6 ml of sterile rat-tail type I collagen dissolved in acetic acid (2.2 mg/ml, First Link Ltd., West Midlands, UK). The collagen solution was instantly neutralized by sequentially introducing 3 µL aliquots of 5 M NaOH. Immediately thereafter, 0.2 ml of D-MEM (at 37 °C) containing the cell population at each investigated density was added and the solution transferred into rectangular molds (13 mm × 33 mm × 4 mm) and left to set at 37 °C for 30 min. Dense collagen scaffolds were produced as previously described by the standardized process of single compression.4–8 Briefly, the resulting hydrated collagen gels were subjected to unconfined compressive load of 1.4 kN · m-2 for 5 min to produce sheets. For the purpose of mechanical analysis, the sheets were rolled along the short axis to give concentrically multilayered cylinder scaffolds of ∼2 mm diameter. 2.3. Evaluation of Postcompression Scaffold Volume. Immediately following compression, acellular dense collagen sheets were embedded in optimal cutting temperature (OCT) compound and snap-frozen in liquid nitrogen at approximately -196 °C. Frozen sections (12 µm) were obtained using a Cryostat (Leitz 1720, Ernst Leitz, Wetzlar GmbH, Germany), mounted on glass slides, and left to dry overnight. The sections were subsequently fixed in ice-cold acetone for 15 min, embedded in an aqueous mountant (Citifluor Ltd., London, UK), and viewed under the light microscope using 10× objective lens. For each section, several areas were selected at random and digital images were

Bitar et al. obtained to assess the collagen sheet depth. This was conducted using Image-Pro Plus software 4.0 (Media Cybernetics UK, Berkshire, UK) whereby all sample sections present in the microscopic field were measured for depth at every 100 µm. 2.4. Cell Proliferation Assay. To assess the impact of cell density on proliferation, AlamarBlue reduction within the seeded population was evaluated. The reduction in AlamarBlue, as a result of cell proliferation, is directly related to the reduced internal environment of the cell. Such reduction results in changing the color of the reagent from the oxidized indigo blue, nonfluorescing state to the reduced fluorescent pink state. The change in color can then be fluorometrically detected. At days 2, 6, 8, and 10 in culture, cell-seeded scaffolds were incubated in 10% concentration of AlamarBlue (Invitrogen Ltd., Paisley, UK) for 4 h at +37 °C. Subsequently and for each sample, 100 µL of the culture medium supernatant was placed in opaque 96-well plates and fluorescently assessed using a Fluroskan Ascent plate reader (Ex 530, Em 590 nm; Labsystems, Helsinki, Finland). For the positive control experiments, cells were seeded at the same densities in noncompressed, hydrated collagen gels. Acellular scaffolds and hydrated gels were used as negative controls. 2.5. Confocal Laser Microscopy. To monitor the changes in cell viability and survival in relation to the seeding density at day 10 in culture, samples were incubated for 45 min in D-MEM supplemented with 1:1000 concentration of Calcein AM and ethidium homodimer-1 (Invitrogen Ltd., Paisley, UK) to stain live and dead cells, respectively. Samples were fixed in 99% ice-cold methanol for 10 min, washed 3 times in phosphate buffered saline (PBS), and mounted between two glass cover slips in citiflour AF2 mounting medium (Citifluor Ltd., London, UK) for examination under the BioRad Radiance 2100 confocal microscope (4-line argon primary laser/green HeNe laser; Carl Zeiss Ltd., Hertfordshire, UK). 2.6. Quantitative Polymerase Chain Reaction (qPCR). The effect of seeding density on osteoblastic differentiation in terms of the regulation patterns for a number of genes was assessed at days 2 and 10 in culture. Scaffolds were placed in RLT cell lysis buffer (including 1% β-mercaptoethanol, Qiagen Ltd., West Sussex, UK) and homogenized immediately, twice for 20 s using FastPrep FP120 cell disrupter (Qbiogene, Inc., Cedex, France). Total RNA was later extracted from the lysate using RNeasy Mini Kit (Qiagen Ltd.). For every sample, RNA concentration was determined using Quanti IT RNA Assay Kit with the Qubit flourimeter (Invitrogen). Subsequently, 200 ng of total RNA for each sample was reverse transcribed into cDNA (high capacity cDNA archive kit, Applied Biosystems, Cheshire, UK). Conjugated to FAM (6-carboxyfluorescein) reporter dye, TaqMan Probes (Applied Biosystemys) were used to target corresponding nucleotide sequences in the cDNA single strands. The presence of osteoblast function-specific nucleotide sequences, and their transcription rate, was assessed by quantifying the amplified FAM signal after 40 thermal cycles. This process took place at 25 µL reaction volume in 96 reaction plates using the ABI PRISM 7300 sequence detection system (all from applied Biosystems). Relative quantification (RQ) of gene expression took place using the 7300 SDS software whereby cDNA amplification curves for all target genes were normalized against a calibrator sample with GAPDH (glyceraldehyde-3-phosphate dehydrogenase, UniGene ID: Hs.479728) as an endogenous housekeeping gene. This comparative CT method was automatically performed using the 7300 SDS software. The cDNA was targeted for sequences of runt-related transcription factor 2 (RUNX2, UniGene ID: Hs.535845), osteonectin encoding SPARC (UniGene ID: Hs.111779), alkaline phosphatase (ALPL, UniGene ID: Hs.75431), and Matrix metallopeptidase 1 (interstitial collagenase, MMP1, UniGene ID: Hs.83169). The data is presented as relative quantification (RQ) of fold increases in cDNA PCR product versus cell density. 2.7. Mechanical Analysis. Tensile testing was carried out to investigate the impact of seeding density on the mechanical properties of the scaffolds immediately following compression and at day 10 in culture. These were carried out on rolled samples, as it allowed for

Biomimetic Dense Collagen Scaffolds

Biomacromolecules, Vol. 9, No. 1, 2008 131 Table 1. Pre- and Postcompression Cell Densities and Their Corresponding Volume precompression cell density (cells/ml) 6 × 103 3 × 104 3 × 105 1 × 106

Figure 1. Continuous descriptive analysis of compressed scaffold depth measurements.

easier handling and mechanical testing. The tests were carried out using a commercial dynamic mechanical analyzer system (DMA-7e, PerkinElmer, Bucks, UK) connected to a PC, and captured data was analyzed through DMA-7e Pyris version 5 software (Perkin-Elmer). This setup allowed uniaxial tensile loading of a sample at a chosen loading rate. Once fabricated, the flat specimen sheets were rolled along their short axis to generate rolls of 2 ( 0.25 mm diameter and 13 mm length. During testing, self-tightening grips were used aided by 2 mm wide stainless steel meshes folded around the specimen ends to allow adequate clamping of the samples. Prior to starting each test, the system was equilibrated at 23 °C. The sample gauge length was then measured and the test started. A ramp load was applied to the specimen, starting from 1 mN and ending at sample failure at a loading rate of 200 mN min-1. The test duration was up to 10 min. The sample was prevented from dehydration by periodic direct application of isotonic PBS during the test period. Because all specimens were prepared and tested identically except for inclusion of the different cell densities and time in culture, an output of force versus strain was chosen to compare the relative changes in break force as an indication of break stress. Acellular scaffolds were used as negative controls. 2.8. Statistical Analysis. Statistical differences were evaluated through applying a one-way ANOVA test using SPSS for windows (release 11, SPSS UK Ltd., Surrey, UK) followed by Fisher-LSD post hoc analysis at 95% confidence interval. Differences in values were considered significant when p < 0.05. SPSS was also used to conduct continuous summary descriptive analysis with 95% confidence interval, 95% percentile plot, and Shapiro-Wilk normality test.

3. Results 3.1. Determination of Cell Density in Compressed Collagenous Scaffolds. A continuous descriptive analysis on the postcompression scaffold depth revealed an average scaffold depth of 50.7 ( 15.6 µm (Figure 1). By using the dimensions of the mold, the postcompression volume of the scaffold was calculated to be ∼10 µL (13 × 103 by 16.5 × 103 by 50.7 (all in µm)). By using this value, the pre- and postcompression cell densities and their corresponding volumes were calculated and are given in Table 1. These values were applied in this work and used in the hydrated gel controls for the cell proliferation

total volume of cell populations (µL)

postcompression cell density (cells/ml)

0.01 0.06 0.59 1.92

6 × 105 3 × 106 3 × 107 1 × 108

study. The effect of cell density on the cell population volume was considered by using the MG-63 cell internal volume of ∼1.92 × 10-6µL.28 Therefore, for example, the highest precompression seeding density of 1 × 106 cells/ml that was used in this study resulted in a total cell population volume of 1.92 µL, which was considered to be of limited potential impact on the final scaffold structure and volume. 3.2. The Effect of Seeding Density on Cell Proliferation. Figure 2 shows a plot of cell proliferation as a function of time in culture for the various dense collagen scaffolds with different seeding densities. As can be seen, scaffolds seeded at 6 × 105 and 3 × 106 cells/ml exhibited an exponentially increasing proliferation pattern as a function of time in culture (parts A and B of Figure 2, respectively). This significant increase in proliferation was demonstrated across almost all the time points tested. Seeding density directly influenced cell proliferation. This was in contrast to cells seeded at the two highest seeding densities of 3 × 107 and 1 × 108 cells/ml, which underwent a limited but significant decrease in proliferation by day 10 in culture and with one exception (1 × 108 cells/ml at day 8). The proliferation trend, however, showed no significant change as a function of time in culture (parts C and D of Figure 2, respectively). The proliferation behavior of cells seeded at similar seeding densities into precompression hyperhydrated collagen gels were used as controls. This was carried out as the proliferation data shown in Figure 2 may have been influenced, following compression, by a decrease in cell numbers due to potential cell death within the dense scaffolds. The data shown in Figure 3 confirmed the above results. For the lowest cell densities, 6 × 105 and 3 × 106 cells/ml, a significant increase in proliferation was exhibited at days 4 and 8 (parts A and B of Figure 3, respectively). However, by day 10, no significant increase in proliferation rate took place. For the highest cell densities of 3 × 107 and 1 × 108 cells/ml, cells also underwent a significant increase at days 4 and 6 in culture (parts C and D of Figure 3, respectively). Nevertheless, for both seeding densities, there was a significant decrease in proliferation rate by day 10. It is noticeable that for both compressed scaffolds and the hydrated controls, proliferation rate was inhibited and was further reduced when the seeding density was highest. Therefore, density-related cell proliferation, as established by using the AlamarBlue assay, exhibited similar trends when comparing compressed and hydrated collagen scaffolds. 3.3. The Effect of Seeding Density on Cell Viability. As a direct confirmation of the AlamarBlue assay data, the cells seeded at 6 × 105 and 3 × 106 cells/ml underwent a considerable increase in proliferation as a function of time in culture resulting in dense cell population throughout the compressed collagenous scaffold (Figure 4A,B). However, due to the relatively high initial seeding densities, no visible increase in cell numbers were established for seeding densities 3 × 107 and 1 × 108 cells/ml by day 10 (Figure 4C,D). It was also clear that for all seeding densities, viable cells (stained green) were predominant within the scaffolds at day 10 in culture.

132 Biomacromolecules, Vol. 9, No. 1, 2008

Bitar et al.

Figure 2. Cell proliferation in compressed collagen scaffolds. A ) 6 × 105, B ) 3 × 106, C ) 3 × 107, and D ) 1 × 108 cells/ml. X axis: time in culture (days). Y axis: fluorescence intensity (Ex 530, Em 590 nm, au). Significant differences (p < 0.05) in proliferation rate are indicated by (*) (n ) 3, error bars + SD).

Figure 3. Cell proliferation in hydrated collagen gels. A ) 6 × 105, B ) 3 × 106, C ) 3 × 107, and D ) 1 × 108 cells/ml. X axis: time in culture (days). Y axis: fluorescence intensity (Ex 530, Em 590 nm, au). Significant differences (p < 0.05) in proliferation rate are indicated by (*) (n ) 3, error bars + SD).

3.4. The Effect of Seeding Density on Osteoblastic Cell Differentiation. As seen from Figure 5A, ALPL transcription at day 2 was expressed at higher levels in cells seeded at 6 × 105 and 3 × 106 cells/ml. For the highest seeding densities of 3 × 107 and 1 × 108 cells/ml, ALPL transcription took place at significantly lower rates. For all seeding densities, by day 10 in culture, ALPL transcription was down-regulated as a function of time. At this time point, cells seeded at 3 × 106 expressed the highest levels of ALPL transcription (p < 0.05). By day 2, RUNX2 transcription was expressed at a significantly higher rate in the scaffolds seeded at the highest cell density, but were down-regulated by day 10 (Figure 5B). For all other densities, RUNX2 transcription was significantly upregulated as a function of time. At day 10, the seeding density of 3 × 107 cells/ml was associated with the highest transcription levels, which was followed by 6 × 105 cells/ml, then by 3 × 106 cells/ml. SPARC transcription took place at equal rates at day 2 in all seeding densities except for 3 × 106 cells/ml, where it displayed

a significantly higher expression rate (Figure 5C). Between days 2 and 10 in culture, SPARC transcription was up-regulated in scaffolds seeded at 3 × 107 cells/ml at a higher level compared to those associated with other densities. This was followed by transcription levels expressed by cells seeded at 3 × 106 cells/ ml. Cells seeded at 6 × 105 and 1 × 108 cells/ml, on the other hand, expressed no changes in SPARC regulation as a function of time and yielded the lowest transcription levels at day 10. 3.5. The Effect of Seeding Density on ECM Degradation and Mechanical Properties. In addition to influencing proliferation and osteoblastic differentiation, cell seeding density is likely to affect the mechanical properties of the scaffold. Immediately, following compression, tensile mechanical analysis showed that only scaffolds seeded at 1 × 108 cells/ml experienced a significant reduction in break force compared to those with other seeding densities and the acellular control (Figure 6A). By day 10 in culture, scaffolds seeded at 3 × 106 and 3 × 107 cells/ml underwent a significant decrease in break force, whereas scaffolds seeded at 6 × 105 cells/ml maintained

Biomimetic Dense Collagen Scaffolds

Biomacromolecules, Vol. 9, No. 1, 2008 133

Figure 4. Confocal laser microscope images of MG-63 cells seeded in compressed dense collagen scaffolds at day 10 in culture. Cell densities were 6 × 105 (A), 3 × 106 (B), 3 × 107 (C), and 1 × 108 (D) cells/ml. Scale bars ) 50 µm.

their break force value. A similar trend was obtained for the acellular controls. As changes in the scaffold mechanical properties could be related to cell-induced break down of the collagenous matrix by collagenase activity, MMP1 gene regulation was assessed as a function of time in culture for the various seeding densities (Figure 6B). Initially, at day 2, higher levels of MMP1 transcription were exhibited by cells seeded at the two lowest densities. By day 10, however, only cells seeded at 6 × 105 cells/ml down-regulated MMP1 transcription. No changes in MMP1 gene regulation were associated with the 3 × 106 cells/ ml density over this time period and, in contrast, there was a significant upregulation in MMP1 transcription by day 10 in cells seeded at 3 × 107 and 1 × 108 cells/ml.

4. Discussion In the development of biomimetic scaffolds for tissue engineering, cell-cell interactions may greatly influence the seeded population behavior and the eventual structure and properties of the engineered tissue.29 The extent of cell-cell interactions may be controlled by the seeding density and eventually induce a specific cell differentiation needed for each tissue type. Therefore, in this study, cell proliferation was used as an indicator of the extent of cell-cell contacts within the three-dimensional scaffold. Cell proliferation is suppressed by the contact-inhibition mechanism modulated primarily by gap junctional intercellular communication (GJIC) between adjacent

cells.28,30 Intercellular communication, on the other hand, particularly through connexin hemichannels, modulates osteoblastic differentiation. For example, it has been shown that the disruption to connexin 43 function significantly reduced osteocalcin synthesis.31,32 Contact inhibition was apparent in this study, as cell proliferation patterns were proportional to their seeding densities. Seeding the compressed collagen scaffolds at higher densities led to the inhibition of cell proliferation and, in some cases, their reduction as a function of time in culture. Only cells seeded at the lowest density of 6 × 105 cells/ml exhibited constant increase in cell division up to day 10. A similar trend was obtained for the hydrated collagen control gels with the same seeding densities, indicating that the compression process resulted in no negative impact of cell proliferation. It is worth considering, nonetheless, that in the case of the 6 × 105 cells/ml density, cell proliferation in the control scaffolds was inhibited by day 10, suggesting that a sufficient cell density was reached to induce contact inhibition in this population. This also implies that, upon compression, the increased collagen density may influence cell proliferation. The synthesis of tissue-nonspecific alkaline phosphatase (TNAP), encoded by the ALPL gene (a tissue-nonspecific isozyme precursor), is vital for bone matrix mineralization. TNAP plays a major role in the removal of mineralization inhibitors within the collagenous ECM. This is achieved through the hydrolysis of extracellular inorganic pyrophosphate (PPi) which inhibits the nucleation and growth of hydroxyapatite (HA)

134 Biomacromolecules, Vol. 9, No. 1, 2008

Figure 5. Analysis of qPCR. (A) ALPL transcripts, (B) RUNX2 transcripts, and (C) SPARC transcripts. Day 2, black bars, and day 10, gray bars. X axis: cell density (cells/ml). Y axis: relative quantification (RQ) of fold increase differences in cDNA PCR product. Significant differences in gene expression among samples at one time point are indicated by (+). Significant differences in gene regulation for each sample as a function of time in culture are indicated by (*) (n ) 3, error bars + SD).

crystals.23,33 By day 2, ALPL transcription appeared to be regulated at higher rates in cells seeded at lower densities. More significantly and for all seeding densities, ALPL transcription was down-regulated as a function of time in culture. Such a phenomenon is not experienced by cells grown in monolayers and may be linked, as previously reported, to the presence of a three-dimensional collagen lattice.34 As an early marker of osteoblastic differentiation, ALPL transcription may be elevated at the early stages of premineralized collagenous bone ECM remodelling and gradually reduced, at later stages, as the ECM becomes capable of supporting HA crystal growth and subsequent mineralization.23,33 The results obtained for ALPL transcription indicate that the sequence of events seen in vivo, related to early stages of matrix mineralization, may be associated with these dense collagenous matrices. This, in turn, further confirms their biomimetic properties by providing a natural environment for cell osteoblastic differentiation. Regulation of RUNX2 (aliases AmL3/CBFA1/PEBP2aA, an osteocalcin transcription factor encoding gene) is also indicative of osteoblastic differentiation because osteocalcin is synthesized when the remodelling of the bone matrix is activated.35,36 With the exception of the seeding density of 1 × 108 cells/ml, RUNX2 transcription was up-regulated with time in culture. SPARC transcription, on the other hand, regulates the synthesis of the noncollagenous glycoprotein osteonectin (ON). ON may be involved in active engagement of HA nucleation during bone matrix mineralization.18,37 In this study, SPARC transcription was up-regulated by day 10 in cells seeded at 3 × 107 cells/ml.

Bitar et al.

Figure 6. (A) Tensile mechanical analysis data showing the effect of seeding density on scaffold break force. Following compression, black bars, and day 10, gray bars. X axis: cell density (cells/ml). Y axis: break force (mN). (B) Analysis of qPCR. MMP1 transcripts. Day 2, black bars, and day 10, gray bars. X axis: cell density (cells/ml). Y axis: relative quantification (RQ) of fold increase differences in cDNA PCR product. Significant differences among samples at one time point are indicated by (+). Significant differences for each sample as a function of time in culture are indicated by (*) (p < 0.05, n ) 3, error bars + SD).

Interestingly, this seeding density was also associated with the highest level of gene transcription at day 10 for both SPARC and RUNX2. Generally, lower osteoblastic gene expression and regulation rates were associated with cells in scaffolds seeded at the highest density (1 × 108 cells/ml). This may be attributed to the relatively high cell number and should be further investigated in future studies. Matrix metalloproteinases are a family of matrix-degrading enzymes of which MMP1 (an interstitial collagenase) degrades collagen type I molecular helix and, subsequently, the fibrillar structure of collagen network.38 This is a key step in the processes of collagen removal during tissue maintenance and repair.39,40 These dense collagen scaffolds should be expected to undergo cell-mediated remodelling balancing collagen degradation and synthesis. In this study, the seeded cells appeared to be actively engaged in the degradation of these scaffold as levels of MMP1 transcription were maintained and up-regulated over time. This was mirrored by changes in mechanical strength, as indicated by the break force value experienced by the scaffolds over time. Seeding density appeared to play a significant role in this factor as cells seeded at the lowest density down-regulated MMP1 transcription with time in culture and corresponded to no changes in the scaffold break force. It was apparent that, at this seeding density, the cells were committed to proliferation with limited matrix degradation activity as also confirmed by the cell proliferation data. On the other hand, an increase in cell density was strongly associated with the concept of density-regulated cell-matrix interaction. Cells seeded at 3 × 106 and 3 × 107 cells/ml appeared to be actively engaged in collagenous matrix degradation, in a time dependent manner, as indicated by the reduction in scaffold strength. It is also worth

Biomimetic Dense Collagen Scaffolds

noting from Figure 6A that the relatively high cell volume associated with the 1 × 108 cells/ml may have also contributed to changes to the scaffold’s mechanical properties, as demonstrated by a significant reduction in the break force immediately following the compression process, which did not change after 10 days in culture.

5. Conclusion A range of “environment” related features must be optimized to ensure potential bone formation within the engineered scaffold. First and foremost, the composition of the threedimensional scaffold is a critically important factor. In this regard, as revealed in this study, plastically compressed dense collagen matrices supported osteoblastic differentiation in a manner that can be related to a native environment. This study has also confirmed, within a connective tissue-like environment, the biological statement that cell density-controlled cell-cell interaction greatly modulates cell function. Controlling the cell seeding density, therefore, within compressed collagen scaffolds, may enhance osteoblastic cell differentiation, thus potentially leading to the functional mineralization and host integration. In this study, it has been demonstrated that densities in the range of 3 × 106 to 3 × 107 cells/ml appeared to be associated with enhanced osteoblastic differentiation. Cells seeded at this range also demonstrated greater cell-mediated matrix degradation activity as confirmed by the mechanical behavior. Therefore, this range should be further refined to investigate the formation and regulation of GJIC and related proteins affecting osteoblastic differentiation with the aim of further developing biomimetic dense collagen matrices for bone tissue engineering. Acknowledgment. This work was funded by The Royal Free and University College London Hospital Clinical Research and Development Committee, the BBSRC, and EPSRC (the TIBS Consortium).

References and Notes (1) Rosso, F.; Giordano, A.; Barbarisi, M.; Barbarisi, A. J. Cell. Physiol. 2004, 199, 174. (2) Rosso, F.; Marino, G.; Giordano, A.; Barbarisi, M.; Parmeggiani, D.; Barbarisi, A. J. Cell. Physiol. 2005, 203, 465. (3) Cancedda, R.; Dozin, B.; Giannoni, P.; Quarto, R. Matrix Biol. 2003, 22, 81. (4) Brown, R. A.; Wiseman, M.; Chuo, C. B.; Cheema, U.; Nazhat, S. N. AdV. Funct. Mater. 2005, 15, 1762. (5) Abou Neel, E. A.; Cheema, U.; Knowles, J. C.; Brown, R. A.; Nazhat, S. N. Soft Matter 2006, 2, 986. (6) Nazhat, S. N.; Abou Neel, E. A.; Kidane, A.; Ahmed, I.; Hope, C.; Kershaw, M.; Lee, P. D.; Stride, E.; Saffari, N.; Knowles, J. C.; Brown, R. A. Biomacromolecules 2007, 8, 543. (7) Mudera, V.; Morgan, M.; Cheema, U.; Nazhat, S. N.; Brown, R. A. J. Tissue Eng. RegeneratiVe Med. 2007, 1, 192. (8) Bitar, M.; Salih, V.; Brown, R. A.; Nazhat, S. N. J. Mater. Sci.: Mater. Med. 2007, 18, 237.

Biomacromolecules, Vol. 9, No. 1, 2008 135 (9) Vacanti, J. P.; Langer, R.; Upton, J.; Marler, J. J. AdV. Drug DeliVery ReV. 1998, 33, 165. (10) Oakes, B. W. Med. J. Aust. 2004, 180, S35. (11) Braddock, M.; Houston, P.; Campbell, C.; Ashcroft, P. News Physiol. Sci. 2001, 16, 208. (12) Cervenkova, K.; Belejova, M.; Vesely, J.; Chmela, Z.; Rypka, M.; Ulrichova, J. Biomed. Pap. Med. Fac. UniV. Palacky Olomouc Czech Repub. 2001, 145, 57. (13) Ciapetti, G.; Ambrosio, L.; Marletta, G.; Baldini, N.; Giunti, A. Biomaterials 2006, 27, 6150. (14) Vogelin, E.; Jones, N. F.; Huang, J. I.; Brekke, J. H.; Toth, J. M. Tissue Eng. 2000, 6, 449. (15) Salgado, A. J.; Coutinho, O. P.; Reis, R. L. Macromol Biosci. 2004, 4, 743. (16) Goh, J. C.; Ouyang, H. W.; Teoh, S. H.; Chan, C. K.; Lee, E. H. Tissue Eng. 2003, 9, S31. (17) Frangos, J. A.; McIntire, L. V.; Eskin, S. G. Biotechnol. Bioeng. 1987, 32, 1053. (18) Burg, K. J.; Porter, S.; Kellam, J. F. Biomaterials 2000, 21, 2347. (19) Brown, T. D. J. Biomech. 2000, 33, 3. (20) Botchwey, E. A.; Pollack, S. R.; Levine, E. M.; Laurencin, C. T. J. Biomed. Mater. Res. 2001, 55, 242. (21) Boden, S. D. Clin. Orthop. Relat. Res. 1999, 367, S84. (22) Zhou, Y. F.; Sae-Lim, V.; Chou, A. M.; Hutmacher, D. W.; Lim, T. M. J. Biomed. Mater. Res., Part A 2006, 78, 183. (23) Ciancaglini, P.; Simao, A. M.; Camolezi, F. L.; Millan, J. L.; Pizauro, J. M. Braz. J. Med. Biol. Res. 2006, 39, 603. (24) Wilson, C. E.; Dhert, J. A.; van Blitterswijk, C. A.; Verbout, A. J.; de Bruijn, J. D. J. Mater. Sci.: Mater. Med. 2002, 13, 1265. (25) Holy, C. E.; Shoichet, M. S.; Davies, J. E. J. Biomed. Mater. Res. 2000, 51, 376. (26) Takagishi, Y.; Kawakami, T.; Hara, Y.; Shinkai, M.; Takezawa, T.; Nagamune, T. Tissue Eng. 2006, 12, 927. (27) Bitar, M.; Salih, V.; Mudera, V.; Knowles, J. C.; Lewis, M. P. Biomaterials 2004, 25, 2283. (28) Banoub, R. W.; Fernstrom, M.; Ruch, R. J. Cancer Lett. 1996, 108, 35. (29) Cheng, G.; Youssef, B. B.; Markenscoff, P.; Zygourakis, K. Biophys. J. 2006, 90, 713. (30) Ruch, R. J.; Guan, X.; Sigler, K. Mol. Carcinog. 1995, 14, 269. (31) Stains, J. P.; Lecanda, F.; Screen, J.; Towler, D. A.; Civitelli, R. J. Biol. Chem. 2003, 278, 24377. (32) Stains, J. P.; Civitelli, R. Birth Defects Res., Part C: Embryo Today ReV. 2005, 75, 72. (33) Balcerzak, M.; Hamade, E.; Zhang, L.; Pikula, S.; Azzar, G.; Radisson, J.; Bandorowicz-Pikula, J.; Buchet, R. Acta Biochimica Pol. 2003, 50, 1019. (34) Rattner, A.; Sabido, O.; Le, J.; Vico, L.; Massoubre, C.; Frey, J. Calcif. Tissue Int. 2000, 66, 35. (35) Karsenty, G. Semin. Cell DeV. Biol. 2000, 11, 343. (36) Hauschka, P. V.; Lian, J. B.; Cole, D. E.; Gundberg, C. M. Physiol. ReV. 1989, 69, 990. (37) Sage, H.; Vernon, R. B.; Funk, S. E.; Everitt, E. A.; Angello, J. J. Cell Biol. 1989, 109, 341. (38) Shapiro, S. D.; Senior, R. M. Am. J. Respir. Cell Mol. Biol. 1999, 20, 1100. (39) Chung, L.; Dinakarpandian, D.; Yoshida, N.; Lauer-Fields, J. L.; Fields, G. B.; Visse, R. EMBO J. 2004, 23, 3020. (40) Abraham, L. C.; Dice, J. F.; Lee, K.; Kaplan, D. L. Exp. Cell Res. 2007, 313, 1045.

BM701112W