Elsevier

Biomaterials

Volume 22, Issue 11, 1 June 2001, Pages 1279-1288
Biomaterials

Effect of convection on osteoblastic cell growth and function in biodegradable polymer foam scaffolds

https://doi.org/10.1016/S0142-9612(00)00280-5Get rights and content

Abstract

Culture of seeded osteoblastic cells in three-dimensional osteoconductive scaffolds in vitro is a promising approach to produce an osteoinductive material for repair of bone defects. However, culture of cells in scaffolds sufficiently large to bridge critical-sized defects is a challenge for tissue engineers. Diffusion may not be sufficient to supply nutrients into large scaffolds and consequently cells may grow preferentially at the periphery under static culture conditions. Three alternative culturing schemes that convect media were considered: a spinner flask, a rotary vessel, and a perfusion flow system. Poly(dl-lactic-co-glycolic acid) (PLGA) foam discs (12.7 mm diameter, 6.0 mm thick, 78.8% porous) were seeded with osteoblastic marrow stromal cells and cultured in the presence of dexamethasone and l-ascorbic acid for 7 and 14 days. Cell numbers per foam were found to be similar with all culturing schemes indicating that cell growth could not be enhanced by convection, but histological analysis indicated that the rotary vessel and flow system produced a more uniform distribution of cells throughout the foams. Alkaline phosphatase (ALP) activity per cell was higher with culture in the flow system and spinner flask after 7 days, while no differences in osteocalcin (OC) activity per cell were observed among culturing methods after 14 days in culture. Based on the higher ALP activity and better cell uniformity throughout the cultured foams, the flow system appears to be the superior culturing method, although equally important is the fact that in none of the tests did any of the alternative culturing techniques underperform the static controls. Thus, this study demonstrates that culturing techniques that utilize fluid flow, and in particular the flow perfusion system, improve the properties of the seeded cells over those maintained in static culture.

Introduction

For the repair of bone defects, the ideal biomaterial is one that has mechanical properties similar to bone, can be fabricated easily into a desired shape, supports cell attachment, contains factors to induce formation of new bone tissue, and biodegrades to permit natural bone formation and remodeling. To date many synthetic polymers have been examined that can meet many of these ideals (for a review see Behravesh et al. [1]), but a continuing challenge is incorporation of osteoinductive factors. One approach has been to incorporate growth factors into the material, and extensive research has focused on the use of bone morphogenic protein-2 [18], [19], [24]. (See Schmitt et al. [30] for a general review of bone morphogenic proteins.) An alternative approach is to culture osteoprogenitor cells in vitro within the biomaterial scaffold. In theory, these cells would proliferate and secrete an osteoinductive matrix onto the surfaces of the scaffold that would facilitate healing even if the cells did not survive implantation of the biomaterial.

Because bone defects that require surgical correction are large (typically many millimeters in size), the biomaterial scaffolds intended to bridge such defects would be sufficiently large to complicate in vitro culture. Diffusional rates of nutrients into the scaffold and metabolites out may not satisfy the metabolic requirements of seeded cells and result in suppression of cell growth and expression of phenotypic markers. Furthermore, gradients in nutrient and metabolite concentration may induce chemotaxis of the cells from the interior. These phenomena would produce an uneven distribution of cells and osteoinductive matrix favoring the exterior surfaces. Clearly, as the size of the scaffold is increased the problem would become more acute. A solution to this problem is to culture seeded scaffolds in a system that can enhance nutrient delivery.

A variety of different culturing systems have been documented, including perfusion [8], spinner flask [33], and rotary vessel [7], [26]. The spinner flask and rotary vessel use convection to ensure that the media surrounding the scaffolds is well-mixed. In addition, fluid flow across the rough external surfaces should form eddies in the superficial pores that will enhance nutrient transport into the pores. One key difference between the two culturing systems is that in the turning rotary vessel the scaffolds are allowed to tumble (i.e., their orientations to change continuously with respect to the direction of flow), whereas in the spinner flask they are fixed in place within the stirring media. The potential limitation with both of these systems is that mixing at the surface of the scaffolds may not be sufficient to deliver the necessary nutrients to the interior of the foam. A perfusion system, because it convects media through the interconnected pores of the scaffolds, should provide the best transport of nutrients. The limitation with the perfusion system is its relative complexity and difficulty in assembling and operating. Comparison of it with the other systems is necessary to justify its utility.

In this study, the growth and osteoblastic function of seeded pre-osteoblastic cells were examined for four different culturing methods: a flow system, a spinner flask, a rotary vessel, and static culture. Large porous foam scaffolds (12.7 mm diameter, 6.0 mm thick, 78.8% porous) of poly(dl-lactic-co-glycolic acid) (PLGA) were prepared by solvent casting followed by compression molding and particulate leaching and seeded with cells expanded from rat marrow stroma [9]. At the time of seeding, osteogenic factors (dexamethasone and l-ascorbic acid) were added to force primitive cells to differentiate into osteoblasts. This differentiation process is marked by sequential stages of proliferation, and expression of alkaline phosphatase (ALP) and osteocalcin (OC) [20]. To evaluate cell growth within the foams, total DNA and 3H-thymidine incorporation were measured. In addition, histological sections of cultured foams were examined at 7 days in culture to identify differences in cell distribution with culturing method. Finally, to evaluate osteoblastic function, ALP activity and OC expression were examined at 7 and 14 days in culture.

Section snippets

Materials

Foams were prepared using 75:25 poly(dl-lactic-co-glycolic acid) (PLGA) Birmingham Polymers (lot D98036, weight average molecular weight of 81 kDa, polydispersity index of 3.7 as determined by gel permeation chromatography) Birmingham, AL and NaCl (Acros, Pittsburgh, PA). 100% ethanol was purchased from Pharmco (Brookfield, CT). Trypsin/EDTA (ethylenediamine tetraacetic acid), phosphate-buffered saline (PBS) and high glucose Dulbecco's modified eagle's medium (DMEM) were purchased from Gibco BRL

Cell density and growth in foams

Cell densities per PLGA foam were determined by a total DNA assay. Cells maintained 1 day in static culture had a measured cell density of 8.6 (±1.2)×105 cells/foam. From the dimensions and porosity of the foam a void space per foam of 0.60 cm3 can be calculated. With a seeding cell suspension of 2×106 cells/cm3, optimal loading would be 1.2×106 cells/foam. For this study a 72% loading efficiency was achieved (neglecting cell growth). After 7 days in culture cell densities had increased roughly

Discussion

In this study four culturing systems were examined to ascertain the effect of fluid flow on cell growth and osteoblastic function in large porous scaffolds. Measurements of cell density and thymidine incorporation revealed similar cell growth in all culturing systems, indicating that nutrient diffusion did not limit cell growth in this study. However, histological sections demonstrated differences in cell distribution: cells were uniformly distributed in foams cultured in the rotary vessel and

Conclusions

In this study four different culturing schemes were examined for their ability to promote growth and osteoblastic function of cells seeded in porous PLGA foams. Although the techniques yielded similar cell densities, foams cultured in a rotary vessel or in a flow perfusion system had the most uniform distribution of cells. In contrast foams cultured in the rotary vessel had the lowest levels of ALP activity whereas those cultured in the perfusion system or in a spinner flask demonstrated

Acknowledgements

This work was funded by the National Institutes for Health (R29-AR42639) and the National Aeronautics and Space Administration (NAGW-5007, NAG5-4072). The authors would like to thank the assistance of Dr. John Olson with DNA measurements.

References (37)

  • J.E. Dennis et al.

    Osteogenesis in marrow-derived mesenchymal cell porous ceramic composites transplanted subcutaneouslyeffect of fibronectin and laminin on cell retension and rate of osteogenic expression

    Cell Transplant

    (1992)
  • P.A. DiMilla et al.

    Maximal migration of human smooth muscle cells on fibronectin and type IV collagen occurs at an intermediate attachment strength

    J Cell Biol

    (1992)
  • G. Ferrari et al.

    Muscle regeneration by bone marrow-derived myogenic progenitors

    Science

    (1998)
  • J. Folkman et al.

    Self-regulation of growth in three dimensions

    J Exp Med

    (1973)
  • J.A. Frangos et al.

    Shear stress induced stimulation of mammalian cell matabolism

    Biotechnol Bioengng

    (1988)
  • L.E. Freed et al.

    Cultivation of cell-polymer constructs in simulated microgravity

    Biotechnol Bioengng

    (1995)
  • A.S. Goldstein et al.

    Effect of osteoblastic culture conditions on the structure of poly(dl-lactic-co-glycolic acid) foam scaffolds

    Tissue Engng

    (1999)
  • R. Gundle et al.

    Isolation and culture of human bone-derived cells

  • Cited by (0)

    1

    Present address: Department of Chemical Engineering, Virginia Polytechnic Institute, Blacksburg, VA 24061, USA.

    View full text