Physiological strains induce differentiation in human osteoblasts cultured on orthopaedic biomaterial.

We have developed an in vitro mechanical stretching model of osteoblastic cells cultured on metallic biomaterials in order to study the effects of mechanical strain on osteointegration of orthopaedic implants. Titanium alloy discs coated with alumina or hydroxyapatite were used as substrates. Three Dynacell devices were especially designed to apply cyclic strains on rigid biomaterials. The regimen (600 mu epsilon strains, 0.25Hz) was defined on the basis of physiological data and estimated deformation on hip stem prostheses. The performances of these apparatus were reproducible and provided controlled deformations. Human osteosarcoma cell line MG-63, human osteoblasts obtained from primary cultures and ROS 17/2.8 rat osteosarcoma cells were used as cell models. Cell behaviour was assessed in terms of growth and alkaline phosphatase (ALP) activity by in situ assays for two regimens: 15-min deformations repeated three times a day to mimic rehabilitation exercises and 24-h continuous deformations. We demonstrated that continuous deformation did not affect the growth and ALP activity of MG-63 cells, in contrast with sequential deformations which had no effect on cell number, but which stimulated ALP activity after 5 days of stretching. This sequential regimen can also modify the behaviour of human bone-derived cells resulting in increased proliferation after 5 days and stimulation of ALP activity after 15 days. ROS 17/2.8 rat osteosarcoma cells submitted to sequential deformations responded faster than other cell lines by increasing their ALP activity only after 1 day of stretching. Like MG-63 cells, proliferation of the ROS 17/2.8 rat osteosarcoma cell line was not affected by sequential deformations. This study suggests that short, repeated deformations defined to mimic rehabilitation exercises recommended after prostheses implantation are more likely to exert beneficial effects on implanted bone than continuous strains.

Practice of a testing bench to study the effects of cyclic stretching on osteoblast-orthopaedic ceramic interactions.

A new experimental method has been used to study the behaviour of human osteoblasts cultured on bioceramics subjected to mechanical strains. The ceramics were alumina, hydroxyapatite (HA) and a duplex system composed of hydroxyapatite-covered alumina. The system applied 400 microdeformations for a 6-h period with a cycle frequency of 0.5 Hz to osteoblasts growing on ceramic-covered disks. The effects of strains on short-term cell viability, cell growth, alkaline phosphatase (ALP) activity, and collagen biosynthesis were assessed. When possible, the parameters (lactate dehydrogenase) were studied along the experiment in samples of the culture medium, in the other cases by comparison of stretched and unstretched cultures on the same ceramics with the same cell line. In relationship with the coating, mechanical strains resulted in a decrease in DNA corresponding to cell number, an LDH release during straining, an unchanged (alumina) or decreased (HA and duplex) ALP activity, a decrease (HA and duplex) of collagen and total protein synthesis or an increase of it (alumina). The stress-producing device and its associated protocol are shown to be suitable for investigating the behaviour of cells, cultured on biomaterials subjected to mechanical strain.

Mineralization and alkaline phosphatase activity in collagen lattices populated by human osteoblasts.

Adult human osteoblastic cells were grown in a native type I collagen gel. Proliferation and viability analyses showed that cells rapidly stopped dividing and became blocked in the G0G1 phase (91% on day 13). Carboxyfluorescein diacetate cell staining and flow cytometry showed that osteoblasts were viable for the first 16 days and then viability decreased (58% viable cells on day 22). Osteoblasts were able to retract the matrix. Betaglycerophosphate (betaGP) stimulated the deposition of mineral particles in the collagen network, and electron probe microanalysis showed that they were principally calcium and phosphorus, with a Ca/P ratio of about 1.7. Various times of betaGP supply were tested. We compared 10 mM betaGP added only once at day 0, or continuously from day 0, day 8, or day 21. Mineralization was observed in conditions where betaGP was added at day 0. Furthermore, 10 mM betaGP added once during gel preparation was sufficient to induce mineralization with mineral accumulation up to day 15 whereas the speed of the gel contraction decreased. In every condition, cultures expressed high alkaline phosphatase (ALP) levels as early as day 3, which decreased afterwards. These kinetics might explain why the other conditions did not prove favorable to the mineralization process. The model was used to study the influence of blocking gel retraction. Blocking retraction delayed the ALP activity decrease, but had no effect on mineralization. In conclusion, human adult osteoblasts cultured in native collagen gel stopped proliferation and underwent mineralization very early. This model should be used to investigate the influence of effectors on the early stages of culture.

Effects of gamma-alumina and hydroxyapatite coatings on the growth and metabolism of human osteoblasts.

The behavior of human osteoblasts cultivated on hydroxyapatite or alumina-coated disks of Ti6AL4V was studied in vitro. Cell anchorage and spreading were observed by scanning electron microscopy. Cell growth was monitored by counting cells and measuring DNA at 5 h and 2,5, and 10 days after cell seeding. Cells grown for 10 days were labeled with 14C-proline and total protein and collagen synthesis were measured; the type of collagen was also determined. Both ceramics showed excellent biocompatibility. At 10 days of culture the cells showed a higher rate of proliferation on alumina than on hydroxyapatite. Neither ceramic altered the collagen biosynthesis or the osteoblast-like properties of the cells, as indicated by the high percentage of type I collagen.

In-vitro responses to ascorbate and manganese in fibroblasts from a patient with prolidase deficiency and iminodipeptiduria: cell growth, prolidase activity and collagen metabolism.

After successful ascorbate and manganese treatment of a female patient with prolidase deficiency and iminodipeptiduria, we attempted to explain the mechanism of action of these drugs in vitro, using them preferentially on skin fibroblasts. Since in vivo, ascorbate and manganese seemed to be responsible for both biochemical and clinical improvement, they were also expected to activate prolidase activity in vitro. Cell growth and prolidase activity were accordingly observed in fibroblast cultures supplemented with these compounds. It seemed that only ascorbate accounted for the successful in vivo response. To understand the mechanism involved, we studied collagen metabolism and found a decreased proline pool, a massive increase of rapidly degraded collagen and moderate enhancement of type III collagen and type I trimer in the patient's fibroblasts. We believe that ascorbate allowed the prolidase-deficient cells to maintain a normal collagen pool by increasing collagen synthesis. Both the massive increase in cell growth in response to ascorbate and the bad response as regards the quality of the collagen produced confirm the secondary nature of this mechanism. However, the relationship between accelerated collagen catabolism and prolidase deficiency remains unclear.