PERICELLULAR PARTIAL OXYGEN PRESSURE (pO₂) MEASUREMENT IN OSTEOPENIC BONE-DERIVED OSTEOBLAST CULTURES

SUMMARY

In order to achieve further information on the in vitro behaviour of osteoblasts derived from osteopenic bone, in the present study comparative measurements of some parameters of cell proliferation, metabolism and differentiation and also of the pericellular partial oxygen pressure (pO₂) were performed on normal and osteopenic bone derived osteoblasts from heathy and osteopenic rats.

The respiration rate was increased in osteoblasts derived from osteopenic bone as compared to normal cells at 48 hours and 7 days, involving a significant decrease in pericellular pO₂ in the culture medium. At 48 hours, in osteopenic bone-derived cells, a significant increase in MTT and a significant decrease of osteocalcin were observed. At 7 days, cell count highlighted a significant slowing down of the proliferation of osteopenic bone-derived osteoblasts. No significant differences were observed for alkaline phosphatase activity, nitric oxide and type I collagen production. The present preliminary results may be taken into consideration also in in vitro comparative biocompatibility or osteointegration studies of biomaterials in normal and osteopenic bone-derived cells because a decrease in pericellular pO₂ in these tissue cultures could influence results on material behaviour.

• Osteoblasts
• Osteopenia
• Pericellular oxygen pressure

INTRODUCTION

During the last 10 years, in vitro studies on the aged and osteopenic or osteoporotic skeleton have improved [1-15]. Thanks to the great advances made in the study of in vitro systems with primary osteoblast and bone marrow stromal cell cultures, it is now possible to investigate numerous aspects of osteopenia and osteoporosis that, due to the ageing population and longer life expectancy, can be considered as one of the major social health problems [[16]].

Osteoblasts derived from osteopenic bone are used for both medical and surgical studies. Medical researches have focused primarily on pathophysiology and innovative treatments of the disease, investigating also drawbacks or abnormalities of osteoblasts derived from osteopenic or aged bone of patients or animals. Surgical researches start from the in vivo experimental and clinical evidence that the same prosthetic device implanted in osteopenic/osteoporotic and normal bone presents a different osteointegration rate [17-21]; using in vitro studies, researchers aim at testing biomaterial biocompatibility and osteointegration in cell cultures derived from osteopenic bone, and these evaluations seem to be promising in predicting the in vivo behaviour, even if many other assessments are mandatory [22-23]. This approach has the advantage of testing biocompatibility and osteointegration of biomaterials in a microenvironment more similar to the clinical situation.

Nowadays, there is still debate over the existence of in vitro differences between osteopenic bone-derived osteoblasts and normal bone-derived ones, and sometimes results obtained by different authors on stimulated and unstimulated cells are controversial. Moreover, because of the great variability and heterogeneity of human osteoporosis, it is difficult to compare data achieved by different authors, who have used osteoblasts derived from osteoporotic patients [24-26].

Polarography is an electrochemical analytical technique commonly used for the qualitative and quantitative evaluation of the ions present in a solution; it is based on the measurement of the electric current through an electrolytic cell, when a potential difference is created between a metallic and a reference electrode. In the proximity of a polarographic electrode, the oxygen (O₂) of the culture medium is converted into hydroxylic ions and intermediate compounds, such as hydrogen peroxide and superoxide ions, thus generating an electric current proportional to the O₂ pressure. Therefore, the availability of O₂ on the cellular levels, pericellular partial O₂ pressure (pO₂), results from the balance between the local (air-medium) diffusion rate of O₂ and its cellular uptake due to the O₂ metabolising mitochondria. The current authors have previously evaluated the possibility to use this method for the evaluation of cytotoxicity by comparing and correlating pericellular pO₂ with MTT data, showing that polarography is an alternative methods to test cell viability [[27]].

The aim of the present study was to increase the knowledge on the in vitro behaviour of osteoblasts derived from osteopenic bone, by doing comparative measurements of some parameters of cell proliferation, metabolism, differentiation and also of the pericellular pO₂, being cellular O₂ availability important also for the interpretation of biochemical and toxicological cell culture studies. Healthy and osteopenic bone-derived cells were obtained from the trabecular bone of the distal femurs of normal and ovariectomized rats (3 months after ovariectomy), with evidence of demonstrated post-menopausal osteopenia in the cell donor area [[23]].

MATERIALS AND METHODS

Animals

The study was performed in compliance with the European and Italian Laws on animal experimentation and with the Animal Welfare Assurance No #A5424–01 of the National Institute of Health (NIH-Rockville Maryland USA) and the animal research protocol was approved by the responsible public authorities as requested by the Italian Law according to the EC rules (Law by Decree, 27 January, 1992 no. 116).

For primary osteoblast cultures, trabecular specimens were obtained from the right distal femurs of two different groups of animals, one group of healthy rats and one group of rats that had been ovariectomized (OVX) 3 months before and had undergone bilateral ovariectomy through lumbar access and under general anaesthesia, by subcutaneous injection of 87 mg/kg ketamine (Ketavet, Farmaceutici Gellini, Aprilia Lt-Italy) and 13 mg/kg xylazine (Rompun, Bayer Italy spa, Milano-Italy). All of the animals were pharmacologically euthanized under general anaesthesia and, in surgical asepsis, the right distal femur was explanted and cleaned of soft and cartilagineous tissues. Fragments of trabecular bone were put in DMEM:F12 serum-free medium for subsequent isolation of cells as described later.

Also the left distal femurs were explanted and cleaned of soft tissues, in order to demonstrate the development of an osteopenic state, particularly in the bone area used for cell cultures: dual X-ray absorptiometry (DXA) at the distal epiphysis below the supracondylar line and histomorphometric evaluation on undecalcified sections were performed. OVX rats were characterised as osteopenic, because, when compared to healthy rats, femoral densitometric data registered a significant decrease of 15.1% in BMD and histomorphometric results showed a cancellous bone loss in the femoral distal epiphysis. Comparison of the two groups of rats revealed that the cancellous bone loss was mainly due to a 27% decrease in trabecular bone volume, an 18% decrease in trabecular thickness, a 35% decrease in the index of spatial connectivity, and a 33% increase in trabecular separation.

Cell Cultures

Trabecular fragments were washed with DMEM:F12 serum-free medium, and digested in F12 medium with 1 mg/ml collagenase for 90 min. at 37°C. The enzymatic reaction was stopped by adding an equal volume of medium with 10% of FCS, and the supernatant containing the released cells was collected. Washing and collecting were repeated three times. The cells obtained were pelletted by centrifugation, resuspended, seeded in 35 mm dishes and incubated at 37°C in a humidified atmosphere of 5% CO₂ / 95% air. Cells were released at confluence with trypsin: EDTA. In order to assess the ability of rat cells to mineralise their extracellular matrix, confluent cultures were grown in chamber slides, where 1,25(OH)₂D₃ (10⁻⁹M) was added. Supernatant was collected to measure osteocalcin (OC: Metra Biosystem, Ca, USA, immunoenzymatic assay) and alkaline phosphatase activity (ALP: Sigma, UK, Kinetic method Kit). The cells were then fixed and stained with the von Kossa method (BioOptica, Milan, Italy), to reveal the deposition of calcium in the extracellular matrix. These tests confirmed that cells had the capacity to differentiate and mineralise in vitro, thus showing an osteoblastic behaviour.

Then, cells were seeded for experiment (1.2 ± 0.1 × 10⁴cell/ml) at the first passage in 4-well chamber slides, and cultured in DMEM:F12 medium to which ascorbic acid (50 μg/ml) and β-glycerophosphate (10⁻⁸M) were added. In most cases, bone cells formed a confluent monolayer afterwards.

The following tests were performed at 48 hours and at 7 days on normal and osteopenic bone-derived cells. The supernatant was collected at the established times for the spectrophotometric dosages of ALP (as previously described), Nitric Oxide production (NO, Sigma, UK, assessed by the Griess reaction) and OC (as previously described). Cell proliferation was assessed by the tetrazolium salt test (MTT). MTT (500 μg/well) was added during the last 4 h of the experiment. After removing supernatants, the formazan crystal was solubilized in 1 ml of DMSO and absorbance was measured spectrophotometrically at 540 nm.

Pericellular pO₂ was measured following a method described previously [27-28]. In summary, pericellular pO₂ was measured by the Licox pO₂ computer (GMS mbH, Kiel, Germany) and its specific Clark rod-type probe (sensitivity _35°C∼6*10⁻¹² A/mmHg pO₂) was placed in each well at 37°C in 5% CO₂ for 2 min. The oxygen in the solution underwent electrolytic reduction at the cathode of the probe, with the production of a current which was measured by an appropriate amplifier system and recorder.

Then cells were fixed and stained for the presence of type I collagen (monoclonal anti-collagen type 1, Sigma, UK, immunofluorescence method). Slides were observed at a fluorescence light microscope (Ziess Axioscope, Carl Zeiss, Jena, Germany) and analysed by an image computerised system (Kontron KS300 v.2, Kontron Elektronik, München, Germany). The amount of collagen I was adjusted for cell number. Cells were counted by Trypan Blue stain and hemocytometer chamber.

Ten cultures were obtained for each group and each culture was performed in triplicates.

Statistical Analysis

Statistical analysis was performed with SPSS v.7.5 software (SPSS/PC Inc., Chicago, Illinois). Data are reported as mean ± SD of triplicates at a significance level of p<0.05. After having analysed data for normal distribution and homogeneity of variances, non parametric Wilcoxon test was used to compare data between normal and osteopenic bone-derived osteoblast cultures at both 48 hours and 7 days. Non parametric Spearman' rho coefficient was used to highlight significant correlation between tested parameters.

RESULTS

Results at 48 hours and at 7 days are reported in Tables 1 and 2, respectively. Wilcoxon W statistic showed significant differences existed between normal and osteopenic bone-derived osteoblast cultures for OC (W=63.0, p=0.044), MTT (W=61.0, p<0.001) and pericellular pO₂ (W=55.5, p<0.0005) at 48 hours, and for cell count (W=49.0, p=0.004) and pericellular pO₂ (W=55.0, p<0.0005) at 7 days.

Table 1. Results of Biochemical Parameters, Gene Expression, and Pericellular pO₂ at 48 Hours in Normal and Osteopenic Bone-Derived Osteoblasts (Mean ± SD; n=10 Triplicates)

	Osteoblasts
	Normal bone-derived	Osteopenic bone-derived
ALP (IU/l)	14.87 ± 2.92	14.79 ± 3.64
NO (μM)	2.70 ± 1.07	2.60 ± 0.37
OC (ng/dl)	15.47 ± 1.85	13.67 ± 2.01
MTT (OD)	0.46 ± 0.05	0.56 ± 0.05
Type I collagen (%)	11.46 ± 1.09	11.17 ± 1.28
Cell count (x10^4 cell/ml)	2.10 ± 0.24	2.27 ± 0.61
pO2 (mm Hg)	111.00 ± 2.16	104.41 ± 2.11

Table 2. Results of Biochemical Parameters, Gene Expression, and Pericellular pO₂ at 7 Days in Normal and Osteopenic Bone-Derived Osteoblasts (Mean ± SD; n=10 Triplicates)

	Osteoblasts
	Normal bone-derived	Osteopenic bone-derived
ALP (IU/l)	12.23 ± 2.56	12.77 ± 3.59
NO (μM)	4.72 ± 0.23	4.48 ± 0.40
OC (ng/dl)	15.23 ± 1.27	16.60 ± 3.45
MTT (OD)	1.47 ± 0.16	1.58 ± 0.22
Type I collagen (%)	10.48 ± 0.40	10.17 ± 0.59
Cell count (x10^4 cell/ml)	5.65 ± 1.20	3.67 ± 1.17
pO2 (mm Hg)	84.50 ± 1.14	80.75 ± 1.18

Significant correlation between tested parameters is reported in Table 3. In particular, highly significant correlation (p<0.01) was observed between MTT and cell count, MTT and pericellular pO₂ as well as pericellular pO₂ and cell count. NO showed significant correlations (p<0.01) with MTT, pericellular pO₂ and type I collagen, as well as OC presented significant correlation with type I collagen.

Table 3. Spearman' rho Coefficient of Correlation Between Tested Parameters (Mean ± SD; n=40 Triplicates)

	ALP	NO	OC	MTT	I collagen	Cell Count
NO	−0.233
OC	0.181	0.299
MTT	−0.442^b	0.676^a	0.187
I collagen	0.507	0.899^a	0.829^a	−0.486
Cell count	−0.496^b	0.646^b	0.059	0.744^a	−0.147
pO2	0.311	−0.660^a	−0.264	−0.878^a	0.257	−0.676^a

^a, p<0.01;.

^b, p<0.05.

DISCUSSION

The scientific community feels strongly the need for reliable in vitro models to study both the pathophysiology of diseases and the behaviour of biomaterials used for relevant treatment. Regarding this latter aspect, the reliability of models is a very important issue and has to be strictly related to the clinical situation of the human beings, in order to overcome one of the main limits of the in vitro systems. For this reason, tremendous progresses have been made in terms of cell cultures and in vitro tests on biomaterials. In our opinion, one of the most recent advances is the use of “pathological” cells, that is to say cells derived from bone with those pathologies that are more commonly registered in patients considered appropriate candidates for an orthopaedic prosthetic implant. The unpredictability of some surgical implant failures in bone could depend on the fact that biomaterials are tested on healthy normal models, while prostheses are usually implanted in a type of bone which is aged, osteopenic, osteoarthritic, etc. However, there is not agreement between results obtained in the comparison between osteopenic and normal bone-derived osteoblast cultures: no differences were reported on morphology, cell metabolism or differentiation [7-9], [[12]], [[14]] while differences were found particularly in cell proliferation [1-2], [8-9], [12-14] depending also on the high, low or normal bone turnover of patients. Variability in osteoblast dysfunction in osteoporotic patients was well documented by means of histomorphometric studies on iliac crest biopsies and it proved that not only differences from patient to patient [[5]] but also intra-individual variability existed [[12]]: consequently, a great number of cases is mandatory, in order to obtain reliable and significant results when using human bone-derived osteoblasts. In vitro researchers focused on osteoporosis, but by means of mesenchymal stem cells (MSCs) from osteoporotic and normal bone were not discussed because of the differences existing between stromal and trabecular bone cultures [[29]]. For these reasons most of authors stated that further and numerous investigations on the in vitro behaviour of “pathological” cells are mandatory in order to avoid misunderstanding and inappropriate inferences on results.

The current results are a little bit in contrast with those obtained in our previously experiences where no significant differences were detected between normal and osteopenic bone-derived cells when incubated at the same experimental times. Differently from previoulsy experiences, significant differences were found for cell viability at 48 hours and for cell proliferation and viability at 7 days. At 48 hours, a significant increase of MTT and a significant decrease of OC expression were observed in osteopenic bone-derived osteoblasts. At 7 days, in these cells, cell count highlighted a significant slowing down of the proliferation rate of osteopenic bone-derived cells.

Regarding the respiration rate, that, to our knowledge has never been tested in osteopenic bone-derived cells, it appears to be increased in osteoblasts derived from osteopenic bone as compared to normal cells both at 48 hours and 7 days, involving a significant decrease in pericellular pO₂ in the culture medium. The explanation for these findings is still unknown and, in reality, when the present study was planned, the authors thought, they would obtain opposite data. In our opinion, pericellular pO₂ increased consumption and increased MTT could be the expression of an increased metabolic rate of the osteoblasts in the osteoporotic bone which runs out progressively in in vitro conditions. The reduction of bone tissue pericellular pO₂, due to O₂ consumption by osteoblasts, could be also one of the mechanisms that modify bone vascularization by means of hypoxia neoangiogenesis stimulation, which seems to be the cause for an increased bone blood flow in osteoporosis [30-33].

Finally, another remark has to be added: when osteopenic bone-derived cells are used to study biocompatibility, present preliminary results may be taken into consideration also by biomaterial science researchers, because a decrease in pericellular pO₂ in tissue cultures of osteoblasts derived from osteopenic bone, could influence results on material toxicity.

ACKNOWLEDGMENTS

Financial support for this research was given by the Istituti Ortopedici Rizzoli grants (ricerca corrente) and Italian Ministry of Health strategic project “Fratture osteoporotiche”. The authors have appreciated the technical assistance of Patrizio Di Denia, Dalfiume Claudio, Patrizia Nini, Nicola Corrado and Franca Rambaldi (Experimental Surgery Department, IOR)