Differences in Osteogenic and Apoptotic Genes Between Osteoporotic and Osteoarthritic Patients

Results

We analyzed data from 12 patients: 6 with osteoporotic hip fracture (OP) and 6 with hip osteoarthritis (OA), as the control group, who fulfilled the enrolment criteria. Clinical characteristics and laboratory data of the OP and OA groups are summarized in Table 1.

In both groups, the anthropometric and biochemical characteristics were comparable. Although patients with fractures tended to be older than those with OA, age differences did not seem to influence the results because we found no statistical significant differences between groups and also we found no correlation between age and gene expression in either group.

OP patients had lower values of total hip bone mineral density (BMD) (0.83 gHA/cm ² ± 0.04 OP; 0.89 gHA/cm ² ± 0.06 OA) and neck BMD (0.59 gHA/cm ² ± 0.03 OP; 0.74 gHA/OA ± 0.08 cm ²⁾ than the control group, and higher bone remodelling serum marker, β-CrossLaps, 0.51 ng/mL ± 0.1 (OP) vs. 0.36 ng/mL ± 0.08 (OA) without rearching statistical differences. In all patients, the levels of 25 (OH) D were < 20 ng/mL.

Histomorphometry and Biomechanics Results

Microstructural indices show a lower bone quality in the group of patients with hip fractures compared to the osteoarthritic group (Table 2).

We observed significant lower number of trabeculae (Tb.N.) in the bone of hip fracture patients (1.5 mm-1 ± 0.1 vs 1.9 mm-1 ± 0.1; p = 0.049) and furthermore, these patients also have a greater separation between trabeculae (Tb.Sp) (0.55 mm ± 0.46 vs. 0.02 mm ± 0.03, P = 0.04). The thickness of the trabeculae (Tb.Th) was similar in both groups.

The trabecular pattern index (inverse of connectivity index) showed higher values in the fractured patient group. The structure model index (SMI), which implies a relative prevalence of tube-cylinder trabeculae in comparison to plate-form ones, was higher in the group of patients with hip fracture, indicating a greater number of trabeculae in tube-shaped cylinder, less resitant, in these patients.

Biomechanical studies showed that the values of Young's modulus, maximum stress and maximum deflection, after applying the compression test, were lower in patients with hip fractures compared with osteoarthritic patients, although these differences did not reach significance (Table 2).

PCR Array Osteogenesis

We examined whether osteoporotic patients would have any alterations in the osteogenesis in comparison with osteoarthritic patients. If we consider biologically significant variations when the induction or repression of the gene is at least 3 times, we found that 23 genes, from 86 genes involved in osteogenesis, are altered (p <0.05). In most cases, we observed a repression of these genes in osteoporotic compared with the control bone (Table 3). Only FGF1 and VEGFA, genes involved in cell cycle regulation, are overexpressed (3.3 and 3.0-fold, respectively). From the family of metalloproteases, only MMP8 was up-regulated, and it is involved in inflammatory joint processes (5.06-fold).

Among the down-regulated genes, it is necessary to emphasize the osteocalcin gene (BGLAP) (− 3.39-fold), a specific gene from osteoblastic cells which function is bone mineralization, CDH11 (−3.3-fold) a typical adhesion molecule of osteoblasts, Runx2 (−2.94-fold) which is a factor of transcription essential for the maturation of osteoblastic cells and BMP6 (−3.3-fold) that encodes a protein of bone formation, also typical from osteoblastic cells. The latter four genes are characteristic of osteoblastic cells and are suppressed on the macerated OP bones.

We also showed down-regulation in several transcription factors such as NFKB1, SMAD2, SMAD3, SOX9 and TNF, which in turn modulate gene expression of several genes.

PCR Array Apoptosis

When analyzing array apoptosis, we observed that most of the genes show a decrease in gene expression in the osteoporotic group in comparison with the osteoarthritic group (Table 4).

As known, apoptosis can be initiated either by the death receptor-mediated pathway or the mitochondria-mediated pathway. In our study, we found that both pathways are altered on macerated OP bones. Of the 84 genes studied, 26 have a significant lower gene expression in the OP group vs. OA, and only two of them show increased expression (p <0.05). Specifically, we observed a lower expression of TNF receptor family genes as CD40 (−7.7-fold OP vs. OA), LTBR (−3.3-fold), TNFRSF11B (−3.5-fold), CD27 (−6.6-fold) and the TNF ligand Family, TNF (−16.7-fold). At a mitochondrial level, there are repressed genes such as BAD (−5.8-fold) and slightly BAX (−1.2-fold).

Caspases are final effector enzymes of programmed cell death. In the OP group, the caspases studied also have a lower gene expression, observing a significant biologically variation for caspase 9 (−6.1-fold OP vs. OA) and caspases 4 and 5 (−5.3- fold and −5.9-fold-respectively).

We have also observed the repression of genes such as Bcl2 (−6.3-fold) and BAG1 (−3.62-fold), both of the bcl-2 family and ABL1 (−6.8-fold) involved in the damage response DNA.

In OP patients, we only observed an increased in the gene expressions of BCL2A1 (9.66-fold) and BNIP1 (2.7-fold), both from the BCL2 family of anti-apoptotic features.

To verify the results obtained in the array plates, we chose two genes, with particular relevance to the activity of bone remodelling, such as Runx2 and OPG. Runx2 is the major transcription factor involved in osteoblast differentiation, and the results of the osteogenesis array plate indicate a down-regulation in OP vs. OA bone of this gene (−2.94-fold, p <0.05) (Table 3). OPG is part of one of the main systems of regulation of bone remodelling (OPG/RANKL/RANK) and the array plate are decreased apoptosis expression in the OP group compared with the OA group (−3.52-fold, p < 0.05) (Table 4).

When analyzed individually OPG and Runx2 gene, we observed that the OP group has almost 2 times less expression of Runx2 and 6.9 times less OPG than the control group (p = 0.004), confirming the results of osteogenesis and apoptosis arrays (Figure 1).

BMC Musculoskelet Disord. 2013;14(41) © 2013  BioMed Central, Ltd.

© 1999-2006 BioMed Central Ltd