Raloxifene ameliorates detrimental enzymatic and nonenzymatic collagen cross-links and bone strength in rabbits with hyperhomocysteinemia

We demonstrate a reduction in enzymatic divalent immature and trivalent pyridinium cross-links and an increase in the nonenzymatic cross-link, pentosidine (Pen), in rabbits with methionine (Met)-induced hyperhomocysteinemia. Such detrimental cross-link formation in bone was ameliorated by raloxifene (RLX) treatment.

INTRODUCTION

Collagen cross-links are determinants of bone quality. Homocysteine (Hcys) interferes with collagen cross-linking. Because RLX is thought to ameliorate bone quality, we investigated whether RLX ameliorated hyperhomocysteinemia-induced cross-link abnormalities using a Met-rich diet rabbit model.

METHODS

We divided New Zealand white rabbits into six groups (n = 6 per group): baseline control, sham operation, sham + 1% Met diet, ovariectomy (OVX), 1% Met diet + OVX, OVX + RLX (10 mg/kg/day), and 1% Met diet + OVX + RLX. RLX was administered for 16 weeks. We measured the amount of enzymatic immature and mature pyridinium cross-links and the nonenzymatic cross-link, Pen, and correlated the cross-link content to bone strength.

RESULTS

Hcys levels were significantly higher in the Met diet groups than in the normal diet groups. Met-fed rabbits with or without OVX showed a significant reduction of enzymatic cross-links, whereas an increase in Pen was observed in Met-fed rabbits with OVX. The cross-link content of the RLX-treated Met-fed rabbits with OVX was restored to similar levels as the sham group, accompanied by an improvement of bone strength.

CONCLUSION

These results demonstrate that hyperhomocysteinemia reduced bone strength via a reduction of enzymatic cross-links and an increase of nonenzymatic cross-links. RLX may ameliorate hyperhomocysteinemia-induced detrimental cross-linking in rabbits with OVX and may improve bone strength via the amelioration of collagen cross-links.

Reductions in degree of mineralization and enzymatic collagen cross-links and increases in glycation-induced pentosidine in the femoral neck cortex in cases of femoral neck fracture

INTRODUCTION

Enzymatic and glycation-induced nonenzymatic cross-links play important roles in the expression of bone strength. The cross-link pattern is affected by tissue maturation and senescence. The aim of our study was to understand the distinctive posttranslational modifications of collagen in areas with different degrees of mineralization with and without hip fracture.

METHODS

Sixteen female cases of intracapsular hip fracture (78+/-6 years) and 16 age- and gender-matched postmortem controls (76+/-6 years) were included in this study. A sample of each femoral neck cortex was fractionated into low (1.7 to 2.0 g/ml) and high (>2.0 g/ml) density portions. The contents of enzymatic cross-links (dihydroxylysinonorleucine, hydroxylysinonorleucine, lysinonorleucine, pyridinoline, and deoxypyridinoline) and nonenzymatic cross-links (pentosidine) and the extent of lysine (Lys) hydroxylation were determined in each fraction.

RESULTS

In the controls, there was no significant difference in the contents of enzymatic cross-links between low- and high-mineralized bone fractions whereas pentosidine content was significantly higher in high-mineralized bone compared with low-mineralized bone (p=0.0014). When comparing enzymatic cross-link contents between controls and fracture cases, a trend toward lower (p=0.0961) cross-link content in low-mineralized bone and a significant reduction (p<0.0001) in high-mineralized bone were observed. Pentosidine content of low-mineralized bone was significantly higher in fracture cases than in controls (p<0.0001). The extent of Lys hydroxylation was significantly higher in fracture cases than in controls (p<0.001). The higher hydroxylation of Lys in collagen from fracture cases relative to controls was associated with significantly higher values of hydroxylysine-derived cross-link such that the enzymatic cross-link patterns correlated with the extent of Lys hydroxylation in the collagen molecules.

CONCLUSIONS

These results suggest that reductions in the degree of mineralization and enzymatic cross-links and excessive formation of pentosidine may play an important role in explaining poor bone quality in osteoporosis.

Effect of low- and high-intensity pulsed ultrasound on collagen post-translational modifications in MC3T3-E1 osteoblasts

Different intensities of pulsed ultrasound have distinct biological effects on bone mineralization in the process of bone fracture repair, even across a narrow range (e.g., 30-120 mW/cm(2)). The aim of our study was to elucidate the effect of low-intensity (30 mW/cm(2)) and high-intensity (120 mW/cm(2)) pulsed ultrasound on collagen metabolism by using MC3T3-E1 osteoblasts. Of special interest was the relationship between posttranslational collagen quality and prostaglandin E(2) activity. Cells with or without a cyclooxygenase-2 inhibitor, NS398, were exposed every day for four consecutive days to high-level or low-level intensities of pulsed ultrasound. We examined the, expression patterns of cyclooxygenase-2, lysyl oxidase, telopeptidyl lysyl hydroxylase (TLH), and helical lysyl hydroxylase by real-time polymerase chain reaction analysis. Quantitative analyses of reducible immature and nonreducible mature cross-links were also performed. Ultrasound at 30 mW/cm(2) upregulated TLH messenger RNA (mRNA) expression and enzyme activity compared to the control and resulted in increased relative concentrations of telopeptidyl hydroxylysine-derived cross-links. In addition to upregulated lysyl oxidase mRNA expression, increased total reducible and nonreducible cross-links were observed by 30 mW/cm(2) exposure compared to the control. In contrast, ultrasound at 120 mW/cm(2) had no obvious effect on collagen metabolism owing to high levels of endogenous prostaglandin E(2) induced by ultrasound. Our results showed that (1) low-intensity, but not high-intensity, ultrasound may accelerate the formation of the unique molecular packing of collagen fibers conducive to bone mineralization and that (2) the high dose of endogenous prostaglandin E(2) induced by pulsed ultrasound may be detrimental to calcifiable cross-link formation.

A MALIGNANT PERIPHERAL NERVE-SHEATH TUMOUR RESPONDING TO CHEMOTHERAPY

Amalignant peripheral nerve-sheath tumour developed in the right S1 nerve root in a man aged 30 causing back pain and sciatica. CT and MRI revealed a destructive tumour of the sacrum invading the retroperitoneal space. The tumour was not resectable with an adequate margin. Chemotherapy, consisting of high-dose ifosfamide followed by a combination of vincristine, doxorubicin and cyclophosphamide, was given with success. Malignant peripheral nerve-sheath tumours are thought to respond weakly to chemotherapy, but the response in our patient was complete.

A malignant peripheral nerve-sheath tumour responding to chemotherapy

A malignant peripheral nerve-sheath tumour developed in the right S1 nerve root in a man aged 30 causing back pain and sciatica. CT and MRI revealed a destructive tumour of the sacrum invading the retroperitoneal space. The tumour was not resectable with an adequate margin. Chemotherapy, consisting of high-dose ifosfamide followed by a combination of vincristine, doxorubicin and cyclophosphamide, was given with success. Malignant peripheral nerve-sheath tumours are thought to respond weakly to chemotherapy, but the response in our patient was complete.

Local effects of transforming growth factor-beta1 on rat calvaria: changes depending on the dose and the injection site

Transforming growth factor-beta1 (TGF-beta1) has opposite effects on osteoblastic cells in vitro, namely an inhibitory or stimulatory effect on cell differentiation. Because these effects are dependent on TGF-beta1 concentration or culture condition, we investigated whether the in vivo effects of TGF-beta1 on bone formation in infant rat calvaria were affected by the dose or the injection site. Human platelet-derived TGF-beta1 was injected subcutaneously onto the periosteal surface of parietal bone of 4-week-old rats at doses of 5 or 20ng/100microl per animal for 14 days, and the local effect on bone formation was examined by bone histomorphometry. TGF-beta1 treatment for 7 days decreased the mineral apposition rate, bone formation rate, and elongated mineralization lag time at the injection site. This change became more prominent when treatment continued for 14 days. These changes were restricted to the TGF-beta1-exposed area. Multiple subcutaneous injections of a relatively high dose (200ng/100microl per animal) of TGF-beta1 induced woven bone formation, in addition to marked inhibition of bone formation rate and prolongation of mineralization lag time. On the other hand, direct exposure of TGF-beta1 in the subperiosteal layer induced woven bone with periosteal cell proliferation even at a single injection of a low dose (5 or 50ng/25 microl) of TGF-beta1. In conclusion, the in vivo effects of TGF-beta1 on bone formation varied depending on its concentration and injection site. Also, subcutaneous injection of relatively low doses of TGF-beta1 inhibited local lamellar bone formation.

Effect of recombinant human granulocyte colony-stimulating factor (rh G-CSF) on rat bone: inhibition of bone formation at the endosteal surface of vertebra and tibia

The effect of recombinant human granulocyte colony-stimulating factor (rh G-CSF) on bone was evaluated by histomorphometry using Sprague-Dawley rats. rh G-CSF was injected at doses of 0, 50, 150, and 450 microg/kg for 6 weeks. In vivo double fluorochrome labeling was performed before sacrifice. No significant change in body weight was observed. Bone mineral density (BMD) of lumbar vertebrae and femora was significantly decreased in G-CSF-treated groups. In the lumbar vertebra, osteoid surface, osteoid thickness, trabecular thickness, and labeled surface in G-CSF-treated groups were also significantly lower. In addition, osteoclast number and osteoclast surface were significantly higher in the G-CSF-treated groups. The endocortical surface at the mid-tibia showed lower labeled surface and mineral apposition rate in G-CSF-treated groups, without significant changes at the periosteal surface. Furthermore, numerous granulocytes fully occupied the bone marrow area. We conclude that proliferating granulocytes in the bone marrow may inhibit bone-forming cells from contacting the bone surface, resulting in reduction of bone formation; and increased osteoclastic bone resorption induced by G-CSF treatment contributed to the reduction of BMD.

Use of a guanidine extract of demineralized bone in the treatment of osteochondral defects of articular cartilage

In order to evaluate the ability of a guanidine extract of demineralized bone to repair osteochondral defects in articular cartilage, plugs made of this extract were implanted into defects in rabbit knees. The repair tissue was examined macroscopically, histologically, and immunohistochemically at 4, 8, 12, and 30 weeks. Controls (defects that were left empty) showed no cartilage formation. Four weeks after implantation of a guanidine extract plug, histological examination showed a nonhomogeneous metachromatically stained region extending from the surface of the repair tissue down to cancellous bone. This region also was labeled by an anti-type-II collagen antibody, indicating that cartilage-like tissue had been induced. At 8 weeks, the newly formed cartilage in the subchondral and cancellous bone had been partially replaced by bone. At 12 weeks, the thickness of the newly formed cartilage layer had decreased, and most of the newly formed cartilage in the subchondral and cancellous bone had been replaced by bone. In addition, a tidemark was observed. At 30 weeks, the repair tissue was a mixture of cartilage and fibrocartilage, and there was severe degeneration of the cartilage surrounding the repaired defects. These findings indicate that osteochondral defects of articular cartilage can be partially repaired by the implantation of a guanidine extract and that the newly formed cartilage-like tissue is not permanent.

An experimental study on transpedicular screw fixation in relation to osteoporosis of the lumbar spine

In order to elucidate the relationship between the severity of osteoporosis and the fixation strength of a pedicle screw, screw pull-out tests were performed using cadaveric lumbar vertebrae. The severity of osteoporosis was evaluated by the Jikei osteoporosis grading scale (Jikei method), bone mineral density, and microdensitometry. When a 7.0-mm screw was used, the pull-out force of the screw was 1,056.4 N in the normal group (as determined by the Jikei method), while it was 495.6 N in the Grade I osteoporosis and 269.5 N in the Grade II osteoporosis groups, respectively. There were also positive correlations between the pull-out force and bone mineral density and each parameter of the microdensitometry method. When bone cement was used in an osteoporotic vertebra, twofold stronger pull-out force was obtained in comparison to that obtained without bone cement.