J. W. M. Chow

Role of Nitric Oxide and Prostaglandins in Mechanically Induced Bone Formation

We have previously shown that prostaglandins (PG) and nitric oxide (NO) are required in the induction of bone formation by mechanical stimulation. We therefore tested the ability of NO donors, S‐nitroso‐N‐acetyl‐D,L‐penicillamine (SNAP), and S‐nitroso‐glutathione (GSNO) to mimic or augment the osteogenic response of bone to a minimal mechanical stimulus. In rats administered vehicle or the vasodilator hydralazine, stimulation of the 8th caudal vertebra increased bone formation. In animals treated with SNAP or GSNO, there was significant potentiation of this osteogenic response. The bone formation rate in nonloaded vertebrae was unaffected by administration of the NO donors. We also found that while inhibition of either PG or NO production at the time of loading caused a partial suppression of c‐fos mRNA expression in the loaded vertebrae, administration of indomethacin and NG‐monomethyl‐L‐arginine together markedly suppressed c‐fos expression. This suggests that although both PG and NO are required in mechanically induced osteogenesis, they appear to be generated largely independently of each other. Moreover, while exogenous NO potentiates the stimulatory effect of mechanical loading on bone formation, the lack of effect in nonloaded vertebrae suggests that NO is necessary but not sufficient for induction of bone formation.

Nitric oxide synthase expression in bone cells.

We have localized the expression of the three main nitric oxide synthases (eNOS, bNOS, and iNOS) in bone cells of rats and humans using immunohistochemistry. The predominant isoform expressed in normal adult bone was the constitutive isoform, eNOS, mainly in cells of osteoblastic lineage. In adult bone, the osteoblast lineage cells exhibiting eNOS expression were flat bone lining cells and osteocytes, but cuboidal osteoblasts were consistently negative. Expression for bNOS was not detected in any bone cells. iNOS expression was not detected in any cells of osteoblastic lineage in normal adult rat or human bone, but was observed in cuboidal osteoblasts of adult rats with experimental colitis, in which the suppression in bone formation may be cytokine mediated. Osteoclasts in normal rat tissue showed expression for both eNOS and iNOS, but these were patchy. As for cells of the osteoblast lineage, osteoclasts were negative for bNOS. Thus, our findings support evidence, from in vitro studies and from animal experiments, that nitric oxide may play an important role in the physiology of bone.

Role for parathyroid hormone in mechanical responsiveness of rat bone

We investigated the relationship between parathyroid hormone (PTH) and mechanical stimulation in mechanically induced osteogenesis. In normal rats, mechanical stimulation of the eighth caudal vertebra induced an osteogenic response. This was augmented by a single injection of human PTH-(1-34) 30-45 min before loading. No osteogenic response was seen in thyroparathyroidectomized (TPTX) rats; the osteogenic response was restored by a single injection of PTH before stimulation, suggesting that physiological levels of PTH are necessary for the mechanical responsiveness of bone. c- fosexpression was detected only in the osteocytes of those rats that were both mechanically stimulated and given PTH. This suggests that PTH supports mechanically induced osteogenesis by sensitizing either the strain-sensing mechanism itself or early responses of bone to strain-generated signals. The osteogenic response was not augmented by two further daily injections of PTH and was not seen in TPTX rats in which PTH administration was started 3 days after loading. These results reveal a major role for PTH in the mechanical responsiveness of rat bone.

Mechanical loading stimulates bone formation by reactivation of bone lining cells in 13-week-old rats.

The bone formation that occurs in response to mechanical stimulation is generally considered to be a means by which bone adapts to changes in its mechanical environment. We have previously shown that the expression of genes for bone matrix proteins is maximal 72 h after a single 5‐minute episode of loading of tail vertebrae of 13‐week‐old female rats, that the predominant increase in mineralization occurs after 3 days, and that the osteogenic response to mechanical stimulation is not dependent on prior bone resorption. We have now investigated the cellular correlates of this osteogenic response. No proliferation was detected, by pulse or flash labeling, in the trabecular bone surface cells of animals killed 1 h to 10 days after the loading episode. Ultrastructural examination revealed that most of the cells covering the trabecular bone surface of control vertebrae were flat bone lining cells. After mechanical stimulation, the trabecular bone surface cells developed ultrastructural features of osteoblastic differentiation and activity, with acquisition of an increasingly cuboidal shape, rounded nuclei, and abundant rough endoplasmic reticulum. Morphometric analysis of the mean cell area, mean nuclear area, and cell and nuclear height showed that they were all maximal 48 h after loading. By 120 h after loading, the appearances of bone surface cells had reverted to those of control vertebrae. Thus, mechanical loading appears to activate lining cells, with a temporal sequence that correlates with bone matrix production.

Early Effects of Hormone Replacement Therapy on Bone

Estrogen replacement is currently the preferred therapy for postmenopausal osteoporosis, although its mechanism of action remains poorly understood. Its primary action on bone is generally considered to be antiresorptive, but there is evidence in animals to suggest a stimulatory effect on bone formation. We have now attempted to detect a similar effect in humans by administering hormone replacement therapy (estradiol valerate 2 mg/day and dydrogesterone 5 mg/day given in a continuous, combined manner) to ten postmenopausal women. We carried out histomorphometric analyses of transiliac bone biopsies after quadruple tetracycline labeling, which was commenced before and continued during the first 4 weeks of hormone replacement therapy. Biochemical markers of bone turnover suggested that bone resorption decreased, but no significant effects on histomorphometric parameters of bone formation were detected. We conclude that hormone replacement therapy at the dose given does not stimulate bone formation in the iliac crest as assessed by histomorphometry.

An assessment of the prevalence of organic material on bone surfaces

SummaryAlthough an unmineralized layer of organic material has been identified on both bone-forming surfaces and surfaces upon which bone formation has ceased (quiescent surfaces), the proportion of bone surfaces that is covered by unmineralized material has not been quantified. Because the unmineralized layer may play a role in the regulation of bone resorption, we undertook a scanning electron microscopy (SEM) assessment to determine its extent. Specimens of adult human ribs were prepared for undecalcified resin sections and SEM. For SEM, cells were removed and the bone surface was inspected and photographed. The same specimen was then immersed in NaOCl to remove organic material, and inspected again in the SEM. We found that the surface of bone appeared quite different before, compared to after, removal of organic material. Before removal, the entire nonresorptive surface was finely fibrillary. After removal of the organic material we observed a minor component showing the finely nodular surface typical of mineralizing bone, and a major component in which the mineral surface was free of such nodules. In only 3 of 1,200 photographs did we identify areas in which the bone surface was not altered by removal of organic material from the specimen. Analysis of histological sections of the ribs showed that approximately 85% of the bone surface was classifiable by light microscopy as quiescent. These results suggest that not only formative but also quiescent surfaces are covered by a layer of unmineralized organic material.

3-amino-1-hydroxypropylidine-1-bisphosphonate (AHPrBP) suppresses not only the induction of new, but also the persistence of existing bone-forming surfaces in rat cancellous bone

Although bone formation is coupled to resorption in both man and the rat, it is not known whether similar mechanisms are involved in these two species. To investigate this, we have attempted to further characterise the suppression of bone formation by the potent bone resorption inhibitor 3-amino-1-hydroxypropylidine-1-bisphosphonate (AHPrBP) in ovariectomised adult female rats. To see whether AHPrBP affects the activity of preexisting bone forming surfaces, as opposed to preventing the induction of new forming surfaces, three consecutive fluorochrome labels were administered 3, 10, and 17 days prior to sacrifice. The percentage of trabecular surface covered by first and second labels only (arrested surface), second and third labels only (induction surface), and all three labels (persisting surface) were recorded at the secondary spongiosa of the proximal tibial metaphysis. As expected, ovariectomy increased the proportion of induction surfaces, with AHPrBP preventing this increase. However, AHPrBP also reduced the proportion of persisting surfaces, implying inhibition of the activity of preexisting bone-forming surfaces. We conclude that resorption inhibitors such as AHPrBP inhibit bone formation in the rat in a manner which appears inconsistent with the site-specific coupling seen in the adult human skeleton.