Bone sialoprotein (BSP) is a phosphorylated skeletal glycoprotein. Here we describe a new procedure for the purification of BSP involving wide-pore reversed-phase HPLC, and the development of a homologous RIA for human BSP. The immunoassay showed linearity between 3 and 120 μg/L, a lower detection limit of 0.7 μg/L, and a mean recovery rate of 99.4%. Intraassay variation was 7.0% (mean = 10.9 μg/L) and 6.1% (mean = 38.8 μg/L), and interassay variation was 9.2% (mean = 11.1 μg/L) and 9.4% (mean = 39.0 μg/L). No cross-reactivity was detected with osteocalcin, osteonectin, or bone alkaline phosphatase. Preliminary clinical evaluation in healthy controls (n = 90) showed a mean serum BSP concentration of 12.1 ± 5.0 μg/L (±SD). BSP was significantly increased in patients with Paget disease of bone, primary and secondary hyperparathyroidism, and also in subjects with renal failure without skeletal involvement. Impairment of hepatic function did not affect serum BSP concentrations.

Bone sialoprotein (BSP) is a phosphorylated glycoprotein with a M(r) of 70-80 kDa that accounts for approximately 5-10% of the noncollagenous proteins of bone. Due to its relatively restricted distribution to mineralized tissues, BSP may serve as a potential marker of bone metabolism. Employing a recently developed RIA, serum BSP was measured in 133 healthy subjects, aged 20-80 yr, and in patients with primary hyperparathyroidism (pHPT; n = 26), Paget's disease of bone (PD; n = 14), untreated multiple myeloma (MM; n = 32), and breast cancer with bone metastases (BC; n = 19). Results were compared to clinical and laboratory data, including serum total alkaline phosphatase as a marker of bone formation, and the urinary cross-links pyridinoline (PYD) and deoxypyridinoline (DPD) as markers of bone resorption. In healthy adults, serum BSP values ranged between 5.0-21.6 ng/mL (5-95% interval), with a median of 10.5 ng/mL (total group). In healthy females, a linear correlation was found between serum BSP and age (r = 0.51; P < 0.001), with significantly higher values in postmenopausal than in premenopausal women (13.3 +/- 4.8 vs. 9.0 +/- 3.8; P < 0.01). In the healthy group, BSP values did not change with body mass index, lumbar bone mineral density, serum calcium, serum creatinine, or serum total alkaline phosphatase levels. In contrast, a weak, but significant, correlation was observed between serum BSP and the urinary excretion of PYD and DPD. Compared to those in healthy controls, serum BSP levels were significantly higher in patients with pHPT, PD, MM, or BC (P < 0.01 for all groups). These differences remained after analyses were adjusted for age and sex. In pHPT, serum BSP levels were closely correlated to urinary PYD and DPD (r = 0.87 and 0.83, respectively; P < 0.01), whereas in PD, no correlation was observed between any of the bone markers. Serum BSP levels were highest in patients with MM, and there was a significant difference between early and advanced stages of the disease (30.2 +/- 8.0 vs. 64.3 +/- 6.8; P < 0.01). In a subgroup of 15 patients with metastatic BC, iv bisphosphonate treatment resulted in a rapid reduction of serum BSP levels to 40% of the baseline values within 4 days of treatment. In conclusion, BSP appears to be a sensitive marker of bone turnover, and the present data suggest that its serum levels predominantly reflect processes related to bone resorption.

Bone sialoprotein (BS), a protein synthesized by osteoblasts and osteoclasts and highly modified posttranslationally, constitutes a predominant fraction of the noncollagenous organic matrix in human bone. We report an assessment of serum concentrations of BS in patients with malignant bone diseases. In patients with bone metastases (according to scintigraphic criteria), serum BS concentrations were greater than in patients without bone metastases (P &lt;0.05). However, ROC curve analysis revealed that serum BS was inferior to serum bone alkaline phosphatase in discriminating between patients with and without bone metastases. Patients with bone metastases showed a weak correlation between serum BS concentrations and bone formation markers. Only “traditional” markers of bone formation—but not BS—were correlated with urinary deoxypyridinoline (P &lt;0.01). Liver and kidney dysfunction had no significant influence on BS values in these patients (as assessed by analysis of variance; P &gt;0.05). In multiple myeloma patients treated with corticosteroids and bisphosphonates, BS concentrations were lower than in tumor patients without bone metastases (P &lt;0.001), and the correlation between BS concentrations and the number of bisphosphonate courses applied was significant (r = −0.578; P &lt;0.05). In postmenopausal women, serum BS concentrations averaged 142% greater than in premenopausal women. Further studies should be done, therefore, to elucidate whether serum BS is able to predict high bone turnover after menopause.

No abstract

No abstract