Activity of the osteocalcin promoter in skeletal sites of transgenic mice and during osteoblast differentiation in bone marrow-derived stromal cell cultures: effects of age and sex.

The bone-specific osteocalcin gene is a well established marker of osteoblast activity. We have studied osteocalcin transcription in transgenic mice carrying rat osteocalcin promoter-chloramphenicol acetyltransferase (CAT) reporter constructs. Transgenic lines carrying each of the 1.7-, 1.1-, 0.72-, or 0.35-kilobase promoter constructs expressed the reporter gene in a tissue-specific manner. However, each of these constructs was sensitive to site of integration effects, reflected by a high frequency of nonexpressing transgenic lines. High expression of the 1.7-kilobase promoter in osseous tissues was accompanied by low ectopic expression in the brain. Analysis of CAT expression in femurs, calvariae, and lumbar vertebrae of this line indicated considerable variability in promoter activity among individual transgenic animals. Analysis of the variance in CAT activity demonstrated a linkage between promoter activities in these distant skeletal sites. Promoter activity was inversely correlated with age, and females exhibited severalfold higher activity than age-matched males. Bone marrow stromal cells from these animals, cultured under conditions that support osteoblast differentiation, exhibited the expected postproliferative onset of osteocalcin promoter activity, as assessed by CAT assay. The ex vivo CAT activity was not dependent on the sex or the age of the donor transgenic mouse. Taken together, our results are consistent with the hypothesis that a common, probably humoral, factor(s) regulates osteocalcin transcription in distant skeletal sites. We suggest that the abundance of this factor(s) is different between males and females and among individual mice at a given time point, and that ex vivo culturing of osteoblasts reduces the variation in osteocalcin promoter activity by eliminating the physiological contribution of this factor.

Site selectivity of osteoblast gene expression response to thyroid hormone localized by in situ hybridization.

We have previously reported that thyroid-stimulating hormone (TSH)-suppressive doses of L-thyroxine (L-T4) decrease femoral, but not vertebral, bone mineral density (BMD) in rats. L-T4-induced decreases in BMD were associated with increased expression of genes, reflecting osteoblast activity in mRNA extracted from whole femurs but not from vertebrae. To document that this skeletal selectivity reflected altered osteoblast activity, we studied gene expression by in situ hybridization in 8-wk-old rats treated with L-T4 (20 microg x 100 g body wt(-1) x day(-1)) for 4 wk. TSH-suppressive doses of L-T4 were associated with decreased femoral (0.299 +/- 0.005 vs. 0.273 +/- 0.005 g/cm2, P < 0.01), but not vertebral (0.222 +/- 0.004 vs. 0.218 +/- 0.003 g/cm2), BMD. In situ hybridization documented that L-T4 administration for 4 wk increased expression of osteocalcin and alkaline phosphatase mRNA in femoral, but not vertebral, osteoblasts. This study demonstrates a differential gene expression response of vertebral and femoral osteoblasts to L-T4. This altered degree of gene expression markers of osteoblast activity documented by in situ hybridization may in part explain the apparent clinical differences in the effect of L-T4 on femoral and vertebral BMD.

Effects of the vitamin D3 analog 1 alpha, 25-dihydroxyvitamin D3-3 beta-bromoacetate on rat osteosarcoma cells: comparison with 1 alpha, 25-dihydroxyvitamin D3.

The actions of the hormonal form of vitamin D, 1 alpha, 25-dihydroxyvitamin D3 [1 alpha, 25-(OH)2 D3], are mediated by both genomic and nongenomic mechanisms. Several vitamin D synthetic analogs have been developed in order to identify and characterize the site(s) of action of 1 alpha, 25-(OH)2D3 in many cell types including osteoblastic cells. We have compared the effects of 1 alpha, 25-(OH)2D3 and a novel 1 alpha, 25-(OH)2D3 bromoester analog (1,25-(OH)2-BE) that covalently binds to vitamin D receptors. Rat osteosarcoma cells that possess (ROS 17/2.8) or lack (ROS 24/1) the classic intracellular vitamin D receptor were studied to investigate genomic and nongenomic actions. In ROS 17/2.8 cells plated at low density, the two vitamin D compounds (1 x 10(-8) M) caused increased cell proliferation, as assessed by DNA synthesis and total cell counts. Northern blot analysis revealed that the mitogenic effect of both agents was accompanied by an increase in steady-state osteocalcin mRNA levels, but neither agent altered alkaline phosphatase mRNA levels in ROS 17/2.8 cells. ROS 17/2.8 cells responded to 1,25-(OH)2-BE but not the natural ligand with a significant increase in osteocalcin secretion after 72, 96, 120, and 144 hr of treatment. Treatment of ROS 17/2.8 cells with the bromoester analog also resulted in a significant decrease in alkaline phosphatase-specific activity. To compare the nongenomic effects of 1 alpha, 25-(OH)2D3 and 1,25-(OH)2-BE intracellular calcium was measured in ROS 24/1 cells loaded with the fluorescent calcium indicator Quin 2. At 2 x 10(-8) M, both 1 alpha,25-(OH)2D3 and 1, 25-(OH)2-BE increased intracellular calcium within 5 min. Both the genomic and nongenomic actions of 1,25-(OH)2-BE are similar to those of 1 alpha,25-(OH)2D3, and since 1,25-(OH)2-BE has more potent effects on osteoblast function than the naturally occurring ligand due to more stable binding, this novel vitamin D analog may be useful in elucidating the structure and function of cellular vitamin D receptors.

Improved hepatocyte culture system for studying the regulation of glycogen synthase and the effects of diabetes.

When 3-4-week-old rats (young rats) are used as a source of hepatocytes, primary culture cells express the adult, differentiated, liver-specific isoform of glycogen synthase. Synthase enzyme protein levels are relatively stable over a 3 day culture period in young but not in adult (> 150 g rat) hepatocyte cultures. Corresponding synthase enzyme activity and mRNA levels decrease over time in culture in adult but not in young hepatocyte cultures. Young rat hepatocytes also have the ability to proliferate in chemically defined medium in the absence of added mitogens. A diabetes-induced increase in total synthase activity has been demonstrated by our lab and others, using cultured hepatocytes, liver homogenates, and perfused livers. In the present study, utilizing synthase-specific antibody and primary cultures of cells from young normal and alloxan diabetic rats, we found that greater total synthase activity in the diabetic cells was associated with higher levels of enzyme protein. Immuneprecipitation of 35S methionine-labeled freshly plated cells demonstrates an increase in the rate of protein synthesis in diabetic as compared with normal cells. Synthase mRNA levels are correspondingly increased in the diabetic relative to normal cells. Chronic exposure of young, normal hepatocytes to increasing levels of glucose induces a dose-dependent increase in total synthase activity, total synthase protein, and synthase message levels. By comparison, cells from diabetic animals do not respond by any of these measures to increased glucose concentrations. We conclude that this defined primary culture system represents a useful model for investigating the regulation of hepatic glycogen synthase and the defects which occur as a result of diabetes.