We examined the association of bone mineral density (BMD) with a polymorphism in the gene encoding the vitamin D receptor (VDR) that causes a change in the predicted protein sequence. The polymorphism results from a C-to-T transition and creates an initiation codon (ATG) three codons proximal to a downstream start site. The polymorphism can be defined by a restriction fragment length polymorphism (RFLP) using the restriction endonuclease FokI. The presence of a FokI site, designated f, allows protein translation to initiate from the first ATG. The allele lacking the site (designated F), initiates from a second ATG site. Thus, translation products from these alleles are predicted to differ by three amino acids with the f variant elongated. In a group of 100 postmenopausal Mexican-American Caucasian women, subjects with the ff genotype (15% of the study population) had a 12.8% lower BMD at the lumbar spine than FF subjects (37% of the population) (p = 0.01). Heterozygote (Ff) subjects (48% of the population) had an intermediate BMD. This association between BMD and genotype was not apparent at the femoral neck or forearm. Over a 2-year follow-up period, a decrease in BMD at the femoral neck was greater in ff compared with FF subjects (-4.7% vs. -0.5%, p = 0.005). This trend was not apparent at the lumbar spine or forearm. There were no differences between genotype groups in measurements of 25-hydroxyvitamin D (25(OH)D), calcitriol, parathyroid hormone (PTH), osteocalcin, or urinary pyridinolines. We conclude that the FokI polymorphism of the VDR gene correlates significantly with decreased BMD at the lumbar spine and with an increased rate of bone loss at the hip in ff subjects. We emphasize that these initial data should be interpreted with caution but that the utility of this polymorphism as a genetic marker to determine BMD and osteoporosis risk warrants further study in larger populations with subjects of diverse ethnic backgrounds.

No abstract

We investigated the associations of vitamin C, calcium and protein intakes with bone mass at the femoral neck and lumbar spine in postmenopausal Mexican American women. Bone mass was measured by dual-energy X-ray absorptiometry (DXA) and expressed as areal (BMD, g/cm2) and volumetric (bone mineral apparent density or BMAD, g/cm3) bone mineral density. Diet was assessed using a modified version of the National Cancer Institute Food Questionnaire, which was administered by trained bilingual interviewers familiar with Mexican dietary practices. Data gathered from 125 subjects were analyzed using multiple linear regression analysis with age, body mass index (BMI), acculturation, years of estrogen use, physical activity, total energy intake, and the nutrient of interest as independent variables. Neither calcium nor calcium/protein ratio was associated with bone mineral density. There was evidence of a positive association between dietary vitamin C intake and femoral neck BMD (beta = 0.0002 g/cm2 per mg/day, SE = 0.00006, p < 0.05), but vitamin C was not associated with lumbar spine bone mass. Further investigation of the role of vitamin C in skeletal health is warranted.