Transforming Growth Factor-β1 Gene Polymorphism, Bone Turnover, and Bone Mass in Italian Postmenopausal Women



Transforming growth factor β1 (TGF-β1) is abundant in bone and is an important regulator of the osteoclastic-osteoblastic interaction (coupling). The sequence variation, 713–8delC in the TGF-β1 gene has previously been found to be associated with very low bone mass in osteoporotic women and with increased bone turnover in both osteoporotic and normal women. The possible association of this polymorphism with bone mass and bone turnover has now been investigated in 256 postmenopausal Italian women. A significant association of TGF-β1 with bone mass was detected in the populations. Subjects carrying the sequence variation 713–8delC (Tt) genotype showed a significantly lower bone mineral density (BMD) at the hip than those without sequence variation in the genotype (TT). Individuals carrying the tt genotype have a more severe osteoporosis (P = 0.0001 vs. TT and Tt genotypes). The frequency of the fragility fractures was significantly lower in individuals with TT genotype than in those with the Tt and tt genotypes (χ2 = 21.9; P = 0.006). Furthermore a significant association was found between 713–8delC and bone turnover. The results suggest a strong evidence for an association among the 713–8delC allele of the TGF-β1 gene and the femoral BMD, the prevalence of osteoporotic fractures, and finally a high bone turnover in a sample of Italian postmenopausal women.


Osteoporosis is a common disease characterized by a reduced bone mass and an increased risk of fracture.(1) Bone mass is determined by a variety of genetic and environmental factors. Family and twin studies have shown that genetic factors account for most of the variance in bone mineral density (BMD) in the general population and the inheritance of bone mass is thought to be under polygenic control.(2–4) Polymorphisms of several genes including those encoding for vitamin D receptor, type I collagen, estrogen receptor, and calcitonin receptor have been associated with BMD.(5–11) However, so far, conflicting results have been obtained and genetic susceptibility to osteoporosis is not fully understood.(6–10,12)

The balance between bone resorption and bone formation seems to be regulated by a variety of growth factors and cytokines, among them transforming growth factor β1 (TGF-β1), interleukin 6, and insulin-like growth factors.(13,14) Bone matrix has the highest concentrations of TGF-β1 of all tissues and has been implicated as a possible mediator of coupling between bone resorption and formation in vitro and in vivo studies.(15,16)

Table Table 1.. Clinical Characteristics, Biochemical Markers of Bone Turnover, and BMD Values According to TGF-b1 Genotype in Unrelated Italian Postmenopausal Women
  1. The data are expressed as mean ± SEM.

  2. * Significance of difference among genotype based on one-way ANOVA.

  3. a χ2 Test = 21.09, degree of freedom 2.

Number of subjects208 (81.2%)45 (17.6%)3 (1.2%) 
Age (years)59.9 ± 0.5959.8 ± 1.362.0 ± 2.80.930
Height (cm)160.9 ± 0.3160.8 ± 0.8162.6 ± 3.90.848
Body weight (kg)62.2 ± 0.865.2 ±1.768.4 ± 5.80.220
BMI (kg/m2)24.2 ± 0.2623.9 ± 0.5925.5 ± 3.70.745
Years after menopause8.99 ± 0.338.4 ± 0.68.9 ± 0.80.595
Dietary calcium (mg/day)580 ± 136623 ± 222595 ± 2030.483
Serum calcium (mg/dl)9.1 ± 0.59.0 ± 0.29.3 ± 0.70.735
ALP (U/liter)93.9 ± 1.6100.3 ± 2.7128.6 ± 6.10.012*
GHYL (mmol/mol Cr)0.81 ± 0.031.64 ± 0.112.88 ± 0.400.0001*
FN-BMD (g/cm2)0.803 ± 0.010.695 ± 0.020.471 ± 0.050.0001*
W-BMD (g/cm2)0.662 ± 0.010.578 ± 0.270.309 ± 0.020.0001*
TH-BMD (g/cm2)0.901 ± 0.010.858 ± 0.030.573 ± 0.010.0001*
Fragility fractures20 (9.6%)16 (35.5%)3 (100%)0.006a

TGF-β1 gene consists of seven exons, and its product is secreted as a propeptide of 390 amino acids by a variety of cell types, includng osteoblasts.(16,17) Several polymorphisms in TGF-β1 have been described.(18–20) Yamada et al. described the association between a T-C transition at nucleotide 29 of the region coding the signal sequence (T29-C) and BMD evaluated at lumbar spine in postmenopausal Japanese women.(21) Langdahl et al. found that a one-base deletion in intron 4 (713–8delC) in the TGF-β1 is associated with severe osteoporosis and increased bone turnover in normal and osteoporotic women.(18) They showed that the mutated allele was an independent risk factor for genetic susceptibility to postmenopausal osteoporosis. Being aware of the constant interaction between genetic and environmental factors in bone mass determination, the aim of this study was to investigate the allelic distribution of 713–8delC polymorphism of the TGF-β1 gene in a postmenopausal Italian women population and its relationship with bone mass and bone turnover.



Two hundred and fifty-six postmenopausal women were selected among 1100 unrelated consecutive postmenopausal women who attended the Metabolic Bone Diseases Unit at the University of Verona for osteoporotic risk evaluation. For all women, a detailed medical and nutritional history was obtained. Only white women were included. Individuals with disorders known to affect bone metabolism and bone mass including diabetes mellitus; renal disorders; rheumatoid arthritis; and thyroid, parathyroid and other endocrinological diseases and who had taken drugs such as progesterone, estrogen, glucocorticoids, vitamin D supplementation, and bisphosphonates were excluded from the study. Blood was available for DNA isolation in the 256 subjects. The mean age of the women was 59.9 ± 7.9 years (range, 46–77 years) and the years since menopause was 8.5 ± 4.7 years (range, 1–22 years). All subjects underwent a lateral spine X-ray examination. A subgroup of 39 (15.2%) women had one or more vertebral compression fracture diagnosed by lateral spine radiographs or hip fracture proved with hospital records.(22) An informed consent was obtained from all subjects and the study protocol was approved by the Local Ethic Committee. Clinical characteristics are reported in Table 1.

Measurement of BMD

BMD expressed as grams per centimeter squared was assessed at the hip by dual-energy X-ray absorptiometry (DXA) with Lunar Expert-XL (Lunar, Madison, WI, U.S.A.).

The CV was 1.2% for femoral neck BMD (FN-BMD), 2.2% for Ward's triangle BMD (W-BMD), and 1.0% for total hip BMD (TH-BMD). The subjects were divided in normal, osteopenic, and osteoporotic on the basis of the T-score values of the femoral neck according to the classification of the World Health Organization (WHO).(1)

Bone turnover

Standard routine profile was performed on all patients by DAX 96 (Bayer Corporation, Tarrytown, NY, U.S.A.) autoanalyser. Serum calcium, phosphate, creatinine, and albumin were measured by standard laboratory methods. Serum intact parathyroid hormone (PTH) was measured by an immunoradiometric assay (Intact PTH, Nichols Institute Diagnostics, San Juan Capistrano, CA, U.S.A.). To assess bone turnover, we measured the total alkaline phosphatase (ATP) in the presence of normal liver function on serum samples and the fasting urinary excretion of galactosylhydroxylysine (GHYL) by high performance liquid chromatography (HPLC) as previously described.(23) Results were corrected by urinary creatinine concentration assessed by a standard colorimetric method.

Table Table 2.. Clinical Characteristics, Markers of Bone Turnover, and Distribution of TGF-b1 Genotypes Among Normal, Osteopenic, and Osteoporotic Patients
  1. * Significance of difference among the groups based on one-way ANOVA.

  2. a Cross-tabulation test χ2 = 14.7, d.f. 4, and P = 0.005.

Number of subjects947785 
Age (years)61.4 ± 0.860.5 ± 0.962.5 ± 0.90.833
Height (cm)161.0 ± 0.5160.8 ± 0.5161.0 ± 0.60.948
Body weight (kg)62.6 ±1.263.5 ± 1.462.4 ±1.10.811
BMI (kg/m2)23.8 ± 0.4124.9 ± 0.4123.8 ± 0.490.125
Total ALP (U/l)93.7 ± 2.498.5 ± 2.594.5 ± 2.60.377
GHYL (mmol/mol Cr)0.72 ± 0.050.84 ± 0.0421.40 ± 0.0760.001*
 TT82 (87.2%)67 (87.0%)59 (69.4%) 
 Tt12 (12.8%)10 (13.0%)23 (27.1%) 
 tt3 (3.5%) 


Genomic DNA was isolated from peripheral blood according to standard protocols. Polymerase chain reaction (PCR) primers were designed to amplify a 221-base pair (bp) product including the 713–8delC TGF-β1 gene sequence variation. PCR product was then cut with two different restriction enzymes: Alu I and Bsy I. Homozygotes for the presence of the polymorphism show three fragments (108, 81, and 32 bp), while homozygotes for the normal sequence show two bands (140 and 81 bp). An intermediate pattern with four bands was detectable in heterozygous women. The polymorphism was coded as T-t, in which the uppercase letter signifies the absence and the lowercase letter signifies the presence of the restriction site.

Statistical analysis

All data are expressed as as mean ± SEM. Differences in all continuous clinical variables, BMD, and markers of bone turnover among the three TGF-β1 genotypes were tested using one-way analysis of variance and Scheffe's multiple range test. Mean BMD values were corrected for potential confounding factors such as age, height, weight, and years since menopause using analysis of covariance. The frequency distribution of TGF-β1 genotypes in osteoporotic, osteopenic, and normal subjects was compared using cross-tabulation and standard χ2 test. A P value of <0.05 was considered statistically significant. All statistical analyses were performed using STATGRAPHICS version 5 (Manugistic Inc., Rockville, MD, U.S.A.).


Genotype frequencies in the population under study were 81.2% TT, 17.6% Tt, and 1.2% tt. Proportions among different genotypes were as expected according to Hardy-Weinberg equilibrium.

Clinical characteristics of patients and their relationship with TGF-β 1 genotypes are shown in Table 1. Results indicated that subjects in the three genotypes were well comparable for age, weight, body mass index, years after menopause, and dietary calcium intake. On the basis of FN-BMD measurements and according to WHO criteria 37% of the 256 subjects were normal and 63% had osteopenia or osteoporosis.(1) A statistically significant segregation of TGF-β1 genotypes with FN-BMD, W-BMD, and TH-BMD was detected (Table 1). Subjects heterozygous or homozygous for the 713–8delC sequence variation showed a lower TH-BMD than that of subjects without the sequence variation. The frequency of the fragility fractures was significantly lower in individuals with TT genotype than in those with the Tt and tt genotypes (χ2 = 21.09; P = 0.006; Table 1). Furthermore, patients with Tt and tt showed a significantly higher bone turnover, expressed as levels of serum ALP and urinary GHYL, than TT subjects (Table 1). Patients carrying the tt genotype presented with the lowest BMD values and the highest levels of GHYL and ALP. After stratifying the sample on the basis of BMD T-score at the femoral neck, genotype frequencies among normal, osteopenic, and osteoporotic women were compared (Table 2). The tt genotype (3.5%) was present only in the osteoporotic group. Cross-tabulation testing for the genotype frequencies was statistically significant (x2 = 14.7; P = 0.004). Higher, but not statistically significant ALP values were detected in osteopenic women as compared with normal controls whereas GHYL levels were significantly higher in osteoporotic than normal and osteopenic women (P = 0.001).


Evidences from epidemiological and twin studies clearly ahowed that osteoporosis is a multifactorial disease with a strong genetic component. BMD, the major factor determining bone strength and consequently osteoporotic fracture risk, can be considered a quantitative polygenic trait. Several genes have been shown to be associated with bone mass and fracture risk but conflicting results make genetic susceptibility to osteoporosis a factor that still needs to be clarified.(6–10,12,24–27) Recently, two different studies, one on the white population and the other on the Japanese population, have shown the presence of an association between polymorphisms of TGF-β1 gene and either osteoporosis or bone turnover.(18,21) We have examined the possible association of a sequence variation 713–8delC of TGF-β1 gene with bone mass and bone turnover in postmenopausal women from Northeastern Italy, a region thought to be quite genetically homogeneous. We found that the genotype TT was the most frequently represented in the study population (81.2%) and that the tt genotype was rare (1.2%). The number of heterozygous individuals was greater both in normal and in osteoporotic patients (12.8% and 27.1%, respectively) as compared with other white populations.(18) Furthermore, we found three osteoporotic patients (3.5%) homozygous for the 713–8delC sequence variation whereas Langdahl et al. did not find any homozygous subject.(18)

Our results clearly show that 713–8delC polymorphism is associated to low bone density. Subjects carrying the Tt genotype show a significantly lower TH-BMD than those carrying the TT genotype. Finally, individuals carrying the tt genotype present with a more severe osteoporosis. This association was maintained when the subjects were stratified on the basis of BMD in normal, osteopenic, and osteoporotic patients. With regards to BMD, our data are only partially in agreement with Langdahl et al. who found an association of 713–8delC only in osteoporotic patients with very low bone mass at the lumbar spine whereas in the total group of osteoporotic patients spine BMD was unaffected by the presence of this polymorphism.(18) Our analysis was restricted to Femoral-BMD, a site mainly characterized by cortical bone, whereas genetic effects on bone mass seem to be stronger at sites characterized by higher proportions of trabecular bone.(28–31) A possible explanation for these findings is that after the age of 50 years vertebral osteophytosis, degenerative disc, and apophyseal joint disease with consequent hyperostosis may induce inaccuracies and overestimation of BMD and this could induce errors in BMD phenotyping.(32)

In our study, we confirm that the TGF-β1 gene polymorphism is associated with an increased bone turnover. Postmenopausal women with 713–8delC sequence variation displayed higher bone turnover than women without the 713–8delC sequence. These data are in full agreement with those reported by Langdhal at al. who also found an increased bone turnover in women carrying this TGF-β1 allele.(18) Thus, these results indicate that TGF-β1 polymorphism is one of the possible genetic loci conditioning the bone turnover interindividual heterogeneity.

TGF-β1, stored in large amounts in bone matrix, is produced by osteoblasts and is released and activated during bone resorption by osteoclasts. It is an important regulator of bone metabolism and plays crucial roles in coupling bone resorption and formation.(33,34) In particular, TGF-β1 has been implicated as a mediator of skeletal effects of estrogen.(35) The production of TGF-β1 by human osteoblastic cells is stimulated by 17β-estradiol and TGF-β1 contributes to the estrogen-induced apoptosis of osteoclasts, which results in reduced bone resorption.(36) The 713–8delC mutation is located in a splice junction sequence upstream to exon 5.(18) Thus, at least from a theoretical point of view, it could result in an exon skipping leading to a truncated propeptide and/or absence of active TGF-β1. Yamada et al. found that serum concentration of TGF-β1 was associated, in osteoporotic as well as normal subjects, with a T-C transition at nucleotide 29 in the signal sequence region of the TGF-β1 gene resulting in a Leu-Pro substitution at amino acid position 10.(21) We have not shown whether and how the 713–8delC allele affects the amounts and/or activation of TGF-β1. Therefore, we may only speculate that 713–8delC sequence variation might affect the normal inhibitory effects of TGF-β1 on osteoclastic bone resorption increasing bone turnover.(37) Several studies have shown significant correlation between baseline levels of bone markers and the subsequent rate of bone loss, a high bone turnover being associated with a fast rate of bone loss.(38,39) If so, the effects of the 713–8delC sequence variation in the TGF-β1 gene on bone mass should be mediated through the high bone turnover. The increased bone turnover also can increase per se the risk of osteoporotic fractures by increasing trabecular perforations and thereby decrease the strength of the bone tissue. This could contribute to the higher incidence of fragility fractures in osteoporotic patients with 713–8delC polymorphism. However, the limited number of subjects with osteoporotic fractures limits statistical power to test the hypothesis of a prevalence of a given genotype in fractured subjects.

In conclusion, our study provides strong evidence for an association between the 713–8delC allele of TGF-β1 gene and (1) the Femoral-BMD, (2) the prevalence of fragility fracture, and finally (3) a high bone turnover in a sample of Italian postmenopausal women. Our findings suggest that this polymorphism may be one of the most important genetic determinants of bone turnover and bone mass. Further studies are required to determine whether 713–8delC polymorphism is associated to differences in TGF-β1 serum or bone concentration. Moreover, these findings warrant further longitudinal study in larger populations and in other ethnic groups before becoming useful in prevention and management of osteoporosis.