Relation of PvuII site polymorphism in the COL1A2 gene to the risk of fractures in prepubertal Finnish girls

Miia Suuriniemi1,4,2, Anitta Mahonen3, Vuokko Kovanen4, Markku Alén5 and Sulin Cheng4

1 Department of Cell Biology, University of Jyväskylä
2 LIKES-Foundation for Sport and Health Sciences, Jyväskylä FIN-40014
3 Department of Medical Biochemistry, University of Kuopio, Kuopio FIN-70211, Finland
4 Department of Health Sciences, University of Jyväskylä, Jyväskylä FIN-40014
5 PEURUNKA-Medical Rehabilitation Center, Jyväskylä FIN-41340, Finland


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 DISCLOSURES
 REFERENCES
 
Genetic susceptibility to fractures may be detectable in early childhood. We evaluated the associations between the polymorphic PvuII site of the COL1A2 gene and bone properties assessed by different modalities (dual-energy X-ray absorptiometry; peripheral quantitative computed tomography; gel coupling scanning quantitative ultrasonometry; ultrasound bone sonometry), bone turnover markers, and the occurrence of fractures in 244 prepubertal Finnish girls. Tanner stage and physical characteristics did not differ significantly among girls with different COL1A2 genotypes. The polymorphism was not significantly associated with different bone properties or any of the bone turnover markers when girls at Tanner stage I (prepuberty) and stage II (early puberty) were considered together, but there was a significant association with spine bone mineral content (BMC) and bone mineral density (BMD), as well as with speed of sound (SOS) (P < 0.05), when girls at Tanner stage I were considered separately, as a purpose to avoid the confounding effect that the pubertal growth spurt has on skeletal development. The distribution of fractures was different between the three genotype groups (P = 0.023). The P alleles were over-represented in girls who had been fractured at least once; 88% of them had at least one copy of the P allele (either PP or Pp). Girls with the PP genotype had 4.9 times higher relative risk for fractures than girls with the pp genotype (95% CI, 1.4 to 17.4; P = 0.015). No significant difference was found between fractured and nonfractured girls in anthropometric measurements, physical activity, or bone mass. However, BMD of the spine and SOS at the radius and tibia were significantly lower in the fractured girls. We conclude that the COL1A2 polymorphism is associated with nonosteoporotic fractures in prepubertal girls independently of bone density.

children; genetics; osteoporosis; type I collagen


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 DISCLOSURES
 REFERENCES
 
ALTHOUGH MUCH IS KNOWN regarding the incidence and pattern of fractures during growth, the pathogenesis of the increased fracture rate during childhood has not yet been defined. Low bone mass is an important component of the risk of fracture, but other abnormalities arise in the skeleton that contribute to skeletal fragility. In addition to the inability of the mineralization process to keep pace with the growth of the long bones, an imperfection in the alignment of collagen fibers with the principal directions of loading may play a role. Furthermore, various nonskeletal factors, such as the liability to fall, contribute to fracture risk. There is, therefore, a distinction to be made between diagnosis of osteoporosis and assessment of fracture risk.

Bone mass and strength are under strong genetic control. Most efforts toward understanding the genetics of bone density have focused on association studies of candidate genes known to be involved in bone metabolism in adult populations (1). During the past years, a few researchers have examined the influence of candidate genes, reported to be associated with adult bone mass, on the phenotypic variability of skeletal development in children. Defining the genetic and environmental factors responsible for variations in bone properties and fracture rate during skeletal growth should help identify children at risk for fractures also later in life (12).

Besides bone mass, the strength and mechanical properties of bone also depend on the architecture and molecular structure of inorganic and organic components (23). Type I collagen is the major structural protein in bone and consists of a heterotrimeric complex of two {alpha}1-polypeptides and one {alpha}2-polypeptide. Polymorphism in the Sp1 binding site of the collagen type I {alpha}1 gene (COL1A1) has been found to be related to decreased bone mass and osteoporotic fractures in women (15, 41). More recently, the COL1A1 gene alleles have been found to be associated with the normal variations in the apparent density of cancellous bone in the axial skeleton of prepubertal girls (34). Virtually all mutations that result in osteogenesis imperfecta (OI) affect the genes that encode the chains of type I procollagen, especially the COL1A1 gene. Nevertheless, OI patients with mutations in the collagen type I {alpha}2 gene (COL1A2) gene have been characterized (6, 32, 38, 39, 42). The well-described strain of mice with a nonlethal recessively inherited mutation in the COL1A2 gene (oim) that results in phenotypic and biochemical features that simulate moderate-to-severe human OI is living evidence of the importance of the collagen type I {alpha}2 polypeptide to the bone structure and strength (10, 25, 31, 33). In this study, we examined the relationship between a single nucleotide polymorphism (A->C) in codon 392 in the COL1A2 gene (11) and various bone properties, bone turnover markers, and the retrospective occurrence of fractures in early and prepubertal girls.


    METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 DISCLOSURES
 REFERENCES
 
Subjects.
The study subjects were 258 (244 for the genomic analysis) healthy early and prepubertal Finnish girls, aged 10–12 yr, who were recruited for an intervention study to evaluate the effects of calcium, vitamin D, and milk product supplementation on bone accrual (the CALEX study). To be eligible for the study, the participants had to have no history of serious medical conditions and no history of medication known to affect bone metabolism. For determination of sexual development by a public health nurse, the Tanner grading system was used (40). Only those girls at Tanner stage I–II (prepuberty and early puberty) were included in the study. Level of physical activity was assessed by a self-reported questionnaire including items on the frequency, duration, and types of exercise done during the subject’s leisure time (19). The data of physical activity is provided in terms of hours per week. Fracture history was clarified by a parents-reported questionnaire including items on the site, time, and cause of each fracture and was confirmed by hospital medical records. Girls with a fracture history of less than 1 yr were excluded from the study. Fractures caused by severe trauma (e.g., serious accidents and crushed fingers/toes) were excluded from this analysis. The investigational protocol was approved by the ethical committee of the University of Jyväskylä, the Central Hospital of Central Finland, and the Finnish National Agency of Medicines. An informed consent was obtained from all subjects and their parents prior to the assessments. The results presented in this report are from the baseline assessments.

Body composition assessments.
Body height was measured with a fixed-scale measuring device. Weight was determined (±0.5 kg) using a calibrated scale. The results of height and weight were then used to determine body mass index (BMI), expressed as weight in kilograms divided by the square of the height in meters (in kg/m2). The fat and lean tissue mass of the total body were analyzed from a total body dual-energy X-ray absorptiometry (DXA) scan.

Bone property assessments.
Bone mineral content (BMC, g) and areal bone mineral density (aBMD, g/cm2) of the total body, left proximal femur (total femur and femoral neck), and lumbar spine (L2–L4) were measured using DXA (Prodigy; GE Lunar, Madison, WI). The percentage coefficient of variation (%CV) for repeated measurements ranged from 0.6 to 1.2 for BMC and from 0.86 to 1.3 for aBMD at the different bone sites. Cross-sectional area (CSA, mm2), cortical thickness (CTh, mm), and volumetric BMD (vBMD, mg/cm3) were measured using peripheral quantitative computed tomography (pQCT) (model XCT 2000; Stratec Medizintechnik, Pforzheim, Germany). We scanned the left distal radius (4% of the total length of the forearm medial to the reference line) and tibia shaft (60% of the lower leg length between the tuberositas tibia and the medial malleolus). Data were then analyzed using BonAlyse software (BonAlyse Oy, Jyväskylä, Finland) (8). The %CV was <3% for CSA and <1% for vBMD. Broadband ultrasound attenuation (BUA, dB/MHz) of the left calcaneus was measured using a gel coupling scanning quantitative ultrasonometer (QUS-2; Quidel, Santa Clara, CA) (7). The speed of sound (SOS, m/s) at the distal third of radius (medial surface) and the midshaft of tibia (anteromedial surface) were determined by a mobile ultrasound bone sonometer (Omnisense; Sunlight Technologies, Rehovot, Israel) (2).

Biochemical markers of bone turnover.
After an overnight fast, blood was drawn in the morning between 7–9 AM for determinations of bone formation [osteocalcin, bone-specific alkaline phosphatase (BAP), and amino-terminal propeptide of type I procollagen (P1NP) in a subgroup, n = 102] and bone resorption [bone-specific tartrate-resistant acid phosphatase (TRAP 5b) in a subgroup, n = 213] markers. The serum intact osteocalcin level was measured by a competitive immunoassay (NovoCalcin; Metra Biosystems, Mountain View, CA) (4); the intra-assay and interassay %CV were 2.4% and 0.62%, respectively. The serum BAP activity was measured by an immunoassay (Alkphase-B; Metra Biosystems, Mountain View, CA) (14); the intra- and interassay %CV were 2.1% and 15.3%, respectively. The serum P1NP level was measured by a competitive radioimmunoassay (Orion Diagnostica, Espoo, Finland) (28); the intra- and interassay %CV were 3.6% and 1.85%, respectively. The serum TRAP 5b activity was measured by a specific immunoassay (16, 17); the intra-assay and interassay %CV were 2.7% and 24.5%, respectively.

Analysis of collagen type I {alpha}2 gene polymorphism.
Genomic DNA was isolated from EDTA-stabilized blood, using a blood kit (QIAamp; Qiagen, Hilden, Germany). The polymorphic PvuII site of the COL1A2 gene was detected by polymerase chain reaction (PCR) followed by enzymatic digestion (11). The upstream primer was 5'-GGGATATAAGGATACACTAGAGG-3', and the downstream primer was 5'-GAAATATCGGCCCCGCTGGAA-3'. The reaction mixture of 20 µl contained 50–200 ng genomic DNA, 50 mM KCl, 10 mM Tris-HCl (pH 8.8), 1.5 mM MgCl2, 0.2 mM dNTPs, 10 pmol of each primer, and 1.5 U of Taq DNA polymerase (Fermentas, Vilnius, Lithuania). The reactions were performed in a DNA thermocycler (T3 combi-block; Biometra, Göttingen, Germany) with a cycling protocol of 94°C, 59°C, and 72°C for 1 min each, for 30 cycles. Prior to the first cycle, initial denaturation was performed at 94°C for 5 min, and the last cycle was followed by an extension step of 7 min at 72°C. The PCR products were digested with 10 U of PvuII restriction enzyme and a buffer, containing 10 mM Tris-HCl (pH 7.5), 50 mM NaCl, and 10 mM MgCl2, by incubation for 4 h at 37°C. The digested products were analyzed on an 1.5% agarose gel. The PCR products containing p alleles were cleaved by PvuII, resulting in bands of 541 bp and 239 bp compared with the uncleaved P alleles band of 771 bp.

Statistical analysis.
Statistical analyses were carried out using the Statistical Package (SPSS) version 9.0 for Windows. Differences in bone properties (CSA, CTh, BMC, aBMD, vBMD, BUA, and SOS) and serum bone turnover markers among the three genotypes as well as between the fracture and nonfracture groups were tested using analysis of covariance, controlling for the effects of BMI and Tanner stage. The likelihood ratio was used to test for the genotype distribution in girls with and without fractures. Odds ratios [with 95% confidence intervals (CI)] were calculated by multivariate logistic regression analysis after stepwise adjustment for potential confounding variables such as Tanner stage, BMI, aBMD of the total body, and physical activity to estimate the relative risk of fracture by COL1A2 genotypes. Under a possible dominant inheritance model, analysis was subsequently performed on the combined PP and Pp genotype groups. To control potential influence of bone density on the SOS, we used site-matched volumetric BMD (vBMD of radius in SOS at radius and vBMD of tibia in SOS at tibia) in addition to Tanner stage and BMI as a covariate in the covariance analysis of SOS in fracture vs. nonfracture girls. All significant tests were two-sided. A P value of less than 0.05 was considered statistically significant. Genotype distribution was studied by the Hardy-Weinberg equilibrium (29).


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 DISCLOSURES
 REFERENCES
 
Physical characteristics.
Table 1 shows physical characteristics according to the COL1A2 genotype. The overall prevalence of the genotypes in this study was 15.6% PP, 52.9% Pp, and 31.6% pp. The genotype distribution was found to be in Hardy-Weinberg equilibrium, suggesting that the subjects represented a homogenous genetic background. There were no significant differences in developmental status or physical activity among the girls in the different COL1A2 genotype groups. The mean values for weight, height, and BMI, as well as the values for total body fat and lean mass were similar in the different COL1A2 genotype groups.


View this table:
[in this window]
[in a new window]
 
Table 1. Physical characteristics in relation to the COL1A2 genotype

 
Bone measurements and bone turnover.
We found no significant effect of the COL1A2 gene polymorphism on BMC or aBMD of the total body, total femur, femoral neck, or lumbar spine (L2–L4) (Table 2). No significant differences were found between the different genotype groups in CSA or vBMD in the radius or tibia, or in the CTh of the tibia, BUA of the calcaneus, or SOS at the radius and tibia. Serum concentrations of biochemical markers for bone turnover were independent of the COL1A2 genotypes (Table 3). No clear relationship was found between the P allele and aforesaid parameters by pooling of the two genotype groups (PP and Pp) (data not shown). However, when Tanner stage I girls were analyzed separately, individuals with the PP genotype had significantly lower BMC (P = 0.010) and aBMD (P = 0.013) of the L2–L4 (Fig. 1), as well as SOS at the tibia (P = 0.036) and a trend toward lower SOS at the radius (P = 0.092), compared with individuals with the pp genotype (Fig. 2). No significant effect of the COL1A2 gene polymorphism on bone properties was found at Tanner stage II (data not shown).


View this table:
[in this window]
[in a new window]
 
Table 2. Bone measurements in relation to the COL1A2 genotype

 

View this table:
[in this window]
[in a new window]
 
Table 3. Bone turnover in relation to the COL1A2 genotype

 


View larger version (16K):
[in this window]
[in a new window]
 
Fig. 1. Bone mineral content (BMC, top) and areal bone mineral density (aBMD, bottom) of the lumbar spine (L2–L4) in relation to the COL1A2 genotype at Tanner stage I. Data are adjusted means ± SE. P values were determined by analysis of covariance, with body mass index (BMI) as a covariate.

 


View larger version (8K):
[in this window]
[in a new window]
 
Fig. 2. Speed of sound (SOS) at tibia in relation to the COL1A2 genotype at Tanner stage I. Data are adjusted means ± SE. P values were determined by analysis of covariance, with BMI as a covariate.

 
Fracture data.
Among our study sample, there were 46 girls (18%) who had sustained at least one fracture. One girl had sustained three fractures over a 2-yr period, and eight girls had had two fractures. The age when the fracture occurred ranged from newborn to 11 yr. Four of the fractures had happened at birth, ten occurred at the ages of 1–5 yr, thirteen at the ages of 6–8 yr, and nineteen at the ages of 9–11 yr. The most frequent fracture site was arm (31 cases). There were 11 collarbone, 7 finger, and 6 toe fractures, and 1 skull fracture. All the fractures, except those that happened at birth and one due to a knock against the bookshelf, were caused by falling down (e.g., from chair, sofa, or bed; or during walking, running, or bicycling). In the analyses we only included those girls (n = 37) who had had a fracture resulting from minimal to moderate trauma (by convention, the equivalent of a fall from standing height or less).

There were no significant differences in physical characteristics, physical activity, or developmental status between the girls with fractures and their counterparts (Table 4). No significant differences were found between the fracture and nonfracture groups in BMC or aBMD of the total body, total femur, or femoral neck, CSA or vBMD in the radius or tibia, CTh of the tibia, or BUA of the calcaneus. The biochemical markers of bone turnover were similar between girls with and without fracture. However, aBMD of the L2–L4 was significantly lower in the fracture group compared with the nonfracture group (P = 0.04). Also, the fracture group had significantly lower SOS values at the radius (P = 0.029) as well as at the tibia (P = 0.036) compared with the nonfracture group after adjusting for site-matched vBMD and Tanner stage as well as BMI (Fig. 3). When all fractures (n = 46) were included in the analysis, the differences were even stronger (BMC of the L2–L4, P = 0.071; aBMD of the L2–L4, P = 0.028; SOS at the radius, P = 0.003; SOS at the tibia, P = 0.039) (data not shown).


View this table:
[in this window]
[in a new window]
 
Table 4. Characteristics in relation to fracture history

 


View larger version (17K):
[in this window]
[in a new window]
 
Fig. 3. SOS in relation to fracture history, for radius (top) and tibia (bottom). Data are adjusted means ± SE. P values were determined by analysis of covariance, Tanner stage, BMI, and volumetric bone mineral density (vBMD) of radius and tibia, respectively, as covariates.

 
Fracture risk according to the COL1A2 genotype.
The distribution of fractures was significantly different in the genotype groups (P = 0.023) (Table 5). Pooling of the PP and Pp genotype groups under a dominant inheritance model showed a clear relationship between the P allele and the occurrence of fractures (P = 0.007). Of the girls who had sustained at least one fracture before the laboratory assessments, 88% had at least one copy of the P allele, in comparison to 65% of the girls who had not had fractures during their lives. The odds ratios showed that the relative risk of fracture among girls with the P allele (either PP or Pp genotype) was 4.0 times higher than among the girls with the pp genotype (Table 5). The crude risk of fracture was 3.8 for the Pp genotype group and 4.9 for the PP genotype group in relation to the pp genotype group. The risk did not change essentially after adjustments for potential confounding factors such as Tanner stage, BMI, and aBMD of the total body in the multivariate regression analysis (Table 5). Stepwise adjustments for the aforementioned parameters yielded similar results as adjustment for all the five factors together (data not shown). When the results were additionally adjusted for physical activity, the risks of fracture still remained similar.


View this table:
[in this window]
[in a new window]
 
Table 5. Fracture risk related to the pp genotype of the COL1A2

 

    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 DISCLOSURES
 REFERENCES
 
We were able to show a significant association between the COL1A2 polymorphism and the retrospective occurrence of fractures in early and prepubertal girls. To the best of our knowledge, this is the first study reporting the genotypic effect of PvuII site polymorphism in the COL1A2 gene on bone properties and the occurrence of fractures.

Differences among COL1A2 gene alleles may be more difficult to demonstrate during puberty, when large increases in skeletal size, bone mass, and bone density occur over a brief period of time. In fact, we found no effect on bone properties in relation to the genotype when girls at Tanner stage I (prepuberty) and II (early puberty) were considered together, but there was an association of the COL1A2 polymorphism with spine BMC and BMD, as well as with SOS when girls at Tanner stage I were considered separately. This finding is consistent with the hypothesis that COL1A2 genotype contributes to bone strength in part by an effect on bone density and in part by an effect on bone quality. As in our work, earlier studies had shown that the genetic effect associated with polymorphism in the type I collagen gene appeared stronger at the spine (15, 20).

We found no effect on the bone properties studied except for the aBMD of the lumbar spine (L2–L4) and the SOS at the radius and tibia in relation to fracture history. Theoretically, the SOS is a measure of BMD and elasticity combined. Mehta et al. (26) found that the organic matrix exerts a profound influence on bone elasticity and that the subtle changes in the organic matrix have effect on ultrasound velocity in vitro. Our previous study in females with diseases related to collagen mutations (Ehlers-Danlos syndrome and systemic sclerosis) showed that collagen abnormalities may impact on bone mass measurements differently depending on skeletal site, modality of the assessment, and the source and nature of collagen defects. Ultrasound assessment was able to detect the differences between patients and matched controls (9). We found that those girls who had sustained a fracture previously had significantly lower SOS values than girls who had not. After adjusting for site-matched vBMD, the significant differences remained. Our results indicate that differences in SOS might be derived from the differences in bone elasticity between the fracture and nonfracture groups.

Girls with the PP genotype had 4.9 times higher relative risk of fracture compared with girls with the pp genotype. In addition, after adjusting for Tanner stage, BMI, and aBMD of the total body, no change in fracture risk was found between the COL1A2 genotype groups. Thus we may assume that COL1A2 genotype predisposes to fracture mostly by an effect on bone quality, such as bone structure or matrix composition, rather than by an effect on bone quantity. The increased risk of fracture could not be a consequence of higher physical activity, since the physical activity did not differ in the COL1A2 genotype groups, and adjustment for physical activity did not affect the association between the COL1A2 polymorphism and the risk of fracture. Even though the fractures were not osteoporotic, the fact of their occurrence was due to minimal to moderate trauma may be related to the structure and strength of bones. The involvement of BMD in bone strength is well established, although the correlation between these two parameters is only partial (23). Conversely, the existence of an association between type I collagen structure and bone strength has been well documented in OI and other clinical syndromes attributable to mutations of type I collagen genes encoding for both {alpha}1- and {alpha}2-chains (3537). Although most OI patients have subnormal BMD, some of them have decreased bone strength but normal BMD (30). This provides evidence that type I collagen structure independently of BMD may be related to changes in the mechanical strength of bone.

The association between COL1A1 gene Sp1 polymorphism and the risk of fractures has recently been found to be independent of BMD (3). Garnero and coworkers have shown more recently that different age-related forms of the COOH-terminal cross-linking telopeptide of type I collagen are associated with increased fracture risk independently of BMD (13). Because the type I collagen molecule consists of two {alpha}1-chains and one {alpha}2-chain, mutations in the {alpha}2-chain are considered to have a smaller potential for deleterious consequences than {alpha}1 mutations. Our results support these findings and show that the association between the COL1A2 polymorphism and the fracture risk is independent of BMD. We conclude that the main effect of the COL1A2 genotype on the fracture risk is mediated by the effect of the genotype on the elastic properties of bone assessed by the quantitative ultrasound, which is believed to provide information about bone quality. Nevertheless, bone density at least at the spine may also be affected by the COL1A2 genotype during the fast pubertal growth period.

The mechanisms by which the different COL1A2 alleles affect fracture risk are not yet known. The sequence change detected is a CpA to CpC transversion in exon 25 which does not affect the encoded proline residue at position 392 of the {alpha}2(I) chain (22). Constantinou et al. (11) observed the same polymorphism in the amplified products of cDNA from a patient with OI and several patients with osteoporosis (11). Our results raise the possibility that this polymorphism may be in linkage disequilibrium with a causal mutation in the same gene or in genes nearby. Another possibility could be a direct influence on the gene regulation. For example, synonymous single nucleotide polymorphisms located in coding regions (cSNPs), although seemingly translationally silent, can have a profound influence on splicing. In fact, cSNPs can disrupt (or eventually create) exonic splicing enhancers and silencers; create new splice sites or strengthen cryptic ones; alter pre-mRNA secondary structures important for exon definition; and, conceivably, modify the pausing architecture of a gene, provoking changes in RNA PolII processivity, which might in turn affect splice site choice (5). Approximately 15% of mutations that cause genetic disease affect pre-mRNA splicing (21). An important fact is that Nicholls et al. (29) found an association between the absence of this PvuII polymorphic site (denoting the P allele) in exon 25 and a splice site mutation (G->A) causing deletion of exon 21 from the pro-{alpha}2(I) chain of type I collagen in a patient with very mild OI. The mutant pre-mRNA was alternatively spliced, yielding both full-length and deleted transcripts. Therefore, further investigation with other polymorphisms within the COL1A2 gene and nearby markers may be useful. If individuals, due to their altered collagen production, have reduced trabecular thickness, then they would be at higher risk of trabecular perforations. This would reduce bone strength proportionally more than the accompanying loss of bone density. This could be an explanation for the finding that carriage of the P allele predicts fractures independently of bone density.

In conclusion, our study demonstrated a significant association between PvuII site polymorphism in the COL1A2 gene and fractures in early and prepubertal girls. This information may contribute to the identification of a subset of the population of normal girls that may be at risk of developing fractures later in life and may ultimately be of value in the planning of early preventive strategies for osteoporosis.


    DISCLOSURES
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 DISCLOSURES
 REFERENCES
 
This study was supported by the Academy of Finland and Finnish Ministry of Education.


    ACKNOWLEDGMENTS
 
We thank the Universities of Jyväskylä and Turku for assessing the biomarkers, and we thank Toritutkain-Medical Center in Jyväskylä for the bone density assessments. We also thank all the investigators, data managers, other support staff, and, most importantly, the children and their parents participating in the study.


    FOOTNOTES
 
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).

Address for reprint requests and other correspondence: S. Cheng, Dept. of Health Sciences, Univ. of Jyväskylä, PO Box 35, FIN-40014 Jyväskylä, Finland (E-mail: cheng{at}sport.jyu.fi).

10.1152/physiolgenomics.00070.2003.


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 DISCLOSURES
 REFERENCES
 

  1. Audi L, Garcia Ramirez M, and Carrascosa A. Genetic determinants of bone mass. Horm Res 51: 105–123,
  2. Barkmann R, Kantorovich E, Singal C, Hans D, Genant HK, Heller M, and Gluer CC. A new method for quantitative ultrasound measurements at multiple skeletal sites: first results of precision and fracture discrimination. J Clin Densitom 3: 1–7, 2000, 1999[ISI][Medline]
  3. Bernad M, Martinez ME, Escalona M, Gonzalez ML, Gonzalez C, Garces MV, Del Campo MT, Martin Mola E, Madero R, and Carreno L. Polymorphism in the type I collagen (COLIA1) gene and risk of fractures in postmenopausal women. Bone 30: 223–228, 2002.[ISI][Medline]
  4. Blumsohn A, Hannon R, and Eastell R. Apparent instability of osteocalcin in serum as measured with different commercially available immunoassays. Clin Chem 41: 318–319, 1995.[Free Full Text]
  5. Cáceres JF and Kornblihtt AR. Alternative splicing: multiple control mechanisms and involvement in human disease. Trends Genet 18: 186–193, 2002.[ISI][Medline]
  6. Campbell BG, Wootton JA, Macleod JN, and Minor RR. Canine COL1A2 mutation resulting in C-terminal truncation of pro-{alpha}2(I) and severe osteogenesis imperfecta. J Bone Miner Res 16: 1147–1153, 2001.[ISI][Medline]
  7. Cheng S, Fan B, Wang L, Fuerst T, Lian M, Njeh C, He Y, Kern M, Lappin M, Tylavsky F, Casal D, Harris S, and Genant HK. Factors affecting broadband ultrasound attenuation results of the calcaneus using a gel-coupled quantitative ultrasound scanning system. Osteoporos Int 10: 495–504, 1999.[ISI][Medline]
  8. Cheng S, Kröger H, Junkala T, Koistinen A, Kuronen P, Renko R, Tylavsky F, and Suominen H. Physical activity and bone mass in prepubertal girls. J Bone Miner Res 15, Suppl 1: S333, 2000.
  9. Cheng S, Tylavsky FA, Orwoll ES, Rho JY, and Carbone LD. The role of collagen abnormalities in ultrasound and densitometry assessment: in vivo evidence. Calcif Tissue Int 64: 470–476, 1999.[ISI][Medline]
  10. Chipman SD, Sweet HO, McBride DJ, Davisson MT, Marks SC, Shuldiner AR, Wenstrup RJ, Rowe DW, and Shapiro JR. Defective pro-{alpha}2(I) collagen synthesis in a recessive mutation in mice: a model of human osteogenesis imperfecta. Proc Natl Acad Sci USA 90: 1701–1705, 1993.[Abstract]
  11. Constantinou CD, Spotila LD, Zhuang J, Sereda L, Hanning C, and Prockop DJ. PvuII polymorphism at the COL1A2 locus. Nucleic Acids Res 18: 5577, 1990.[ISI][Medline]
  12. Ferrari S, Rizzoli R, Slosman D, and Bonjour JP. Familial resemblance for bone mineral mass is expressed before puberty. J Clin Endocrinol Metab 83: 358–361, 1998.[Abstract/Free Full Text]
  13. Garnero P, Cloos P, Sornay-Rendu E, Qvist P, and Delmas PD. Type I collagen racemization and isomerization and the risk of fracture in postmenopausal women: the OFELY prospective study. J Bone Miner Res 17: 826–833, 2002.[ISI][Medline]
  14. Gomez B, Ardakani S, Ju J, Jenkins D, Cerelli MJ, Daniloff GY, and Kung VT. Monoclonal antibody assay for measuring bone-specific alkaline phosphatase activity in serum. Clin Chem 41: 1560–1566, 1995.[Abstract/Free Full Text]
  15. Grant SF, Reid DM, Blake G, Herd R, Fogelman I, and Ralston SH. Reduced bone density and osteoporosis associated with a polymorphic Sp1 binding site in the collagen type I {alpha}1 gene. Nat Genet 14: 203–205, 1996.[ISI][Medline]
  16. Halleen J, Hentunen TA, Hellman J, and Väänänen HK. Tartrate-resistant acid phosphatase from human bone: purification and development of an immunoassay. J Bone Miner Res 11: 1444–1452, 1996.[ISI][Medline]
  17. Halleen JM, Hentunen TA, Karp M, Kakonen SM, Pettersson K, and Väänänen HK. Characterization of serum tartrate-resistant acid phosphatase and development of a direct two-site immunoassay. J Bone Miner Res 13: 683–687, 1998.[ISI][Medline]
  18. Hans D, Srivastav SK, Singal C, Barkmann R, Njeh CF, Kantorovich E, Gluer CC, and Genant HK. Does combining the results from multiple bone sites measured by a new quantitative ultrasound device improve discrimination of hip fracture? J Bone Miner Res 14: 644–651, 1999.[ISI][Medline]
  19. Hickman M, Roberts C, and Gaspar de Matos M. Exercise and leisure-time activities. In: Health and Health Behaviour Among Young People, edited by Currie C, Hurrelmann K, Settertotulte W, Smith R, and Todd J. Health Promotion and Investment for Health, World Health Organization Regional Office for Europe. Copenhagen: 2000, p. 73–82.
  20. Keen RW, Woodford-Richens KL, Grant SFA, Ralston SH, Lanchbury JS, and Spector TD. Association of polymorphism at the type I collagen (COL1A1) locus with reduced bone mineral density, increased fracture risk, and increased collagen turnover. Arthritis Rheum 42: 185–190, 1999.
  21. Krawczak M, Reiss J, and Cooper DN. The mutational spectrum of single base-pair substitutions in mRNA splice junctions of human genes: causes and consequences. Hum Genet 90: 41–54, 1992.[ISI][Medline]
  22. Kuivaniemi H, Tromp G, Chu ML, and Prockop DJ. Structure of a full-length cDNA clone for the prepro {alpha}2(I) chain of human type I procollagen. Comparison with the chicken gene confirms unusual patterns of gene conservation. Biochem J 252: 633–640, 1988.[ISI][Medline]
  23. Landis WJ. The strength of a calcified tissue depends in part on the molecular structure and organization of its constituent mineral crystals in their organic matrix. Bone 16: 533–544, 1995.[ISI][Medline]
  24. Martin RB and Ishida J. The relative effects of collagen fiber orientation, porosity, density, and mineralization on bone strength. J Biomech 22: 419–426, 1989.[ISI][Medline]
  25. McBride DJ, Shapiro JR, and Dunn MG. Bone geometry and strength measurements in aging mice with the oim mutation. Calcif Tissue Int 62: 172–176, 1998.[ISI][Medline]
  26. Mehta SS, Oz OK, and Antich PP. Bone elasticity and ultrasound velocity are affected by subtle changes in the organic matrix. J Bone Miner Res 13: 114–121, 1998.[ISI][Medline]
  27. Melkko J, Kauppila S, Niemi S, Risteli L, Haukipuro K, Jukkola A, and Risteli J. Immunoassay for intact amino-terminal propeptide of human type I procollagen. Clin Chem 42: 947–954, 1996.[Abstract/Free Full Text]
  28. Muller RF and Young ID. Emery’s Elements of Medical Genetics. Edinburgh, Scotland: Churchill Livingston, 1995, p. 93–104.
  29. Nicholls AC, Oliver J, McCarron S, Winter GB, and Pope M. Splice site mutation causing deletion of exon 21 sequences from the pro {alpha}2(I) chain of type I collagen in a patient with severe dentinogenesis imperfecta but very mild osteogenesis imperfecta. Hum Mutat 7: 219–227, 1996.[ISI][Medline]
  30. Paterson CR and Mole PA. Bone density in osteogenesis imperfecta may well be normal. Postgrad Med J 70: 104–107, 1994.[Abstract]
  31. Phillips CL, Bradley DA, Schlotzhauer CL, Bergfeld M, Libreros-Minotta C, Gawenis LR, Morris JS, Clarke LL, and Hillman LS. Oim mice exhibit altered femur and incisor mineral composition and decreased bone mineral density. Bone 27: 219–226, 2000.[ISI][Medline]
  32. Rose NJ, Mackay K, Byers PH, and Dalgleish R. A Gly238Ser substitution in the {alpha}2 chain of type I collagen results in osteogenesis imperfecta type III. Hum Genet 95: 215–218, 1995.[ISI][Medline]
  33. Saban J, Zussman MA, Havey R, Patwardhan AG, Schneider GB, and King D. Heterozygous oim mice exhibit a mild form of osteogenesis imperfecta. Bone 19: 575–579, 1996.[ISI][Medline]
  34. Sainz J, Van Tornout JM, Sayre J, Kaufman F, and Gilsanz V. Association of collagen type 1 {alpha}1 gene polymorphism with bone density in early childhood. J Clin Endocrinol Metab 84: 853–855, 1999.[Abstract/Free Full Text]
  35. Shapiro J. An osteopenic nonfracture syndrome with features of mild osteogenesis imperfecta associated with the substitution of a cysteine at the triple helix position 43 in the pro {alpha}1 (I) chain of type I collagen. J Clin Invest 89: 567–573, 1992.[ISI][Medline]
  36. Shapiro JR, Burn VE, Chipman SD, Velis KP, and Bansal M. Osteoporosis and familial idiopathic scoliosis: association with an abnormal {alpha}2(I) collagen. Connect Tissue Res 21: 117–124, 1989.[Medline]
  37. Spotila LD, Colige A, Sereda L, Constantinou-Deltas CD, Whyte MP, Riggs BL, Shaker JL, Spector TD, Hume E, and Olsen N. Mutation analysis of coding sequences for type I procollagen in individuals with low bone density. J Bone Miner Res 9: 923–932, 1994.[ISI][Medline]
  38. Superti-Furga A, Pistone F, Romano C, and Steinmann B. Clinical variability of osteogenesis imperfecta linked to COL1A2 and associated with a structural defect in the type I collagen molecule. J Med Genet 26: 358–362, 1989.[Abstract]
  39. Superti-Furga A, Raghunath M, Pistone FM, Romano C, and Steinmann B. An intronic deletion leading to skipping of exon 21 of COL1A2 in a boy with mild osteogenesis imperfecta. Connect Tissue Res 29: 31–40, 1993.[ISI][Medline]
  40. Tanner J. Physical growth and development. In: Textbook of Pediatrics, edited by Forfar JO and Arnell CC. Edinburgh, Scotland: Churchill Livingstone, 1978, p. 249–303.
  41. Uitterlinden AG, Burger H, Huang Q, Yue F, McGuigan FE, Grant SF, Hofman A, van Leeuwen JP, Pols HA, and Ralston SH. Relation of alleles of the collagen type I {alpha}1 gene to bone density and the risk of osteoporotic fractures in postmenopausal women. N Engl J Med 338: 1016–1021, 1998.[Abstract/Free Full Text]
  42. Zhuang J, Tromp G, Kuivaniemi H, Nakayasu K, and Prockop DJ. Deletion of 19 base pairs in intron 13 of the gene for the pro {alpha}2(I) chain of type-I procollagen (COL1A2) causes exon skipping in a proband with type-I osteogenesis imperfecta. Hum Genet 91: 210–216, 1993.[ISI][Medline]