Skeletal Diseases Genome Research Center, Kyungpook National University Hospital, Daegu, Korea
Department of Pathology and Regenerative Medicine, School of Dentistry, Kyungpook National University, Daegu, Korea
Address reprint requests to: Eui Kyun Park, PhD Department of Pathology and Regenerative Medicine, School of Dentistry, Kyungpook National University, 188-1 Samduk 2-ga, Jung-gu, Daegu 700-412, Republic of Korea
The authors state that they have no conflicts of interest.
The genetic effects of FLT3 polymorphisms on BMD and fracture risk in postmenopausal women were studied. We found that FLT3+13348C>T polymorphism and haplotype 2 were significantly associated with low BMD and high risk of fracture.
Introduction: FMS-related tyrosine kinase 3 (FLT3) has been shown to play a critical role in the development of myelolymphoid progenitors and in the development of osteoclasts, but any possible genetic effect of FLT3 on bone metabolism has not been studied.
Materials and Methods: To study a possible genetic effect of FLT3, we directly sequenced the FLT3 gene in 24 Korean individuals and identified 23 sequence variants. Seven polymorphisms were selected and genotyped in Korean postmenopausal women (n = 946).
Results: We found that FLT3+13348C>T was associated with low BMD at the lumbar spine (p = 0.04) and femoral neck (p = 0.04). Haplotype analysis revealed that FLT3-ht2 (TTCTT) containing the rare allele in the +13348 position also showed significant association with low BMD in the lumbar spine (p = 0.04) and femoral neck (p = 0.05). Consistent with these results, the FLT3+13348C>T polymorphism and FLT3-ht2 were also significantly associated with high risk of fracture in the vertebrae (OR = 1.44–1.58; p = 0.03–0.04 and OR = 1.45–1.59; p = 0.02–0.03, respectively) and in any sites (OR = 1.34–1.81; p = 0.02–0.03 and OR = 1.34–1.81; p = 0.02–0.03, respectively).
Conclusions: These results suggest that FLT3 polymorphisms play a role in determination of BMD and subsequent fractures in postmenopausal women.
Osteoporosis is a systemic bone disease characterized by low BMD, low bone strength, microarchitectural deterioration of bone tissue, and a consequent increase in fragility. BMD is frequently used as a proxy measure and accounts for ∼70% of bone strength.(1) Although multiple risk factors influence BMD and thus the pathogenesis of osteoporosis, genetic factors are mainly implicated and account for ∼50–85% of the variance in BMD based on twin and family studies.(2–8) However, the genes that contribute to pathogenesis of osteoporosis are still largely unknown. Recently, we have shown that genes involved in osteoclast differentiation and function are potential candidate genes influencing BMD and osteoporosis.(9,10) Through continuous efforts to identify and further understand the genetic factors involved in osteoporosis, it would be possible to identify subjects at risk for osteoporosis and to target preventive treatment.(11)
The gene for human FMS-related tyrosine kinase 3 (FLT3; MM_004119) is located on chromosome 13q12 (67 kb). FLT3, a member of the class III receptor tyrosine kinases (RTKs), is expressed on the cell surface of hematopoietic progenitors(12) and plays an important role in the maintenance, proliferation, and differentiation of these cells. FLT3 binding to FLT3 ligand (FL) results in dimerization of the receptors, autophosphorylation, and the subsequent phosphorylation of cytoplasmic substrates that are involved in signaling pathways regulating the proliferation and differentiation of immature hematopoietic cells.(13,14) In addition, FLT3-null mice are born healthy with normal peripheral blood components, but have reduced numbers of early B-cell precursors in bone marrow. Bone marrow primitive cells from these mice have a reduced ability to reconstitute B-cell, T-cell, and myeloid lineages when transplanted into irradiated mice,(15) indicating that FLT3 is involved in the development of lymphoid and myeloid cells. In vivo studies(16–19) have shown that FL administration enhances the production of mononuclear phagocytes that share common precursors with osteoclasts. Interestingly, FL synergistically or additively enhances proliferation of progenitor cells when acting with stem cell factor or other cytokines.(13)
FLT3 also functions as a regulator of osteoclast differentiation and function. FLT3-positive macrophage precursors can differentiate sequentially into osteoclasts, dendritic cells, and microglia,(20) implying possible involvement of FLT3 in the development of osteoclasts. In fact, it has been shown that FL can induce osteoclast differentiation by substitution for macrophage-colony stimulating factor (M-CSF), which plays a critical role in the development and function of osteoclasts.(21) These results suggest that FL/FLT3 play a role in development of osteoclast and thus contribute to bone turnover. Despite the important role of FLT3 as a bone regulator, a possible genetic effect of FLT3 on BMD or bone metabolism has not been studied to date. Here, we present the genetic polymorphisms found in FLT3 and describe a positive association of FLT3 polymorphism with low BMD and high-fracture risk in a Korean osteoporosis cohort.
MATERIALS AND METHODS
Subjects and measures
The study population was comprised of 946 postmenopausal women of Korean ethnicity who visited the Asan Medical Center (AMC) in Seoul, Korea. Menopause was defined as the absence of menstruation for at least 6 mo and was confirmed by measurement of serum follicle-stimulating hormone. Women with premature menopause (before 40 yr of age) were excluded. Women who had taken drugs that might affect bone metabolism for >6 mo or within the previous 12 mo were also excluded. Subjects were excluded if they had suffered from any disease that might affect bone metabolism. Women who had suffered a stroke or who exhibited dementia were also excluded because of concerns related to their limited physical activity. Women were also excluded if they had osteophyte formation above the fourth grade of the Nathan classification(22) and/or severe facet joint osteoarthritis in the lumbar spine as determined by conventional spine radiographs. This study was approved by the AMC Ethics Review Committee, and written informed consent was obtained from all subjects.
Areal BMD (g/cm2) values at the lumbar spine (L2–L4) and femoral neck were measured using DXA (Expert XL; Lunar, Madison, WI, USA) in 637 women. In the remaining 309 women, BMD was measured using Hologic equipment (QDR 4500-A; Hologic, Waltham, MA, USA). The accuracies of the Lunar and Hologic equipment, given by CVs were 0.82% and 0.85% for the lumbar spine and 1.12% and 1.20% for the femoral neck, respectively. These values were obtained by scanning 17 volunteers who were not part of the study; each volunteer underwent five scans on the same day, getting on and off the table between examinations. To derive cross-calibration equations between the two systems, BMD values were measured using both machines in 109 healthy Korean women (age, 55 ± 11 yr; range, 31–75 yr), and cross-calibration equations were calculated as follows(23):
The assessment of vertebral fractures was made in accordance with the recommendations of the Working Group on Vertebral Fractures.(24) Lateral radiographs of the thoracic and lumbar spines (T4–L4) were performed using standard X-ray equipment. Radiographs were assessed at AMC by expert radiologists blinded to this study. The radiographic assessment was performed using a visual semiquantitative technique, whereby a vertebral fracture was defined quantitatively as a loss of 15% of the vertebra (compared with data on intact vertebrae from previous examinations) or a decrease of 4 mm in any of the measured vertebral heights (anterior, middle, or posterior) in subjects without a previous history of major trauma such as traffic accidents. In addition, we assessed the occurrence of nonvertebral fracture (wrist, hip, forearm, humerus, rib, and pelvis) after 50 yr old by self-report. Fractures clearly caused by major trauma such as motor vehicle accidents or a fall from more than standing height were excluded.
Sequencing of human FLT3
We sequenced all exons, including exon-intron boundaries, and the promoter region (∼1.5 kb) to detect SNPs in 24 Korean DNA samples using the ABI PRISM 3730 DNA analyzer (Applied Biosystems, Foster City, CA, USA). The primer sets for the amplification and sequencing analysis were designed based on GenBank sequences (Ref. Seq. of FLT3 mRNA: MM_004119 and contig: NT_024524). Sequence variants were verified by automated sequencing chromatograms. SNPs were detected by multiple alignments of sequences using the Phred/Phrap/Consed package(25,26) and polyphred.(27)
Genotyping with fluorescence polarization detection
For genotyping of polymorphic sites, amplification primers and probes were designed for TaqMan.(28) Primer Express (Applied Biosystems) was used to design both the PCR primers and the MGB TaqMan probes. One allelic probe was labeled with the 6-carboxyfluorescein (FAM) dye and the other with the fluorescent 2′-chloro-7′-phenyl-1,4-dichloro-6-carboxyfluorescein (VIC) dye. PCR reactions contained TaqMan Universal Master mix without uracil-N-glycosylase (Applied Biosystems), with PCR primer concentrations of 900 nM and TaqMan MGB probe concentrations of 200 nM. Reactions were performed in a 384-well format in a total volume of 5 μl using 20 ng of genomic DNA. The plates were placed in a thermal cycler (PE 9700; Applied Biosystems) and heated at 50°C for 2 min and 95°C for 10 min followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. The TaqMan assay plates were transferred to a Prism 7900HT instrument (Applied Biosystems) in which the fluorescence intensity in each well was measured. Fluorescence data files from each plate were analyzed using automated software (SDS 2.2). The average call rate of seven SNPs genotyping was 98.8%.
The χ2 tests were used to determine whether individual variants were in equilibrium at each locus in the population (Hardy-Weinberg equilibrium). We examined Lewontin's D' (|D'|) and the linkage disequilibrium (LD) coefficient r2 between all pairs of biallelic loci.(29) Haploview version 3.2 (Whitehead Institute for Biomedical Research, Cambridge, MA, USA) was used to examine the structure of the LD block.(30) This program uses two-marker expectation maximization (EM) to estimate the maximum-likelihood values of the four gamete frequencies from which the D' and LOD values are derived. Haplotypes of each individual were inferred using the algorithm PHASE developed by Stephens et al.,(31) which uses a Bayesian approach incorporating a priori expectations of haplotypic structure based on population genetics and coalescence theory. Phase probabilities of all polymorphic sites for haplotypes were calculated for each individual using the PHASE software. Individuals with phase probabilities of <97% were excluded from the analysis. The genetic effects of inferred haplotypes were analyzed in the same way as was applied to analysis of polymorphisms. Multiple regression analyses were performed for BMD, controlling for age (continuous variable), years since menopause (YSM; continuous variable), weight, and height as covariates. The genotype distributions of FLT3 polymorphisms between fracture and nonfracture groups were analyzed with logistic regression models controlling for age, YSM, weight, and height as covariates. Genotypes of major homozygotes, heterozygotes, and minor homozygotes were given codes of 0, 1, and 2; 0, 1, and 1; and 0, 0, and 1 in the codominant, dominant, and recessive models, respectively. Haplotype associations were analyzed using the algorithm developed by Schaid et al. (Haplo.Score and Haplo.GLM).(32)
Through direct sequencing of all exons and their boundaries of the FLT3 gene including the −1500 bp of the 5′ flanking region in 24 independent Korean individuals, 23 polymorphisms (3 in the 5′ flanking region, 17 in introns, and 3 in exons) were identified. Genotype distributions of all loci were in Hardy-Weinberg equilibrium (p > 0.05). Among the polymorphisms identified, seven (−686A≥G, ±13348C≥T, ±20457T≥C, ±34926C≥T, ±34965T≥A, ±52586T≥C, and ±55242G≥A) were selected for larger scale genotyping (n = 946) by criteria such as one among polymorphisms in tight LDs and allele frequencies >20% (Fig. 1). Two SNPs, +34926C>T and +34965T>A, were not absolute LDs in sequencing stage, but both SNPs were in almost absolute LDs after larger-scale genotyping. Therefore, only +34965T>A polymorphism was selected for further analysis (Fig. 1). Haplotypes in the FLT3 gene were constructed by using PHASE software.(31) The clinical profiles of postmenopausal women are shown in Table 1. The mean age of the participants was 58.9 ± 7.5 yr (range, 44–84 yr), and the mean YSM was 9.6 ± 7.9 yr (range, 1–34 yr). Multiple linear regression analysis showed that age and YSM correlated negatively with BMD at both the lumbar spine (β = −0.005, p = 0.004 and β = −0.003, p = 0.01, respectively) and femoral neck (β = −0.003, p = 0.01 and β = −0.004, p < 0.001, respectively), and weight correlated positively with BMD at both sites (β = 0.005, p < 0.001 and β = 0.003, p < 0.001, respectively). Height correlated positively with lumbar BMD (β = 0.003, p < 0.02). BMD measured by Lunar equipment (0.887 ± 0.145 and 0.734 ± 0.105 g/cm2 at the lumbar spine and femoral neck, respectively) were significantly higher than values given by the Hologic equipment (0.785 ± 0.092 and 0.608 ± 0.067 g/cm2, respectively; p < 0.001 for both). However, the values were not significantly different after the application of cross-calibration equations (data not shown). Therefore, age, YSM, weight, and height were used as covariates in multiple regression analyses for association with BMD at the lumbar spine and femoral neck.
Table Table 1.. Clinical Profiles in Postmenopausal Korean Women (n = 946)
Using multiple regression analysis, the association of FLT3 polymorphisms with adjusted BMD values of lumbar spine and femoral neck controlling for age, YSM, weight, and height was analyzed. Among polymorphisms, the FLT3+13348C>T polymorphism showed significant association with lumbar spine (p = 0.04) and femoral neck BMD (p = 0.04) in a recessive model (Table 2). The level of lumbar spine BMD in individuals bearing a minor homozygous genotype (0.87 ± 0.17 g/cm2 in T/T of +13348C≥T) was lower than that in individuals bearing common homozygous and heterozygous genotypes (0.89 ± 0.17 and 0.89 ± 0.16 g/cm2, respectively; p = 0.04; Table 2). Similarly, individuals bearing minor homozygous genotypes had lower femoral neck BMD values (0.71 ± 0.13 g/cm2) than those bearing heterozygous and common homozygous genotypes (0.73 ± 0.12 and 0.73 ± 0.12 g/cm2, respectively; Table 2). In further haplotype analysis, FLT3-ht2 was also significantly associated with low BMD at the lumbar spine (p = 0.04) and the femoral neck (P = 0.05; Table 2).
Table Table 2.. Regression Analyses for Age, Weight, Height, and Year Since Menopause Adjusted BMD of Lumbar Spine and Femoral Neck With FLT3 Polymorphisms Among the Korean Population
The genetic effects of FLT3 polymorphisms on the risk of fracture in both vertebral and other sites were also analyzed. The rare allele FLT3+13348C>T was higher in subjects with vertebral fracture (freq. = 0.357) than in those without fracture (freq. = 0.290; OR = 1.44–1.58, p = 0.03–0.04). FLT3-ht2 bearers also had a high frequency of vertebral fracture (freq. = 0.357) compared with those who did not have the FLT3-ht2 allele (freq. = 0.290; OR = 1.45–1.59, p = 0.03–0.02; Table 3). When the risk for any fracture in either vertebral or other bone sites was considered, individuals bearing the rare allele FLT3+13348C>T showed higher fracture frequencies (0.344) than those bearing common alleles or those who were heterozygotes for FLT3+13348C>T (0.286). A higher probability for the occurrence of any fracture was also observed in individuals bearing haplotype FLT3-ht2 (OR = 1.34–1.81, p = 0.03–0.02; Table 3).
Table Table 3.. Logistic Analysis of FLT3 Polymorphisms While Controlling for Age, Weight, Height, Year Since Menopause as Covariates Between Fracture Patients and Others
Bone remodeling is tightly controlled by a balance between bone resorption by osteoclasts and bone formation by osteoblasts. An imbalance in the remodeling process, in which resorption usually exceeds formation, results in accelerated bone loss.(33–35) Therefore, balanced control of osteoclast formation is one of the popular treatments for osteoporosis and an understanding of molecular or genetic contributions of critical genes to the formation of osteoclasts is pivotal to the development of preventive and therapeutic interventions for osteoporosis. Although osteoporosis is believed to be a multifactorial disease involving genetic, environmental, and nutritional factors, studies have shown that genetic effects contribute primarily to the development of osteoporosis. Osteoporosis is a polygenic disease.(36) However, numerous studies have shown that a single gene or single polymorphism is substantially associated with risk of osteoporosis or BMD.(11) These results suggest that identification of the single gene or polymorphism associated with osteoporosis, followed by combinatorial analysis of important polymorphisms, may dramatically contribute to an understanding of the pathogenesis underlying osteoporosis, and the data may be used to predict patient susceptibility to osteoporosis.
In an effort to identify and characterize polymorphisms associated with osteoporosis, we sequenced the promoter region and the exons and their boundaries of the FLT3 gene. FLT3 is well known as an RTK expressed in a majority of acute leukemias.(37) Activating mutations of FLT3, including an internal tandem duplication in the juxtamembrane region, and an Asp-to-Tyr substitution in the activation loop, have been detected in these malignancies.(37–40) Several reports have shown that FLT3 is also implicated in bone metabolism. FL can substitute for M-CSF,(21) and FL in combination with the receptor activator of NF-κB synergistically stimulates osteoclast formation.(21) These results clearly show that FLT3 functions in osteoclast formation and thus in bone metabolism. In addition, we found a genetic effect of FLT3 on BMD and fracture. Through direct sequencing and association analyses, we showed that FLT3+13348C>T and FLT3-ht2 are significantly associated with low BMD values and risk for osteoporosis. It seems that FLT3-ht2 is primarily influenced by the FLT3+ 13348C>T polymorphism because FLT3-ht2 was tagged by the FLT3+13348C>T minor allele. Multiple regression analyses revealed that FLT3+13348C>T and FLT3-ht2 were significantly associated with low BMD at the lumbar spine and femoral neck. These results strongly suggest that the FLT3+13348C>T polymorphism affects BMD at both sites, and thus, that the FLT3+13348C>T polymorphism and FLT3-ht2 are genetic factors influencing BMD and the risk for osteoporosis.
Because osteoporosis is characterized by low BMD and a consequent increase in fractures, we further analyzed genetic effects of the FLT3+13348C>T polymorphism and FLT3-ht2 on fractures. Logistic analyses showed that the FLT3+13348C>T polymorphism and FLT3-ht2 were associated with vertebral and any fractures (Table 3). Despite the relatively low heritability of fracture (25% and 35%),(41,42) we noted that the rare alleles FLT3+13348C>T polymorphism and FLT3-ht2 were higher in subjects with fractures than in those without. Our results showing consistent and significant associations of the FLT3+13348C>T polymorphism and FLT3-ht2 with low BMD, high risk of osteoporosis, and high frequency of fractures strongly support the reliability and significance of the results.
The +13348C>T polymorphism may be a marker rather than a direct contributor to the genetic functions because the +13348C>T polymorphism is located in the intron and does not cause an amino acid change. It is possible that the effects of other genetic variations linked to this polymorphism may have functional significance. We also cannot exclude a possibility that this intronic polymorphism has a role on its genetic function through the change of alternative splicing.(43)
It can be argued that Bonferroni correction should be applied to the p values obtained. If Bonferroni correction were strictly adopted, associated p values could not retain the significances. However, although there is a chance of type 1 error because of multiple comparisons, when considering the facts (1) that the comparisons were not totally independent of each other because of tight LDs among SNPs/haplotype and related phenotypes and (2) that consistent positive signals at the same site (FLT3+13348C>T) with related phenotypes (BMD in various bone sites (femoral neck and vertebrae), risk of osteoporosis, and risk of radiological fracture (Table 3) were detected, the significance of associations might be acceptable.
In summary, we identified 23 polymorphisms of FLT3 in a Korean population and showed that one polymorphism (FLT3+13348C>T) and one haplotype (FLT3-ht2) were significantly associated with low BMD values and risk for osteoporosis and fractures. To our knowledge, this is the first report suggesting a genetic role for FLT3 in bone metabolism, and our observations suggest that the FLT3+ 13348C>T polymorphism of the FLT3 gene represents a possible genetic risk factor for osteoporosis and subsequent fractures in postmenopausal women.
This work was supported by a grant from the Korea Health 21 R&D Project, Ministry of Health & Welfare, Republic of Korea (Project A010252) and intramural grants from the Korea National Institute of Health, Korea Center for Disease Control, Republic of Korea (Project 347-6111-211).