SEARCH

SEARCH BY CITATION

Keywords:

  • low-density lipoprotein receptor-related protein 5 gene;
  • peak BMD;
  • genetic association;
  • single nucleotide polymorphisms

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

The role of the LRP5 gene in rare BMD-related traits has recently been shown. We tested whether variation in this gene might play a role in normal variation in peak BMD. Association between SNPs in LRP5 and hip and spine BMD was measured in 1301 premenopausal women. Only a small proportion of the BMD variation was attributable to LRP5 in our sample.

Introduction: Mutations in the low-density lipoprotein receptor-related protein 5 (LRP5) gene have been implicated as the cause of multiple distinct BMD-related rare Mendelian phenotypes. We sought to examine whether the LRP5 gene contributes to the observed variation in peak BMD in the normal population.

Materials and Methods: We genotyped 12 single nucleotide polymorphisms (SNPs) in LRP5 using allele-specific PCR and mass spectrometry methods. Linkage disequilibrium between the genotyped LRP5 SNPs was measured. We tested for association between these SNPs and both hip and spine BMD (adjusted for age and body weight) in 1301 healthy premenopausal women who took part in a sibling pair study aimed at identifying the genes underlying peak bone mass. Our study used both population-based (ANOVA) and family-based (quantitative transmission disequilibrium test) association methodology.

Results and Conclusions: The linkage disequilibrium pattern and haplotype block structure within the LRP5 gene were consistent with that observed in other studies. Although significant evidence of association was found between LRP5 SNPs and both hip and spine BMD, only a small proportion of the total variation in these phenotypes was accounted for. The genotyped SNPs accounted for ∼0.8% of the variation in femoral neck BMD and 1.1% of the variation in spine BMD. Results from our sample suggest that natural variation in and around LRP5 is not a major contributor to the observed variability in peak BMD at either the femoral neck or lumbar spine in white women.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

LOW BMD is a major risk factor for spine and proximal femur fractures.(1,2) Peak BMD is a major determinant of BMD and bone strength in later life and is highly heritable.(3) Therefore, peak BMD is a trait that lends itself to studies designed to identify the genes underlying its normal variation. Variation in peak BMD is likely to be under complex genetic and environmental regulation.

Recently, different mutations in the low-density lipoprotein receptor-related protein 5 (LRP5) gene were reported to result in two different single gene disorders with opposite phenotypes: high BMD and low BMD. A missense mutation in LRP5 that leads to a gain in function results in an autosomal dominant, high BMD phenotype (HBM, Z score > +3.0) that has no clinical sequellae.(4,5) A second gain-of-function mutation in LRP5 has been found in a family with a distinct high BMD phenotype.(6) Further screening of the LRP5 gene has identified mutations that may result in other rare bone and BMD-related phenotypes.(7,8)

In contrast, a variety of loss-of-function mutations in LRP5 result in the autosomal recessive disorder osteoporosis pseudoglioma syndrome (OPPG), which is characterized by low BMD.(9) Importantly, in the loss-of-function mutations, there is a clear gene dosage effect, with heterozygous mutation carriers having BMD measures intermediate between those of the affected and normal homozygotes.(10) These results suggest that LRP5 may contribute to the variation in BMD in the normal healthy population.

LRP5 is a cell surface protein and functions as a co-receptor for Wnt proteins. It is expressed in osteoblasts, among numerous other tissues, and can transduce Wnt signaling in vitro through the canonical pathway.(5,9) Furthermore, Lrp5 knockout mice also developed a low bone mass phenotype.(11) These observations indicate that LRP5 may be important for the attainment of peak bone mass during growth.

We have previously observed linkage of peak BMD in premenopausal women to the human chromosome 11q region, where the LRP5 gene is located.(12) Therefore, to examine whether the LRP5 gene contributes to the variation in BMD in the normal population, we genotyped 12 single nucleotide polymorphisms (SNPs) in LRP5 and tested for their association with hip and spine BMD in 1301 premenopausal healthy women who took part in sibling pair study aimed at identifying the genes underlying peak bone mass. Our study used both population- and family-based association methodology.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Subjects and measurement of BMD

A sample of white premenopausal sister pairs was recruited through advertisement at Indiana University as part of a sibling pair study to identify genes underlying peak bone mass. A total of 1301 women in 588 families were ascertained. The subjects ranged in age from 20 to 50 years. A blood sample for DNA extraction was obtained from each study participant. In addition, 265 parents provided a blood sample for DNA extraction for use in family-based studies of association, but did not complete any phenotypic assessments. All studies were performed on the General Clinical Research Center at Indiana University Medical School, and all subjects gave written, informed consent before participation. The study was approved by the Indiana University-Purdue University Indianapolis IRB (8502-23). Before enrolling in the study, a detailed medical history was obtained from prospective subjects. Women who had conditions known to affect BMD or cause artifactual readings of BMD, such as spinal deformity, diseases, or use of drugs known to affect BMD, intake of high doses of calcium or vitamin D, irregular menses, pregnancy, lactation, or a history of pregnancy or lactation within 3 months before the study, were excluded from study participation. Women taking oral contraceptives were not excluded from the study.

BMD was measured by DXA (DPXL; Lunar, Madison, WI, USA) at lumbar vertebrae L2-L4 and at the femoral neck. Subjects were measured on either of two instruments, which were cross-calibrated weekly over 73 weeks using a step-wedge phantom with BMD ranging from 0.5 to 2.8 g/cm2. Linear regression analysis of the phantom data for the two machines resulted in an intercept of 0.003 g/cm2 and a slope of 0.99, with an r2 value of 0.998. All subjects were measured on the same instrument as their sister(s), usually at the same visit. Image analysis was performed using Lunar software 4.6/4.7. CV in our laboratory was 1.0% for femoral neck BMD and 0.52% for lumbar spine BMD. Height and weight were measured using a Harpenden Stadiometer and a Scale-Tronix weighing scale, respectively, which were regularly calibrated throughout the study.

Genotyping

SNPs located within the LRP5 gene were selected from NCBI's LocusLink (http://www.ncbi.nlm.nih.gov/LocusLink/) and recent freezes of the Human Genome Browser Gateway at the University of California at Santa Cruz (http://genome.ucsc.edu/cgi-bin/hgGateway). The SNPs were validated by restriction fragment length polymorphism or direct sequencing and genotyped by either fluorescent allele-specific PCR (nine SNPs) or matrix-assisted laser disorption/ionization time-of-flight (MALDI-TOF) mass spectrometry of allele-specific extension products (seven SNPs). The genotyping method used for each SNP is indicated in Fig. 1.

thumbnail image

Figure FIG. 1. Diagram of the physical map and intron/exon boundaries for the human LRP5 gene. SNPs genotyped in this study are indicated at the top of the figure. Exon number and base pair coordinate from transcription start site are shown below the diagram. Circles denote SNPs genotyped by allele-specific PCR; squares denote SNPs genotyped by MALDI-TOF mass spectrometry.20

Download figure to PowerPoint

Fluorescent allele-specific PCR:

Allele-specific PCR was performed as previously described.(13) In brief, PCR primers were designed using Primer3 (Whitehead Institute). Forward primers were designed so that each allele can be differentiated by both size and color of the PCR products: each 5′-end was labeled with either FAM or HEX; the 5′-end of the FAM-labeled primer had two additional bases; the final 3′-nucleotide specifically matched the targeted allele. Only when the 3′-end of the primer matched the sequence exactly did extension occur. PCR was performed with two forward primers and one reverse primer in a single reaction. Primer sequences and PCR conditions are available on request. Multiple PCR products were pooled into two sets and separated using the 3100 Genetic Analyzer (PE Applied Biosystems, Foster City, CA, USA). The data were analyzed by two independent researchers using Genescan Analysis 3.7 software (PE Applied Biosystems).

MALDI-TOF mass spectrometry:

Genotyping of seven SNPs was performed using MALDI-TOF mass spectrometry of allele-specific primer extension products (MassARRAY System; SEQUENOM, San Diego, CA, USA). The multiplex assays were designed using the SpectroDESIGNER 2.0 software (SEQUENOM). All PCR and extension reactions were run under conditions slightly modified from the manufacturer's instructions (SpectroPREP User's Guide for Homogeneous MassEXTEND; SEQUENOM). In brief, genomic DNA was first amplified by PCR. After removal of residual dNTPs, allele-specific primer extension products were generated with a mixture of ddNTPs and dNTPs chosen such that one allele of each SNP would be extended by a single nucleotide, and the other by at least two nucleotides. Aliquots of the extension products were spotted onto SpectroCHIPs, and the alleles were determined by MALDI-TOF mass spectrometry. PCR and extension conditions are available on request.

Statistical analysis

Stepwise regression analysis was used on both lumbar spine and femoral neck BMD using height, weight, oral contraceptive use, pack-years of smoking, and age to identify significant covariates with BMD. A p value of ⩽0.10 was required for retention in the model. Regression residuals, representing covariate-adjusted BMD values, were computed and used in all analyses.

Haplotypes were generated from the SNP family data using the SIMWALK2 computer package.(14) Disequilibrium statistics (D′) for each pair of SNPs genotyped were calculated based on the observed haplotype and allele frequencies using the HAPLOXT program.(15)

By necessity, different samples were used for the population-based association test and the family-based test (quantitative transmission disequilibrium test [QTDT]). Because our sample consisted of sibling data, traditional tests requiring independent observations could not be directly applied. Rather, one sister from each of the 588 families was randomly selected to generate an unrelated sample for testing the population-based association hypothesis. ANOVA was performed for each of the genotyped SNP markers versus femoral neck and lumbar spine BMD. The independent variable in the ANOVA was genotype, which took on three levels corresponding to the three genotypes observed for each SNP (1,1; 1,2; 2,2). The ANOVA analysis was repeated on a second sample, consisting of a sister not selected from each sibship for the initial analysis.

The genotype and BMD data on sibpairs were also used to take advantage of family-based statistical approaches that are not prone to false positive results because of population stratification. Genotype data on parent(s), when available, were used in these approaches as well. The QTDT using the orthogonal model(16) was used to test for association between the LRP5 SNPs and BMD phenotypes. This method, as implemented in the QTDT computer software package,(16) extends the trio-based TDT(17) to quantitative trait data and allows for the inclusion of sibling and parental data.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Characteristics of the subjects ascertained for our study are shown in Table 1. The 12 LRP5 SNPs used in this study (designated SNP 1 to SNP 12 from the 5′- to 3′-ends) are shown schematically in Fig. 1. Detailed information about each polymorphism is given in Table 2. The disequilibrium coefficients (D′) estimated from our sibpair data are shown in Table 3. Our data show two regions of substantial linkage disequilibrium (LD) within LRP5. The first region containing SNPs 2, 3, and 4 spans ∼45 kb of the 5′-end of the LRP5 gene and has D′ values of ⩾0.76 in each pairwise comparison. The second region, located at the 3′-end of the gene, spans ∼40 kb and harbors SNPs 6-12. SNPs 6-12 have D′ values of ⩾0.65 with at least one other SNP in this group. Two SNPs, 1 and 5, are not in significant disequilibrium with any other SNP, and each forms its own haplotype block.

Table Table 1.. Characteristics of Study Subjects*
Thumbnail image of
Table Table 2.. Properties of LPR5 SNPs Employed in Study
Thumbnail image of
Table Table 3.. Linkage Disequilibrium (D′) Among SNPs in the LRP5 Gene
Thumbnail image of

Body weight and age were the only covariates tested that reached significance (p < 0.10) in regression model fitting versus BMD. Regression residuals representing age- and weight-adjusted BMD values were used in all further analyses. Body weight and age together explained 12.1% and 17.2% of the variation in spine and neck BMD, respectively, in our sample.

ANOVA was performed to test whether any of the 12 SNPs in LRP5 were associated with either femoral neck BMD (Table 4) or spine BMD (Table 5). The most significant population-based association result (p = 0.009) for femoral neck BMD was with SNP 5, located in intron 7. Although the test for association is significant, this SNP only accounts for 0.82% of the variation in femoral neck BMD in our sample as measured by the ANOVA r2 value. Significant evidence of association (p = 0.001) with lumbar spine BMD was also obtained for SNPs 5 and 12. Similar to the hip BMD linkage, each of these SNPs explained only a very small proportion of the variation in spine BMD (1.18% and 1.14%, respectively). Results for all SNPs were nearly identical when we performed the same test in a confirmatory sample, comprised of one sister from each sibship who was not selected for the initial analysis. Results were also essentially unchanged when BMD values unadjusted for body weight were analyzed.

Table Table 4.. Association Results for Femoral neck BMD
Thumbnail image of
Table Table 5.. Association Results for Lumbar Spine BMD
Thumbnail image of

In all cases, haplotypes constructed using LRP5 SNPs explained less of the BMD variation than did the individual SNPs. Haplotypes constructed from the four SNPs suggesting association to BMD (SNPs 1, 5, 11, and 12) are shown in Table 6 along with their observed frequencies in our sample. Mean femoral neck and lumbar spine BMD values are also presented by genotype (Table 7) for the LRP5 SNPs showing the strongest evidence of association to these phenotypes.

Table Table 6.. Observed Frequencies for a Haplotype Constructed from SNPs 1, 5, 11, and 12
Thumbnail image of
Table Table 7.. Mean BMD Values by Genotype for the Most Strongly Associated LRP5 SNPs in the Study
Thumbnail image of

Subsequently, the QTDT test was performed to determine if any of the SNPs in LRP5 were associated with peak BMD in the sample of premenopausal sister pairs. There was no evidence of association (all p > 0.05) between any of the 12 SNPs tested in the LRP5 gene and either femoral neck Table 4) or spine BMD (Table 5). The QTDT test was likewise nonsignificant (p > 0.05) for the four-locus haplotype constructed from SNPs 1, 5, 11, and 12.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

We tested whether sequence variation in the LRP5 gene accounts for a significant proportion of the variation in normal peak BMD. Our study has several notable strengths. First, we used a relatively large sample of 1301 premenopausal white women from 588 families who have undergone rigorous BMD phenotyping. Second, rather than testing only a few SNPs in the candidate gene, we have tested 12 SNPs distributed throughout the LRP5 gene. Third, we deliberately selected SNPs that had high heterozygosity to maximize our power to detect association with BMD. Fourth, we analyzed LD between SNPs and used this information to help interpret the results of association analyses. Fifth, we used complementary analytic methods to test for association between LRP5 and peak BMD.

The LD pattern we observed between the 12 genotyped LRP5 SNPs in our study was in excellent agreement with that observed by Twells et al.(18) Both studies found substantial LD throughout the 3′ portion of the LRP5 gene and the presence of a region of very low LD near the middle of the coding region (SNP 5 in our study). The only discrepancy between these two studies is the observation by Twells et al. of additional low LD areas between SNPs 2 and 3, resulting in an additional haplotype block. This difference is most likely because of the greater SNP density achieved in this region in their study compared with our own, which highlights the complexity of haplotype blocks and linkage disequilibrium.(19)

To test for association between SNPs in LRP5 and peak BMD, we used two complementary statistical designs. First, we used a population-based association test that tests for significant differences in mean BMD among the three SNP genotypes. This test has maximal power to detect association, but is susceptible to false positive results because of population stratification.(20) Second, we used a family-based test of association, the QTDT, which calculates the difference between the value of the quantitative trait of the offspring and the average quantitative trait of all offspring in all families while simultaneously considering the allele transmission from parent to offspring. Examining transmission of alleles from parent to offspring is a particularly robust method to test for association and is less susceptible to the effects of population stratification.(17) Unfortunately, the test is less powerful for the detection of association.(21)

Using population-based methods of analyses, we detected significant (p < 0.05) evidence of association with both femoral neck and lumbar spine BMD. These results are in agreement with a recent study of Japanese women.(22) Despite our significant results, these SNPs in the LRP5 gene explain a very small proportion of the variation (∼1%) in femoral neck and lumbar spine BMD in this sample of white premenopausal women. In addition, examination of the mean BMD values by genotype for the SNPs with significant evidence of association (Table 7) shows minimal BMD differences (0.01-0.03 g/cm2) between genotype groups of reasonable size (n > 40). In some cases, the pattern of mean BMD values for a SNP with evidence of association does not agree with a plausible biological model (e.g., the decreased heterozygote mean for SNP 1 and femoral neck BMD). These results suggest that our tests for association led to significance because of our large sample size and do not suggest a clinically relevant effect for the tested LRP5 variants. Consistent with this interpretation of the population-based association tests, we found no evidence of association when we used the less-powerful family-based tests of association.

Our findings in premenopausal white women are in contrast to those of Ferrari et al.,(23) who recently reported association of LRP5 with BMC in a sample of white men. Their study found that LRP5 variation accounted for 15% of the variability in BMC compared with the effect of ∼1% for BMD in our study. Both the exon 9 and exon 18 SNPs associated with BMC in the study of males were genotyped in our study as well. It should be noted that our association study in women had >99% power to detect an effect of the size reported by Ferrari et al. in male subjects. Results from our sample suggest that natural variation in and around LRP5 is not a major contributor to the observed variability in peak BMD, at either the femoral neck or lumbar spine, in white women.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

The authors thank the sisters and parents who participated in this study, as well as the study coordinators, without whom this work could not have been accomplished. This work was supported by NIH Grants PO1 AG-18397, ROI AR-43476, MO1 RR-00750, K24 AR-02095, AR-4370, and T32 HD-07373 and the Eli Lilly and Co. Centre for Women's Health. SNP genotyping by MALDI-TOF used the facilities of the Center for Medical Genomics at Indiana University School of Medicine, which is supported in part by a grant from the Indiana Genomics Initiative (INGEN is supported in part by the Lilly Endowment, Inc.).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  • 1
    Melton LJ III, Atkinson EJ, O'Fallon WM, Wahner HW, Riggs BL 1993 Long-term fracture prediction by bone mineral assessed at different skeletal sites. J Bone Miner Res 8: 12271233.
  • 2
    Marshall D, Johnell O, Wedel H 1996 Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ 312: 12541259.
  • 3
    Peacock M, Turner CH, Econs MJ, Foroud T 2002 Genetics of osteoporosis. Endocr Rev 23: 303326.
  • 4
    Johnson ML, Gong G, Kimberling W, Recker SM, Kimmel DB, Recker RB 1997 Linkage of a gene causing high bone mass to human chromosome 11 (11q12–13). Am J Hum Genet 60: 13261332.
  • 5
    Little RD, Carulli JP, Del Mastro RG, Dupuis J, Osborne M, Folz C, Manning SP, Swain PM, Zhao SC, Eustace B, Lappe MM, Spitzer L, Zweier S, Braunschweiger K, Benchekroun Y, Hu X, Adair R, Chee L, FitzGerald MG, Tulig C, Caruso A, Tzellas N, Bawa A, Franklin B, McGuire S, Nogues X, Gong G, Allen KM, Anisowicz A, Morales AJ, Lomedico PT, Recker SM, Van Eerdewegh P, Recker RR, Johnson ML 2002 A mutation in the LDL receptor-related protein 5 gene results in the autosomal dominant high-bone-mass trait. Am J Hum Genet 70: 1119.
  • 6
    Boyden LM, Mao J, Belsky J, Mitzner L, Farhi A, Mitnick MA, Wu D, Insogna K, Lifton RP 2002 High bone density due to a mutation in LDL-receptor-related protein 5. N Engl J Med 346: 15131521.
  • 7
    Van Hul E, Gram J, Bollerslev J, Van Wesenbeeck L, Mathysen D, Andersen PE, Vanhoenacker F, Van Hul W 2002 Localization of the gene causing autosomal dominant osteopetrosis type I to chromosome 11q12–13. J Bone Miner Res 17: 11111117.
  • 8
    Van Wesenbeeck L, Cleiren E, Gram J, Beals RK, Benichou O, Scopelliti D, Key L, Renton T, Bartels C, Gong Y, Warman ML, De Vernejoul MC, Bollerslev J, Van Hul W 2003 Six novel missense mutations in the LDL receptor-related protein 5 (LRP5) gene in different conditions with an increased bone density. Am J Hum Genet 72: 763771.
  • 9
    Gong Y, Slee RB, Fukai N, Rawadi G, Roman-Roman S, Reginato AM, Wang H, Cundy T, Glorieux FH, Lev D, Zacharin M, Oexle K, Marcelino J, Suwairi W, Heeger S, Sabatakos G, Apte S, Adkins WN, Allgrove J, Arslan-Kirchner M, Batch JA, Beighton P, Black GC, Boles RG, Boon LM, Borrone C, Brunner HG, Carle GF, Dallapiccola B, De Paepe A, Floege B, Halfhide ML, Hall B, Hennekam RC, Hirose T, Jans A, Juppner H, Kim CA, Keppler-Noreuil K, Kohlschuetter A, LaCombe D, Lambert M, Lemyre E, Letteboer T, Peltonen L, Ramesar RS, Romanengo M, Somer H, Steichen-Gersdorf E, Steinmann B, Sullivan B, Superti-Furga A, Swoboda W, van den Boogaard MJ, Van Hul W, Vikkula M, Votruba M, Zabel B, Garcia T, Baron R, Olsen BR, Warman ML 2001 LDL receptor-related protein 5 (LRP5) affects bone accrual and eye development. Cell 107: 513523.
  • 10
    Lev D, Binson I, Foldes AJ, Watemberg N, Lerman-Sagie T 2003 Decreased bone density in carriers and patients of an Israeli family with the osteoporosis-pseudoglioma syndrome. Isr Med Assoc J 5: 419421.
  • 11
    Kato M, Patel MS, Levasseur R, Lobov I, Chang BH, Glass DA, Hartmann C, Li L, Hwang TH, Brayton CF, Lang RA, Karsenty G, Chan L 2002 Cbfa1-independent decrease in osteoblast proliferation, osteopenia, and persistent embryonic eye vascularization in mice deficient in Lrp5, a Wnt coreceptor. J Cell Biol 157: 303314.
  • 12
    Koller DL, Rodriguez LA, Christian JC, Slemenda CW, Econs MJ, Hui SL, Morin P, Conneally PM, Joslyn G, Curran ME, Peacock M, Johnston CC, Foroud T 1998 Linkage of a QTL contributing to normal variation in bone mineral density to chromosome 11q12–13. J Bone Miner Res 13: 19031908.
  • 13
    Carn G, Koller DL, Peacock M, Hui SL, Evans WE, Conneally PM, Johnston CC Jr, Foroud T, Econs MJ 2002 Sibling pair linkage and association studies between peak bone mineral density and the gene locus for the osteoclast-specific subunit (OC116) of the vacuolar proton pump on chromosome 11p12–13. J Clin Endocrinol Metab 87: 38193824.
  • 14
    Sobel E, Lange K 1996 Descent graphs in pedigree analysis: Applications to haplotyping, location scores, and marker-sharing statistics. Am J Hum Genet 58: 13231337.
  • 15
    Abecasis GR, Cookson WO 2000 GOLD-graphical overview of linkage disequilibrium. Bioinformatics 16: 182183.
  • 16
    Abecasis GR, Cardon LR, Cookson WO 2000 A general test of association for quantitative traits in nuclear families. Am J Hum Genet 66: 279292.
  • 17
    Spielman RS, Ewens WJ 1996 The TDT and other family-based tests for linkage disequilibrium and association. Am J Hum Genet 59: 983989.
  • 18
    Twells RC, Mein CA, Phillips MS, Hess JF, Veijola R, Gilbey M, Bright M, Metzker M, Lie BA, Kingsnorth A, Gregory E, Nakagawa Y, Snook H, Wang WY, Masters J, Johnson G, Eaves I, Howson JM, Clayton D, Cordell HJ, Nutland S, Rance H, Carr P, Todd JA 2003 Haplotype structure, LD blocks, and uneven recombination within the LRP5 gene. Genome Res 13: 845855.
  • 19
    Phillips MS, Lawrence R, Sachidanandam R, Morris AP, Balding DJ, Donaldson MA, Studebaker JF, Ankener WM, Alfisi SV, Kuo FS, Camisa AL, Pazorov V, Scott KE, Carey BJ, Faith J, Katari G, Bhatti HA, Cyr JM, Derohannessian V, Elosua C, Forman AM, Grecco NM, Hock CR, Kuebler JM, Lathrop JA, Mockler MA, Nachtman EP, Restine SL, Varde SA, Hozza MJ, Gelfand CA, Broxholme J, Abecasis GR, Boyce-Jacino MT, Cardon LR 2003 Chromosome-wide distribution of haplotype blocks and the role of recombination hot spots. Nat Genet 33: 382387.
  • 20
    Schork NJ, Fallin D, Thiel B, Xu X, Broeckel U, Jacob HJ, Cohen D 2001 The future of genetic case-control studies. Adv Genet 42: 191212.
  • 21
    Wacholder S, Rothman N, Caporaso N 2002 Counterpoint: Bias from population stratification is not a major threat to the validity of conclusions from epidemiological studies of common polymorphisms and cancer. Cancer Epidemiol Biomarkers Prev 11: 513520.
  • 22
    Mizuguchi T, Furuta I, Watanabe Y, Tsukamoto K, Tomita H, Tsujihata M, Ohta T, Kishino T, Matsumoto N, Minakami H, Niikawa N, Yoshiura K 2004 LRP5, low-density-lipoprotein-receptor-related protein 5, is a determinant for bone mineral density. J Hum Genet 49: 8086.
  • 23
    Ferrari SL, Deutsch S, Choudhury U, Chevalley T, Bonjour JP, Dermitzakis ET, Rizzoli R, Antonarakis SE 2004 Polymorphisms in the low-density lipoprotein receptor-related protein 5 (LRP5) gene are associated with variation in vertebral bone mass, vertebral bone size, and stature in whites. Am J Hum Genet 74: 866875.