Skip to main content
PMC home page
BoneKEy Reports logoLink to BoneKEy Reports
. 2015 Jul 8;4:714. doi: 10.1038/bonekey.2015.83

Genetic regulation of bone strength: a review of animal model studies

Douglas J Adams 1,a, Cheryl L Ackert-Bicknell 2,b
PMCID: PMC4495779  PMID: 26157577

Abstract

Population- and family-based studies have established that fragility fracture risk is heritable; yet, the genome-wide association studies published to date have only accounted for a small fraction of the known variation for fracture risk of either the femur or the lumbar spine. Much work has been carried out using animal models toward finding genetic loci that are associated with bone strength. Studies using animal models overcome some of the issues associated with using patient data, but caution is needed when interpreting the results. In this review, we examine the types of tests that have been used for forward genetics mapping in animal models to identify loci and/or genes that regulate bone strength and discuss the limitations of these test methods. In addition, we present a summary of the quantitative trait loci that have been mapped for bone strength in mice, rats and chickens. The majority of these loci co-map with loci for bone size and/or geometry and thus likely dictate strength via modulating bone size. Differences in bone matrix composition have been demonstrated when comparing inbred strains of mice, and these matrix differences may be associated with differences in bone strength. However, additional work is needed to identify loci that act on bone strength at the materials level.

Introduction

It is now understood that hip, vertebral and wrist fracture risk are influenced by genetic factors.1,2,3,4 Although fracture incidence can and has been used for heritability and forward genetic mapping studies,3,5,6,7,8,9 data collection is restricted to retrospective measurements from clinical exams and medical records. Retrospective data often miss non-reported fractures and/or may miscategorize individuals who are phenotypically at risk but have not actually incurred a fracture.8 Reduced skeletal strength has primarily been associated with loss of bone mass and concomitant changes in morphology. Extensive work has been carried out to establish the genetic underpinnings of bone mineral density (BMD) as a surrogate measure of strength, successfully identifying novel genes that have key roles in bone biology (reviewed in Hsu and Kiel).6 Bone mass and geometry are phenotypes that can be easily, reliably and noninvasively measured in large populations of patients. This is a major reason for the success of genome-wide association studies (GWASs) for these phenotypes, as sufficient statistical power could be achieved and data were available for tens of thousands of genotyped individuals. However, the genetic correlations between BMD measurements in the spine and vertebral fracture incidence are modest. This suggests that studies in which genetic loci are mapped for surrogate phenotypes of strength2 will not capture fracture incidence completely. As is true for any structure, bone strength is a function of the magnitude and architectural distribution of its mass, as well as the inherent compositional quality and resultant material strength of its constituent tissue matrix. Currently, measurement of these phenotypes requires invasive and typically destructive techniques, thus relegating their application to animal models, which allow collection of a more complete spectrum of genetic and phenotypic data. Given that osteoporotic fractures remain a substantial health burden in developed nations and are likely to increase in number with a proportionally aging population, there has been great interest in identifying the genes that regulate bone strength.

Animal Models Used for Genetic Studies of Bone Strength

Among model organisms, mice and rats are considered to be the most important for the study of human genetic skeletal diseases. The mouse genome, although 14% smaller than the human genome, is remarkably comparable at the nucleotide level. At the gene level, ∼17 770 mouse genes have a known direct human ortholog (http://www.informatics.jax.org). Organizationally, the mouse and human genomes remain highly syntenic despite a quite long evolutionary distance between them.10 Thus, genetic findings in mice are often concordant with genetic findings in humans11 and have been used extensively in studies of the inheritance of bone mass and geometry.11,12,13,14,15,16,17 Although larger than the mouse genome, the rat genome is still smaller than the human, but like the mouse the rat genome is highly orthologous to human. The majority of genes found in the human genome can be found in the rat without duplication or deletion.18 Although many genetic mapping studies for bone phenotypes have been conducted using larger livestock species such as a sheep, cows and pigs, no direct measures of bone strength have been reported. Numerous studies have reported that, as in humans, fracture risk is highly heritable in horses,19,20 but as for other large mammals no loci have been reported for direct measures of breaking strength.

Moving away from mammalian models, substantial work has been carried out examining the genetics of bone in chickens. Although this work was conducted in part for the benefit of the agricultural industry, there are sufficient similarities in basic bone physiology and anatomical patterning when comparing with mammals such that genetic findings from studies in poultry may be informative for interpreting mammalian genetics. At the genome level, there is homology between chickens and humans, and between chickens and rodents. Although only 2.5% of the chicken genome sequence could be directly aligned to the human genome, 75% of coding and between 30 and 40% of gene regulatory regions are conserved between species.21 This degree of homology, although not perfect, may be sufficient to be informative for cross-species comparisons of genetic loci.

Types of Genetic Studies Conducted

Genetic studies can have two main starting points. In forward genetics studies one starts with a phenotype, with the goal of identifying the gene(s) and genetic polymorphisms that are responsible for that phenotype. In contrast, reverse genetics attempts to understand what phenotypes are affected by known polymorphisms in a gene already identified. In animal models, the historical workhorse of forward genetic screening has been the two-strain intercross. In short, two inbred strains are interbred once (resulting in F1 generation animals), and these F1 animals are either backcrossed to one of the original founder strains (N1F1 generation) or the F1 animals are bred together to make second-generation intercross animals (F2 generation). Then, either the N1F1 or the F2 animals (depending on study design) are phenotyped, genotyped at a reasonable marker density, and regions of the genome associated with the phenotype are identified. These so-called quantitative trait loci (QTL) harbor polymorphic differences when comparing the founder strains that are causative for some, or all, of the phenotypic differences between the two original strains.22 There are a number of variations on this strategy, including the recombinant inbred lines (RI). Each RI population is a series of strains of mice descended from two or more progenitor strains but which has been bred to homozygousity.23 Thus, unlike an F2 mouse, wherein each one is a genetic one in a million, the combination of alleles found in any one RI mouse is mirrored by all other RI mice of that same strain.

The drawback of these genetic mapping approaches has been that, although numerous genomic regions have been found that influence a phenotype, these regions are large and often contain hundreds of genes. Indeed, the mean QTL interval size for BMD, as mapped in mice, was found to be ∼32cM.11 One cM in genetic distance is roughly equivalent to 2 Mb in genomic distance for mice. At an average of 10 genes per Mb,24 this equates to over 600 candidate genes, on average, per QTL.25 As a result, the use of the two-strain, two-generation cross has started to fall out of favor for forward genetics studies. These two-strain populations are increasingly being replaced by studies using outbred populations of mice and rats. These include the use of commercially available outbred animals or ‘designer' outbred populations, which are created by breeding two or more strains over the course of multiple generations in a manner that minimizes inbreeding.26,27,28,29

Additional study designs have been used to better define the genetic etiology of bone strength, including the use of congenic and consomic genetic reference populations.23 A consomic, which is sometimes referred to as chromosome substitution, is a strain wherein all of the alleles for an entire chromosome from one inbred strain of mice or rats have been moved onto an otherwise pure background of a second strain. Similarly, a congenic is a strain with part of a chromosome moved from one genetic background to another. Both consomic and congenic strains are generated by selective breeding and thus take several years to generate.30 However, the resulting animals allow confirmation or de novo establishment that genetic regions contain polymorphism(s) impacting phenotype and allow study of the biology associated with a QTL without necessarily knowing the causative gene.

Phenotypes of Bone Strength

The mechanical integrity of a whole bone, as for any force-bearing structure, is a product of geometrical size and shape, mechanical properties of its constituent material (that is, bone matrix) and the forces it must sustain. The most informative phenotypes describing bone strength are measurements of mechanical integrity derived from direct, destructive testing.31 Surrogate measures obtained by imaging, such as dual X-ray absorptiometry or computed tomography, can provide only indirect measures of BMD, morphometry and size. Thus, the exercise of defining and measuring phenotypes describing the mechanical strength and integrity of whole bones (structural integrity) and constituent bone matrix (material integrity) inherently requires test methods of engineering mechanics. The regions of primary interest in skeletal phenotyping include anatomical sites comprising relatively large volumes of cancellous bone that are prone to osteoporotic fracture, such as the vertebral centrum, femoral neck and the distal radius. An engineering approach to measuring the mechanical integrity of these anatomical structures would include separate treatment of the geometry and constituent material integrity, the latter requiring careful preparation of uniform test specimens. As the bones from rodents are quite small, genetic studies have largely relied heavily on whole bone structural tests and predominantly on long bone cortical diaphyses subjected to flexural loading (bending).

Flexural tests of whole bones such as femur and tibia subject the central diaphysis to bending by resting a whole bone freely on end supports and subjecting the mid-span to force applied at one or two contact points (thus, a total of either 3 or 4 points of contact, providing the so-called 3-point or 4-point bending schemes).32 This chosen loading scheme places the top span of bone length in compression and the bottom span in tension. Force is applied at a constant rate of deflection until failure, providing a force versus deflection relationship from which whole bone flexural stiffness is measured as the linear slope of the relationship (unit force per unit length of deflection). The maximum force attained is regarded as the flexural strength of the whole bone, and the integrated area under the force-displacement ‘curve' is defined to be the whole bone fracture energy. Further subdivisions of mechanical behavior can be described as well, including the force and displacement at which whole bone behavior ceases to respond elastically and begins to ‘yield' or deform plastically such that it will no longer retain its original shape. The force, displacement and energy might thus be parsed into that occurring during elastic versus ‘post-yield' deformation. Importantly, the magnitudes of these mechanical test parameters depend on the size and shape of the diaphysis, the test configuration (for example, support span length and anatomical orientation of the specimen with respect to loading) and the inherent integrity of the constituent bone matrix as a material. As such, parameters derived from mechanical integrity tests of whole bones may reflect differences in size and shape or differences in constituent bone matrix integrity. Thus, a structural test performed in the absence of bone geometry measurement is insufficient to separate the effects of size and shape versus matrix integrity.

To estimate the material integrity of cortical bone matrix from the structural flexure test, accurate measurements of transverse (cross-sectional) geometry are required. Such geometrical measurements are most accurately and conveniently achieved with digital images acquired by X-ray computed tomography.33 Notably, neither the cross-sectional area of the diaphysis nor the cortical thickness provides measures sufficient to estimate constituent bone matrix integrity from flexural tests. Mathematical descriptions from engineering mechanics beam theory provide all of the tools required for interpreting the test at the material scale, dictating that the cross-sectional second moment of area (colloquially referred to as moment of inertia) drives the mechanical behavior of flexure (or polar second moment of area for torsion). This measurement of cross-sectional size and shape provides a geometrical measure of diaphyseal stiffness and strength, which in combination with the structural test provides an approximation to a material test.

Many stipulations of beam flexure are violated by the cross-sectional asymmetry and non-constant size and shape throughout the length of the diaphysis, as well as its very small length:diameter ratio (which is ∼4:1 for rodent long bones but should be a minimum of 16:1 by ASTM D 790 standard.34 As such, bone geometry is not captured by a single mid-diaphysis measurement, and whole bone stiffness measurement is compromised by shear forces imposed by the short length:diameter ratio. The mathematical reduction in the structural test loading configuration to mechanical stress (the concentration of force through a unit area) as an estimate of material strength is confounded by the lack of pure flexure due to the short length:diameter ratio. Moreover, calculation of mechanical strain and estimates of matrix material elastic modulus are greatly compromised by not considering the entire span of changing cross-sectional geometry.35

Summary of Genetic Mapping Studies

Studies in rats

In total, we identified four study populations in which bone strength was directly measured in a de novo cross between inbred strains for the purpose of conducting a forward genetic mapping study (Table 1), and these preliminary mapping studies have served as the foundation for more complex bioinformatics explorations.36,37 As is highlighted in Table 1, 75 QTL for bone strength and geometry phenotypes were identified in these populations. It is immediately apparent that there is substantial co-mapping of loci for these various phenotypes, and, although true epistasis cannot be established from these data, for discussion they can be collapsed into ∼41 discrete loci. Of these 41, 23 represent loci to which only geometry and/or bone size phenotypes map (for example, total cross-sectional area, cortical area, width and polar moment of area), and 7 more co-map with one or more geometry phenotypes. On the basis of the known confounding issues for flexural bone breaking tests of long bones and femoral neck, it is reasonable to postulate that the majority of these 41 loci do not represent loci controlling bone matrix quality. Although they do represent a genetic region controlling whole bone strength, they do so by modulating the bone size, which is important for consideration when attempting to ascertain the biological processes modulated by these loci. This leaves 11 loci for which no geometry phenotype locus is reported as co-mapping. It is tempting to consider these loci as regulating bone matrix quality, but one must remain cognizant of the fact that this table lists only the reported QTL and makes no attempt toward identifying sub-significant loci for any phenotype. Likely, some of these 11 loci are impacting the bone matrix, but identifying such matrix loci may require refining test methods beyond those used in these studies.

Table 1. Quantitative trait loci for size and strength mapped in rats.
Chr Phenotypea Bone Peak or confidence interval (cM)b Genderc Strains Reference
1 Flexural strength Tibia 17–42 Male and female GKxF344 54
1 Cortical area Tibia 10–59 Male and female GKxF344 55
1 Polar 2nd moment of area Tibia 10–59 Male and female GKxF344 55
1 Flexural stiffness Tibia 18–43 Male and female GKxF344 54
1 Cortical area Tibia 36–93 Male and female GKxF344 55
1 Polar 2nd moment of area Tibia 36–93 Male and female GKxF344 55
1 Flexural strength Femur 101.4 Female F344xLEW 56
1 Cortical area Tibia 95–124 Male and female GKxF344 55
1 Polar 2nd moment area Femur 110 Male and female COPxDA 57
1 Total area Femur 111 Male and female COPxDA 57
1 Cortical area Femur 112 Male and female COPxDA 57
1 Neck width Femur Neck 113 Male and female COPxDA 57
1 Flexural strength Femur 114 Male and female COPxDA 57
1 Fracture energy Femur 114 Male and female COPxDA 57
1 Fracture force Femur Neck 114 Male and female COPxDA 58
1 Flexural stiffness Femur 117 Male and female COPxDA 57
2 Fracture Femur 36 Male COPxDA 58
  Force Neck   Female   58
2 Fracture energy Femur neck 88 Male and female COPxDA 58
2 Flexural strength Femur 102.2 Female F344xLEW 56
2 Fracture energy Femur 102.2 Female F344xLEW 56
2 Polar 2nd moment of area Femur 102.2 Female F344xLEW 56
2 Cortical area Femur 102.2 Female F344xLEW 56
3 Cortical area Tibia 9–42 Male and female GKxF344 55
4 Polar 2nd moment of area Femur neck 27.2–48.7 Female F344xLEW 59
4 Fracture force Femur neck 27.2–48.7 Female F344xLEW 59
4 Fracture energy Femur neck 27.2–48.7 Female F344xLEW 59
4 Total area Femur neck 27.2–48.7 Female F344xLEW 59
4 Neck width Femur neck 27.2–55.7 Female F344xLEW 59
4 Polar 2nd moment of area Femur 34–55 Male PxNP 38
4 Flexural strength Femur 34–55 Male PxNP 38
4 Fracture force Femur neck 34–55 Male PxNP 38
4 Polar 2nd moment of area Femur 57.7 Female F344xLEW 56
4 Cortical area Femur 57.7 Female F344xLEW 58
4 Fracture force Femur neck 87 Male and female COPxDA 58
5 Cortical area Femur neck 3 Male and female COPxDA 58
5 Cortical area Tibia 0–12 Male and Female GKxF344 55
5 Flexural strength Femur 68.3 Female F344xLEW 56
5 Fracture energy Femur 68.3 Female F344xLEW 56
5 Flexural stiffness Femur 68.3 Female F344xLEW 56
5 Polar 2nd moment of area Femur 68.3 Female F344xLEW 56
5 Cortical area Femur 68.3 Female F344xLEW 56
5 Flexural strength Femur 84 Male and female COPxDA 57
6 Cortical area Femur 16 Male and female COPxDA 57
6 Polar 2nd moment of area Femur 20 Male and female COPxDA 57
6 Total area Femur 20 Male and female COPxDA 57
6 Bone area (longitudinal) Tibia 30–60 Male GKxF344 54
7 Polar 2nd moment of area Femur 31 Female F344xLEW 56
7 Total area Femur 37 Male and female COPxDA 57
7 Polar 2nd moment of area Femur 40 Male and female COPxDA 57
7 Total area Femur neck 52 Male and female COPxDA 58
7 Polar 2nd moment of area Femur neck 53 Male and female COPxDA 58
8 Bone area (longitudinal) Tibia 43–59 Male and female GKxF344 54
10 Fracture energy Femur 8 Male and female COPxDA 57
10 Cortical area Tibia 5–18 Male and female GKxF344 55
10 Polar 2nd moment of area Tibia 5–18 Male and female GKxF344 55
10 Total area Femur neck 13 Male and female COPxDA 58
10 Polar 2nd moment of area Femur neck 43 Male and female COPxDA 58
10 Compression strength L5 58.5 Female F344xLEW 56
10 Polar 2nd moment of area Femur 73 Male and female COPxDA 57
10 Total area Femur 73 Male and female COPxDA 57
12 Total area Femur neck 43 Male and female COPxDA 58
13 Polar 2nd moment of area Femur 41 Male and female COPxDA 57
13 Total area Femur 41 Male and female COPxDA 57
15 Flexural strength Tibia 10–32 Male and female GKxF344 54
15 Fracture energy Tibia 25–37 Male and female GKxF344 54
15 Cortical area Femur 32 Male and female COPxDA 57
15 Fracture force Femur neck 36 Male and female COPxDA 58
15 Fracture energy Femur neck 36 Male and female COPxDA 58
15 Cortical area Tibia 55–69 Male and female GKxF344 55
15 Polar 2nd moment of area Femur 65.3 Female F344xLEW 56
17 Cortical area Femur neck 21 Male and female COPxDA 58
18 Cortical area Tibia 21–31 Male and female GKxF344 55
18 Polar 2nd moment of area Femur 29 Male and female COPxDA 57
19 Compression strength L5 5.2 Female F344xLEW 56
X Bone area (longitudinal) Tibia 52–74 Male and female GKxF344 54
Open in a new tab

aMechanical integrity phenotypes were derived from long bone 3-point bending (flexure) tests or combined compression bending for the femoral neck. Mechanical phenotypes for flexure tests include whole bone flexural strength, stiffness and fracture energy. Phenotypes for femoral neck and vertebrae include fracture force and energy. Phenotypes describing long bone size and shape include bone width, cross-sectional total and cortical areas, and cross-sectional polar 2nd moment of area (the geometrical measure of bone strength and stiffness). No phenotypes are reported for bone matrix (material) mechanical integrity.

bAs provided in the literature. When more than one model was calculated, the peak for at the maximum LOD is provided.

cIndicates gender of animals phenotyped. Not specificity of the locus.

At this stage, these studies represent loci, and loci represent statistical associations with a phenotype. Further, these loci are large and encompass too many genes at this point to name the causative gene for the most part. Toward this first point, Alam et al.38 established that the locus on Chr 4 was indeed causative using congenic rats, and this has not been done for the other loci. In other species, using a series of nested congenic strains has been a successful method to narrow loci down to near single gene for other key bone phenotypes,39 but this method is laborious and takes years to accomplish. For two of these loci, follow-up gene expression work was conducted. In these studies, expression of all genes within the loci was examined by microarray and genes demonstrating differential expression were examined for correlation with the phenotype(s) of interest.36,37 A candidate gene must impact a phenotype by either altering the amount of a gene product or by altering its function. Thus, expression studies are an efficient tool for narrowing large loci to a testable number of candidate genes.

A plethora of additional bioinformatics techniques have been developed to narrow loci. The rat genome has now been sequenced for many of these strains (http://rgd.mcw.edu). These data can be used to narrow such large QTL to eliminate genes that are not polymorphic from the loci interval and may identify key genes harboring non-sense or mis-sense mutations that could be tested for causality (altered function). Using the assumption that a single gene would be responsible for a loci mapped by two different crosses to the same spot, one can combine the allele information for all strains to further eliminate genes from the interval.40 Unfortunately, there is a paucity of co-mapping loci for strength wherein the same loci were found in multiple crosses and generating additional crosses could be beneficial. However, one needs to consider QTL from other species for the same or similar phenotypes that may have mapped to homologous regions. A wealth of information exists from mouse studies (see below). Combining species information has not been done for strength QTL but may be another successful method for narrowing these QTL. Covering the possible bioinformatics approaches that could be applied is beyond the scope of this review. The reviews by Peters et al.23 and DiPetrillo et al.40 provide details of these methods. Many of these techniques were originally developed for use with mouse QTL and could just as easily be applied to the QTL described in Table 1. Once one has found the gene(s) causative for a QTL, one needs to study the involvement of that gene in the phenotype(s) of interest. For mice, that means turning to transgenic models. This is now possible with rats41,42 but is not a strategy that has been widely adopted.

Moving forward, populations such as the heterogeneous stock (HS) rat outbred population will likely replace the F2 model for forward genetics studies. The HS population is descended from eight genetically diverse founder strains: ACI/N, BN/SsN, BUF/N, F344/N, M520/N, MR/N, WKY/N and WN/N. Each HS rat represents a unique combination of alleles, analogous to each F2 intercross animal. As these rats have been bred for over for 50 generations, the density of visible genomic recombinations is higher than is possible with only two generations of interbreeding, leading to increased genetic mapping resolution, decreasing the number of candidate genes per loci and thereby increasing the likelihood of identifying the causative polymorphism(s).43 Preliminary studies with this population show a high degree of heritability for key phenotypes such as flexural strength at the mid-diaphysis and femoral neck, yet a low degree of heritability for flexural fracture energy at mid-diaphysis.44 Geometric phenotypes were not reported, precluding extrapolation of these findings toward making any inferences regarding bone matrix quality. Although loci for bone strength phenotypes have not yet been reported in the HS rat, QTL for bone mass have been, and the narrow confidence intervals of these QTL demonstrate the superior mapping resolution possible when using this population.26 This highlights the superiority of the HS rat for any forward genetic mapping study.

Studies in mice

Genetic mapping studies in mice have followed similar strategies and approaches as for rats, in that whole bone strength was measured using similar methods (Table 2). Further, many more studies examining the genetic control of bone mass have been conducted in mice, but the review of these studies that lack a bone strength assessment as well is beyond the scope of this report. Thus, we are only reporting bone geometry phenotypes in Table 2 for mouse populations in which whole bone strength QTL were mapped directly. This listing of 119 QTL can be binned conservatively into ∼50 QTL (based on peak locations provided by authors, assuming that QTL mapped 10 or more cM apart represent independent loci). As was observed for rats, in many instances, QTL for readouts of bone strength co-map with loci for bone size. Again, this demonstrates that many of these readouts for what are often attributed as capturing bone strength are in fact merely indirect measures of bone size. For example, the locus mapped to chromosome (Chr) 4 in mice, centered at ∼60 cM, is most likely a locus that affects bone size, leading to the mapping of whole bone strength to this genetic region. This hypothesis is supported by the phenotype of congenic mice carrying c3h alleles at this locus, on an otherwise C57BL/6J (B6) background. Female congenic mice have increased polar 2nd moment of inertia and concomitant increased flexural strength of the femur.45

Table 2. Quantitative trait loci for size and strength mapped in mice.
Chr Phenotypea Bone Peak or confidence interval (cM)b Genderc Strains References
1 Flexural strength Tibia 12–17 Female B6xC3H 60
1 Flexural strength Femur 20 Female NZBxRF 17
1 Bone width Femur 20 Female NZBxRF 17
1 Yield stress Femur 41 Male and female HcB-8xHcB-23 16
1 Maximum stress Femur 56 Male and female HcB-8xHcB-23 16
1 2nd moment of area Femur 66 Male and female HcB-8xHcB-23 61
1 Post-yield strain Femur 66 Male and female HcB-8xHcB-23 16
1 Cortical area Femur 67 Male and female HcB-8xHcB-23 61
1 Flexural strength Femur 103.8 Female MRLxSJL 15
2 Maximum stress Femur 36 Male and female HcB-8xHcB-23 16
2 Cortical area Femur 43 Male and female HcB-8xHcB-23 61
2 Elastic modulus Tibia 50–56 Female B6xC3H 60
2 Flexural strength Femur 54.6 Female MRLxSJL 15
3 Fracture energy Femur 5 Male and female HcB-8xHcB-23 61
3 Flexural strength Femur 11 Male and female HcB-8xHcB-23 61
3 Flexural stiffness Femur 11 Male and female HcB-8xHcB-23 61
3 Cortical area Femur 11 Male and female HcB-8xHcB-23 61
3 Fracture energy Femur 34 Male and female HcB-8xHcB-23 61
3 Deflection at fracture Femur 38 Male and female HcB-8xHcB-23 61
3 Post-yield deflection Femur 38 Male and female HcB-8xHcB-23 61
3 Yield strain Femur 38 Male and female HcB-8xHcB-23 16
3 Maximum strain Femur 38 Male and female HcB-8xHcB-23 16
3 Yield force Femur 47 Male and female HcB-8xHcB-23 61
3 Yield stress Femur 47 Male and female HcB-8xHcB-23 16
3 Flexural strength Femur 53 Male and female HcB-8xHcB-23 61
3 Post-yield strain Femur 73 Male and female HcB-8xHcB-23 16
3 Flexural strength Femur 85 Female NZBxRF 17
4 Flexural strength Tibia 17.9–79 Female B6xC3H 60
4 Yield force Tibia 48.5 Male and female B6xDBA 14
4 Flexural strength Tibia 48.5 Male and female B6xDBA 14
4 Flexural strength Femur 58 Female B6xC3H 62
4 Polar 2nd moment of area Femur 58 Female B6xC3H 62
4 Fracture energy Femur 58 Female B6xC3H 62
4 Flexural stiffness Femur 58 Female B6xC3H 62
4 Flexural strength Femur 61.9 Male and female B6xDBA 14
4 Flexural strength Femur 65 Female NZBxRF 17
4 Cortical area Femur 65 Female NZBxRF 17
4 Yield force Femur 66 Male and female HcB-8xHcB-23 61
4 Flexural strength Femur 66 Male and female HcB-8xHcB-23 61
4 Flexural stiffness Femur 66 Male and female HcB-8xHcB-23 61
4 Cortical area Femur 66 Male and female HcB-8xHcB-23 61
4 2nd moment of area Femur 66 Male and female HcB-8xHcB-23 61
5 Flexural strength Femur 30 Female NZBxRF 17
5 Fracture energy Femur 40 Female NZBxRF 17
5 Bone width Femur 45 Female NZBxRF 17
5 Cortical area Femur 50 Female NZBxRF 17
5 Bone width Femur 50 Female NZBxRF 17
5 Flexural strength Tibia 60–81 Female B6xC3H 60
6 Flexural stiffness Femur 7 Male and female HcB-8xHcB-23 61
6 2nd moment of area Femur 9 Male and female HcB-8xHcB-23 61
6 Cortical area Femur 10 Male and female HcB-8xHcB-23 61
6 Maximum stress Femur 10 Male and female HcB-8xHcB-23 16
6 Fracture energy Femur 11 Male and female HcB-8xHcB-23 61
6 Flexural strength Femur 12 Male and female HcB-8xHcB-23 61
6 Bone width Femur 30 Female NZBxRF 17
6 Maximum stress Femur 32 Male and female HcB-8xHcB-23 16
6 Flexural strength Tibia 51.5–74 Female B6xC3H 60
7 Flexural stiffness Femur 11 Male and female B6xDBA 14
7 Yield force Femur 25 Male and female B6xDBA 14
7 Flexural strength Femur 25 Male and female B6xDBA 14
7 Flexural strength Femur 30 Female NZBxRF 17
7 Flexural strength Tibia 50–65.4 Female B6xC3H 60
7 Fracture energy Femur 55 Female NZBxRF 17
7 Cortical area Femur 55 Female NZBxRF 17
7 Bone width (ML dimension) Femur 55 Female NZBxRF 17
7 Bone width (AP dimension) Femur 55 Female NZBxRF 17
8 Flexural strength Femur 15.7 Female MRLxSJL 15
8 Flexural strength Femur 64 Female B6xC3H 62
8 Flexural stiffness Femur 64 Female B6xC3H 62
8 Polar 2nd moment of area Femur 64 Female B6xC3H 62
9 Flexural strength Femur 40 Female NZBxRF 17
9 Flexural strength Femur 41.5 Female MRLxSJL 15
9 Fracture energy Femur 50 Female NZBxRF 17
9 Bone width Femur 50 Female NZBxRF 17
9 Yield force Tibia 53 Male and female B6xDBA 14
9 Flexural strength Tibia 53 Male and female B6xDBA 14
9 Flexural strength Tibia 50–71 Female B6xC3H 60
10 Yield force Femur 13 Male and female HcB-8xHcB-23 61
10 Flexural strength Femur 16 Female B6xC3H 62
10 Fracture energy Femur 16 Female B6xC3H 62
10 Post-yield strain Femur 17 Male and female HcB-8xHcB-23 16
10 Flexural toughness Femur 18 Male and female HcB-8xHcB-23 16
10 Post-yield deflection Femur 19 Male and female HcB-8xHcB-23 61
10 Deflection at fracture Femur 21 Male and female HcB-8xHcB-23 61
10 Flexural strength Femur 22 Male and female HcB-8xHcB-23 61
10 Flexural stiffness Femur 22 Male and female HcB-8xHcB-23 61
10 Maximum strain Femur 22 Male and female HcB-8xHcB-23 16
10 Cortical area Femur 24 Male and female HcB-8xHcB-23 61
10 2nd moment of area Femur 28 Male and female HcB-8xHcB-23 61
10 Yield stress Femur 28 Male and female HcB-8xHcB-23 16
10 Maximum stress Femur 37 Male and female HcB-8xHcB-23 16
10 Elastic deflection Femur 41 Male and female HcB-8xHcB-23 61
10 Yield stress Femur 42 Male and female HcB-8xHcB-23 16
10 Flexural strength Femur 50.3 Female MRLxSJL 15
10 Bone width Femur 60 Female NZBxRF 17
11 Elastic modulus Tibia 1.1–71 Female B6xC3H 60
11 Fracture energy Femur 60 Female NZBxRF 17
11 Bone width (ML dimension) Femur 60 Female NZBxRF 17
11 Bone width (AP dimension) Femur 65 Female NZBxRF 17
11 Cortical area Femur 75 Female NZBxRF 17
12 Bone width Femur 0 Female NZBxRF 17
12 Flexural strength Femur 5 Female NZBxRF 17
12 Cortical area Femur 5 Female NZBxRF 17
12 Flexural stiffness Femur 20 Female NZBxRF 17
13 Flexural strength Tibia 9–54 Female B6xC3H 60
13 Fracture energy Femur 56 Female B6xC3H 62
13 Flexural strength Femur 56 Female B6xC3H 62
13 Max:Min 2nd moment of area Femur 56 Female B6xC3H 62
13 Flexural stiffness Femur 56 Female B6xC3H 62
14 Flexural stiffness Femur syntenic Female B6xC3H 62
14 Flexural strength Femur syntenic Female B6xC3H 62
15 Elastic modulus Tibia 9.9–55.5 Female B6xC3H 60
15 Flexural strength Tibia 9.9–55.5 Female B6xC3H 60
16 Flexural strength Tibia 43–56 Female B6xC3H 60
17 Flexural strength Femur 6.6 Female MRLxSJL 15
17 Flexural strength Tibia 4.1–33.5 Female B6xC3H 60
18 Flexural strength Tibia 5–16 Female B6xC3H 60
18 Cortical area Femur 55 Female NZBxRF 17
19 Elastic modulus Tibia 26–55.7 Female B6xC3H 60
Open in a new tab

aAll mechanical integrity phenotypes were derived from long bone 3-point bending (flexure) tests. Mechanical phenotypes include whole bone flexural strength, stiffness and fracture energy. Phenotypes describing long bone size and shape include bone width, cross-sectional cortical area and cross-sectional 2nd moment of area (the geometrical measure of bone strength and stiffness). Phenotypes describing bone matrix (material) mechanical integrity combine the whole bone structural test with cross-sectional 2nd moment of area to estimate yield stress, maximum stress, post-yield strain and elastic modulus (matrix material stiffness).

bAs provided in the literature. When more than one model was calculated, the peak for at the maximum LOD is provided.

cIndicates gender of animals phenotyped. Not specificity of the locus.

Often body weight is used as a corrective factor in these studies, as is also done in the rat studies. Although generally there is a positive correlation between body weight and cortical cross-sectional area across growth within an inbred strain,46 the relationship between cortical area and body weight breaks down when the alleles for body weight and cortical area begin to segregate independently (correlation coefficient=0.521 (ref. 17)). It is unclear what bias this correction may introduce, as correlated traits are not necessarily caused by the same genes.47 Indeed, this relationship between bone size and body size has been exploited in studies aimed at differentiating between bone robustness (total bone area/length), morphological compensation (cortical bone area/body weight) and the degree of bone mineralization (a surrogate for tissue quality). Using a combination of genetic mapping studies together with recombinant inbred lines and phenotyping of consomic mice, it was shown that these traits are independently inherited, demonstrating the complex genetic regulation of what is ultimately considered the strength of bone matrix.13

Given that bone strength is a function of size, shape and geometry, there is interest in determining the genetic control of bone quality and tissue level determinants of bone strength. The protein component of bone is the product of genes, and as such it stands to reason that there may be differences in bone matrix quality that are controlled at the level of genetic differences. This hypothesis is bolstered by the observation of interspecies differences in bone matrix composition. In a study by Courtland et al,48 it was shown that the C57BL/6J, A/J and C3H/HeJ strains of mice have divergent bone matrix composition. To assess these differences from a strength point of view, this group milled a longitudinal section of cortical bone within the femur of a defined dimension and tested this section in tension to measure matrix strength directly. The authors concluded that A/J cortical bone matrix is more stiff and brittle than the other two strains. Similarly, Blank et al.49 showed that there were differences in collagen cross-links and crystallinity when comparing two strains of recombinant congenic strains that also altered the biomechanical performance. Together, these studies highlight that the relationship between size, shape and material properties of bones is complex and that there is much work to be carried out to truly understand fracture susceptibility.

Studies in chickens

Although a number of QTL mapping studies have been conducted that report strength QTL using chicken bones, few of these studies actually conducted strength phenotyping. Rather, these studies examined bone mineral content and/or bone geometry as surrogate measures. Three studies reported direct measurements of bone strength in chickens. The first study crossed the domestic RJ and WL breeds of chickens and tested femurs in torsion. A single significant locus was reported for twist angle at fracture (in degrees) on Chr 20. This locus is difficult to interpret given that no loci were found for torsional strength or stiffness.50 In a second study, a QTL on chicken Chr 1 for long bone flexural strength was mapped in two white Leghorn lines.51 However, bone size and shape were not measured in this study, clouding the interpretation of these results. The third study was substantially larger and involved Cobb–Cobb broilers bred to White Leghorn layers. QTL for tibial strength were reported on Chr 3, 11, 12, 15 and 26.52 The QTL on Chr 3, 11 and 12 co-localized with loci for bone size, as was observed for the rodent studies. In sum, although promising findings have been made using chickens, additional work is required to further validate these results.

Summary and conclusions

In summary, there has been extensive and varied work conducted in an attempt to identify genes that control the strength of long bone diaphyses. Less work has been conducted for more trabeculated sites such as the femoral neck and the vertebral body. A recurring theme from these studies is that QTL for whole bone structural strength co-map for bone size or/and bone shape QTL, a logically obvious and requisite outcome at such macroscopic length scale. As such, these QTL are not likely providing readouts of genetic loci governing bone matrix quality. Although there have been a few studies that have used body weight and/or bone length as putative correction for bone size, there have been few attempts made to account for actual cross-sectional size and geometry in order to refine structural phenotypes into matrix (material) quality phenotypes. Regardless, the limitations of any whole bone strength test must be carefully considered when interpreting these QTL results if the end goal is to determine the mechanism of action and biological processes the gene is acting onto manifest in the phenotype. At the matrix level, FTIR studies show that there are genetic differences in bone matrix composition, suggesting that variations may result in measurable differences in matrix strength. The whole bone mechanical strength tests commonly used in genetic studies may not adequately provide for reduction in structural properties to matrix (material) level properties that are necessary to elucidate differences in bone matrix quality (that is, matrix strength and stiffness). Mechanical test methods that can isolate and test matrix level properties with greater fidelity and precision will be beneficial toward identifying QTL and the underlying specific genes that regulate bone matrix quality. Because of the destructive nature of methods used to test bone material strength, this work will likely largely remain restricted to animal studies; however, if an alternative phenotype measurement technique was developed that could be used in human populations, GWAS in humans remains a possibility for the identification of bone strength genes. Our experience with bone mass phenotypes suggests that any such method would likely have to be applicable to large cohorts. GWAS for fracture risk (the closest clinical phenotype to bone strength) has been highly informative but does suffer from low statistical power issues. However, genes identified in animal studies can be tested in human populations as candidates for surrogate phenotypes such as fracture risk, thereby reducing the impact of statistical multiple testing penalties that must be applied53 and increasing the information that can be extracted from GWAS data. In conclusion, although a great deal of work has been carried out to characterize the genetic control of bone strength, this work is only just the beginning.

Acknowledgments

This publication was made possible by grant number AR060234 (to CLA) from NIAMS/NIH.

Footnotes

The authors declare no conflict of interest.

References

  1. Deng HW, Chen WM, Recker S, Stegman MR, Li JL, Davies KM et al. Genetic determination of Colles' fracture and differential bone mass in women with and without Colles' fracture. J Bone Miner Res 2000; 15: 1243–1252. [DOI] [PubMed] [Google Scholar]
  2. Liu CT, Karasik D, Zhou Y, Hsu YH, Genant HK, Broe KE et al. Heritability of prevalent vertebral fracture and volumetric bone mineral density and geometry at the lumbar spine in three generations of the Framingham study. J Bone Miner Res 2012; 27: 954–958. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. MacGregor A, Snieder H, Spector TD. Genetic factors and osteoporotic fractures in elderly people. Twin data support genetic contribution to risk of fracture. BMJ 2000; 320: 1669–1670. [PMC free article] [PubMed] [Google Scholar]
  4. Michaelsson K, Melhus H, Ferm H, Ahlbom A, Pedersen NL. Genetic liability to fractures in the elderly. Arch Intern Med 2005; 165: 1825–1830. [DOI] [PubMed] [Google Scholar]
  5. Estrada K, Styrkarsdottir U, Evangelou E, Hsu YH, Duncan EL, Ntzani EE et al. Genome-wide meta-analysis identifies 56 bone mineral density loci and reveals 14 loci associated with risk of fracture. Nat Genet 2012; 44: 491–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Hsu YH, Kiel DP. Clinical review: genome-wide association studies of skeletal phenotypes: what we have learned and where we are headed. J Clin Endocrinol Metab 2012; 97: E1958–E1977. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Lee SH, Kang MI, Ahn SH, Lim KH, Lee GE, Shin ES et al. Common and rare variants in the exons and regulatory regions of osteoporosis-related genes improve osteoporotic fracture risk prediction. J Clin Endocrinol Metab 2014; 99: E2400–E2411. [DOI] [PubMed] [Google Scholar]
  8. Oei L, Estrada K, Duncan EL, Christiansen C, Liu CT, Langdahl BL et al. Genome-wide association study for radiographic vertebral fractures: a potential role for the 16q24 BMD locus. Bone 2014; 59: 20–27. [PMC free article] [PubMed] [Google Scholar]
  9. Oei L, Hsu YH, Styrkarsdottir U, Eussen BH, de Klein A, Peters MJ et al. A genome-wide copy number association study of osteoporotic fractures points to the 6p25.1 locus. J Med Genet 2014; 51: 122–131. [DOI] [PubMed] [Google Scholar]
  10. Mouse Genome Sequencing C. Waterston RH, Lindblad-Toh K, Birney E, Rogers J, Abril JF et al. Initial sequencing and comparative analysis of the mouse genome. Nature 2002; 420: 520–562. [DOI] [PubMed] [Google Scholar]
  11. Ackert-Bicknell CL, Karasik D, Li Q, Smith RV, Hsu YH, Churchill GA et al. Mouse BMD quantitative trait loci show improved concordance with human genome-wide association loci when recalculated on a new, common mouse genetic map. J Bone Miner Res 2010; 25: 1808–1820. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Farber CR, Kelly SA, Baruch E, Yu D, Hua K, Nehrenberg DL et al. Identification of quantitative trait loci influencing skeletal architecture in mice: emergence of Cdh11 as a primary candidate gene regulating femoral morphology. J Bone Miner Res 2011; 26: 2174–2183. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Jepsen KJ, Courtland HW, Nadeau JH. Genetically determined phenotype covariation networks control bone strength. J Bone Miner Res 2010; 25: 1581–1593. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Lang DH, Conroy DE, Lionikas A, Mack HA, Larsson L, Vogler GP et al. Bone, muscle and physical activity: structural equation modeling of relationships and genetic influence with age. J Bone Miner Res 2009; 24: 1608–1617. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Li X, Masinde G, Gu W, Wergedal J, Mohan S, Baylink DJ. Genetic dissection of femur breaking strength in a large population (MRL/MpJ x SJL/J) of F2 Mice: single QTL effects, epistasis and pleiotropy. Genomics 2002; 79: 734–740. [DOI] [PubMed] [Google Scholar]
  16. Saless N, Lopez Franco GE, Litscher S, Kattappuram RS, Houlihan MJ, Vanderby R et al. Linkage mapping of femoral material properties in a reciprocal intercross of HcB-8 and HcB-23 recombinant mouse strains. Bone 2010; 46: 1251–1259. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Wergedal JE, Ackert-Bicknell CL, Tsaih SW, Sheng MH, Li R, Mohan S et al. Femur mechanical properties in the F2 progeny of an NZB/B1NJ x RF/J cross are regulated predominantly by genetic loci that regulate bone geometry. J Bone Miner Res 2006; 21: 1256–1266. [DOI] [PubMed] [Google Scholar]
  18. Worley KC, Gunaratne P. Rat Genome (Rattus norvegicus). Reviews in Cell Biology and Molecular Medicine 2006; 10.1002/3527600906.mcb.200500054. [DOI] [Google Scholar]
  19. Blott SC, Swinburne JE, Sibbons C, Fox-Clipsham LY, Helwegen M, Hillyer L et al. A genome-wide association study demonstrates significant genetic variation for fracture risk in Thoroughbred racehorses. BMC Genomics 2014; 15: 147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Welsh CE, Lewis TW, Blott SC, Mellor DJ, Stirk AJ, Parkin TD. Estimates of genetic parameters of distal limb fracture and superficial digital flexor tendon injury in UK Thoroughbred racehorses. Vet J 2014; 200: 253–256. [DOI] [PubMed] [Google Scholar]
  21. Burt DW. Chicken genome: current status and future opportunities. Genome Res 2005; 15: 1692–1698. [DOI] [PubMed] [Google Scholar]
  22. Broman KW. Review of statistical methods for QTL mapping in experimental crosses. Lab Anim (NY) 2001; 30: 44–52. [PubMed] [Google Scholar]
  23. Peters LL, Robledo RF, Bult CJ, Churchill GA, Paigen BJ, Svenson KL. The mouse as a model for human biology: a resource guide for complex trait analysis. Nat Rev Genet 2007; 8: 58–69. [DOI] [PubMed] [Google Scholar]
  24. Ridgway WM, Healy B, Smink LJ, Rainbow D, Wicker LS. New tools for defining the 'genetic background' of inbred mouse strains. Nat Immunol 2007; 8: 669–673. [DOI] [PubMed] [Google Scholar]
  25. Ackert-Bicknell CL, Hibbs MA. The need for mouse models in osteoporosis genetics research. Bonekey Rep 2012; 1: 98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Alam I, Koller DL, Canete T, Blazquez G, Lopez-Aumatell R, Martinez-Membrives E et al. High-resolution genome screen for bone mineral density in heterogeneous stock rat. J Bone Miner Res 2014; 29: 1619–1626. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Flint J, Valdar W, Shifman S, Mott R. Strategies for mapping and cloning quantitative trait genes in rodents. Nat Rev Genet 2005; 6: 271–286. [DOI] [PubMed] [Google Scholar]
  28. Gatti DM, Svenson KL, Shabalin A, Wu LY, Valdar W, Simecek P et al. Quantitative trait locus mapping methods for diversity outbred mice. G3 (Bethesda) 2014; 4: 1623–1633. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Zhang W, Korstanje R, Thaisz J, Staedtler F, Harttman N, Xu L et al. Genome-wide association mapping of quantitative traits in outbred mice. G3 (Bethesda) 2012; 2: 167–174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Silver LM. Mouse Genetics: Concepts and Applications Oxford University Press1995; . [Google Scholar]
  31. van der Meulen MC, Jepsen KJ, Mikic B. Understanding bone strength: size isn't everything. Bone 2001; 29: 101–104. [DOI] [PubMed] [Google Scholar]
  32. Turner CH, Burr DB. Basic biomechanical measurements of bone: a tutorial. Bone 1993; 14: 595–608. [DOI] [PubMed] [Google Scholar]
  33. Bouxsein ML, Boyd SK, Christiansen BA, Guldberg RE, Jepsen KJ, Muller R. Guidelines for assessment of bone microstructure in rodents using micro-computed tomography. J Bone Mineral Res 2010; 25: 1468–1486. [DOI] [PubMed] [Google Scholar]
  34. ASTM Standard D 790-10. Flexural Properties of Unreinforced and Reinforced Plastics and Electrical Insulating Materials. American Society for Testing and Materials: West Conshohocken, Pennsylvania, USA2010; .
  35. van Lenthe GH, Voide R, Boyd SK, Muller R. Tissue modulus calculated from beam theory is biased by bone size and geometry: implications for the use of three-point bending tests to determine bone tissue modulus. Bone 2008; 43: 717–723. [DOI] [PubMed] [Google Scholar]
  36. Alam I, Sun Q, Koller DL, Liu L, Liu Y, Edenberg HJ et al. Differentially expressed genes strongly correlated with femur strength in rats. Genomics 2009; 94: 257–262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Alam I, Sun Q, Liu L, Koller DL, Liu Y, Edenberg HJ et al. Genomic expression analysis of rat chromosome 4 for skeletal traits at femoral neck. Physiol Genomics 2008; 35: 191–196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Alam I, Carr LG, Liang T, Liu Y, Edenberg HJ, Econs MJ et al. Identification of genes influencing skeletal phenotypes in congenic P/NP rats. J Bone Miner Res 2010; 25: 1314–1325. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Beamer WG, Shultz KL, Coombs HF 3rd, DeMambro VE, Reinholdt LG, Ackert-Bicknell CL et al. BMD regulation on mouse distal chromosome 1, candidate genes, and response to ovariectomy or dietary fat. J Bone Miner Res 2011; 26: 88–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. DiPetrillo K, Wang X, Stylianou IM, Paigen B. Bioinformatics toolbox for narrowing rodent quantitative trait loci. Trends Genet 2005; 21: 683–692. [DOI] [PubMed] [Google Scholar]
  41. Tong C, Huang G, Ashton C, Wu H, Yan H, Ying QL. Rapid and cost-effective gene targeting in rat embryonic stem cells by TALENs. J Genet Genomics 2012; 39: 275–280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Rudemiller N, Lund H, Jacob HJ, Geurts AM, Mattson DL, PhysGen Knockout P. CD247 modulates blood pressure by altering T-lymphocyte infiltration in the kidney. Hypertension 2014; 63: 559–564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Solberg Woods LC, Stelloh C, Regner KR, Schwabe T, Eisenhauer J, Garrett MR. Heterogeneous stock rats: a new model to study the genetics of renal phenotypes. Am J Physiol Renal Physiol 2010; 298: F1484–F1491. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Alam I, Koller DL, Sun Q, Roeder RK, Canete T, Blazquez G et al. Heterogeneous stock rat: a unique animal model for mapping genes influencing bone fragility. Bone 2011; 48: 1169–1177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Turner CH, Sun Q, Schriefer J, Pitner N, Price R, Bouxsein ML et al. Congenic mice reveal sex-specific genetic regulation of femoral structure and strength. Calcif Tissue Int 2003; 73: 297–303. [DOI] [PubMed] [Google Scholar]
  46. Jepsen KJ, Hu B, Tommasini SM, Courtland HW, Price C, Cordova M et al. Phenotypic integration of skeletal traits during growth buffers genetic variants affecting the slenderness of femora in inbred mouse strains. Mammalian Genome 2009; 20: 21–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Karasik D, Dupuis J, Cupples LA, Beck TJ, Mahaney MC, Havill LM et al. Bivariate linkage study of proximal hip geometry and body size indices: the Framingham study. Calcif Tissue Int 2007; 81: 162–173. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Courtland HW, Nasser P, Goldstone AB, Spevak L, Boskey AL, Jepsen KJ. Fourier transform infrared imaging microspectroscopy and tissue-level mechanical testing reveal intraspecies variation in mouse bone mineral and matrix composition. Calcif Tissue Int 2008; 83: 342–353. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Blank RD, Baldini TH, Kaufman M, Bailey S, Gupta R, Yershov Y et al. Spectroscopically determined collagen Pyr/deH-DHLNL cross-link ratio and crystallinity indices differ markedly in recombinant congenic mice with divergent calculated bone tissue strength. Connect Tissue Res 2003; 44: 134–142. [DOI] [PubMed] [Google Scholar]
  50. Rubin CJ, Brandstrom H, Wright D, Kerje S, Gunnarsson U, Schutz K et al. Quantitative trait loci for BMD and bone strength in an intercross between domestic and wildtype chickens. J Bone Miner Re 2007; 22: 375–384. [DOI] [PubMed] [Google Scholar]
  51. Dunn IC, Fleming RH, McCormack HA, Morrice D, Burt DW, Preisinger R et al. A QTL for osteoporosis detected in an F2 population derived from White Leghorn chicken lines divergently selected for bone index. Anim Genet 2007; 38: 45–49. [DOI] [PubMed] [Google Scholar]
  52. Schreiweis MA, Hester PY, Moody DE. Identification of quantitative trait loci associated with bone traits and body weight in an F2 resource population of chickens. Genet Sel Evol 2005; 37: 677–698. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Ackert-Bicknell CL, Demissie S, Tsaih SW, Beamer WG, Cupples LA, Paigen BJ et al. Genetic variation in TRPS1 may regulate hip geometry as well as bone mineral density. Bone 2012; 50: 1188–1195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Lagerholm S, Park HB, Luthman H, Grynpas M, McGuigan F, Swanberg M et al. Identification of candidate gene regions in the rat by co-localization of QTLs for bone density, size, structure and strength. PloS One 2011; 6: e22462. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Lagerholm S, Li LS, Jiao H, Park HB, Ohlsson C, Akesson K et al. Genetic regulation of bone traits is influenced by sex and reciprocal cross in F(2) progeny from GK and F344 rats. J Bone Miner Res 2009; 24: 1066–1074. [DOI] [PubMed] [Google Scholar]
  56. Alam I, Sun Q, Liu L, Koller DL, Fishburn T, Carr LG et al. Whole-genome scan for linkage to bone strength and structure in inbred Fischer 344 and Lewis rats. J Bone Miner Res 2005; 20: 1589–1596. [DOI] [PubMed] [Google Scholar]
  57. Sun Q, Alam I, Liu L, Koller DL, Carr LG, Econs MJ et al. Genetic loci affecting bone structure and strength in inbred COP and DA rats. Bone 2008; 42: 547–553. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Alam I, Sun Q, Liu L, Koller DL, Carr LG, Econs MJ et al. Sex-specific genetic loci for femoral neck bone mass and strength identified in inbred COP and DA rats. J Bone Miner Res 2008; 23: 850–859. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Alam I, Sun Q, Liu L, Koller DL, Fishburn T, Carr LG et al. Identification of a quantitative trait locus on rat chromosome 4 that is strongly linked to femoral neck structure and strength. Bone 2006; 39: 93–99. [DOI] [PubMed] [Google Scholar]
  60. Jiao F, Chiu H, Jiao Y, de Rijk WG, Li X, Eckstein EC et al. Quantitative trait loci for tibial bone strength in C57BL/6J and C3H/HeJ inbred strains of mice. J Genet 2010; 89: 21–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Saless N, Litscher SJ, Lopez Franco GE, Houlihan MJ, Sudhakaran S, Raheem KA et al. Quantitative trait loci for biomechanical performance and femoral geometry in an intercross of recombinant congenic mice: restriction of the Bmd7 candidate interval. FASEB J 2009; 23: 2142–2154. [DOI] [PubMed] [Google Scholar]
  62. Koller DL, Schriefer J, Sun Q, Shultz KL, Donahue LR, Rosen CJ et al. Genetic effects for femoral biomechanics, structure, and density in C57BL/6J and C3H/HeJ inbred mouse strains. J Bone Miner Res 2003; 18: 1758–1765. [DOI] [PubMed] [Google Scholar]

Articles from BoneKEy Reports are provided here courtesy of Nature Publishing Group

RESOURCES