Osteogenesis imperfecta (OI) is a heritable connective tissue disorder caused by mutations in one of the two type I procollagen genes, COL1A1 and COL1A2. The resulting phenotype is short stature, bone deformity and numerous lifetime fractures. The oim/oim model mice are the most widely studied mouse model of OI. Oim/oim mice are homozygous for a null mutation in the COL1A2 gene of type I collagen and have significantly reduced femoral biomechanical strength and increased femoral stiffness as well as altered bone mineral composition. Heterozygous mice (+/oim) have a phenotype intermediate to oim/oim and wildtype (Wt) mice. G610C OI model mice represent a new mouse model of OI based on a large Amish population in which 64 individuals all carry the same COL1A2 mutation.

Bone is inherently mechanosensitive, responding and adapting to its mechanical environment. Bone formation occurs in response to high mechanical loads, often changing its geometry to strengthen the skeleton. The largest physiological loads bones typically experience are from muscles with bone strength directly proportional to muscle mass. Myostatin (mstn) is a member of the transforming growth factor-ß (TGF-ß) super family and is a negative regulator of skeletal muscle growth. When the myostatin protein is missing or non-functional, the result is increased muscle growth with a concomitant increase in bone strength. Recently a completely myostatin deficient child was described whose quadriceps muscles were 7.2 standard deviations above normal but without any detrimental health consequences. Myostatin deficient mice have a similar phenotype: increased muscle mass and bone strength compared to Wt mice. Previous studies demonstrated that mice completely deficient for myostatin (mstn/mstn) have a 40% increase quadriceps muscle mass, an 11% increase in bone mineral density, and a 20% increase in radial ultimate force relative to Wt mice. When mstn/mstn mice were subject to treadmill exercise, radial ultimate force improved by an additional 17%. Exercise has also been shown to be beneficial for human patients at an increased risk for fracture (i.e. osteoporosis, osteogenesis imperfecta). Based on these previous reports, this project was designed to determine if a combination of exercise and myostatin haploinsufficiency could improve the bone strength in mice with compromised bone (+/oim and G610C OI). The aims of this project are threefold: 1) Determine if +/oim and G610C OI mice are able to tolerate and respond to treadmill exercise, 2) Determine if myostatin haploinsufficiency will ameliorate the bone phenotype of +/oim and G610C OI mice and 3) Determine if a combination of exercise coupled with myostatin haploinsufficiency will increase muscle mass and bone strength in +/oim and G610C OI mice. The long term goal of this project is to determine if a combination of these two therapies may be additive, further improving the bone phenotype beyond either therapy alone. This combination would be represent a novel treatment approach for OI patients, astronauts returning from microgravity, bed rest patients and others with compromised bone.

Preliminary data demonstrates the reduction in bone strength seen in female +/oim animals was partially rescued when those mice either underwent treadmill exercise or were also haploinsufficient for the myostatin protein (+/mstn). Treadmill exercise did not appear to impact femoral geometry. However, modest improvements were seen in femoral torsional ultimate strength and energy to failure in both Wt and +/oim mice, though these improvements were not significant. Treadmill exercise did significantly reduce whole bone and material stiffness in both Wt and +/oim mice. Curiously, treadmill exercise did not appear to be beneficial for male Wt or +/oim mice. Myostatin haploinsufficiency marginally improved femoral torsional ultimate strength and energy to failure in +/oim mice, though not significantly. Myostatin haploinsufficiency did significantly reduce whole bone and material stiffness in +/oim mice. Taken together, these data indicate that both treadmill exercise and myostatin deficiency have the potential to improve bone strength and reduce stiffness in +/oim mice with compromised bone.

In order to complete this study, additional female +/oim mice will be evaluated to asses the impact of myostatin haploinsufficiency on both muscle size and bone strength and stiffness. Additionally, the same studies will be performed on male +/oim mice to determine if a gender difference exists in the ability of the bone to respond to myostatin haploinsufficiency. Male and female G610C OI mice will also be evaluated for their ability to respond to treadmill exercise. The potential impact of maternal myostatin haploinsufficiency on the muscle size and bone strength of the offspring will also be evaluated. Finally, +/oim mice of both genders that are also haploinsufficient for myostatin will be subjected to treadmill exercise to determine if exercise coupled with myostatin haploinsufficiency further improves bone strength in +/oim mice.

Toward Aim 2, we are currently analyzing femurs collected from female and male +/oim mice that were also haploinsufficient for the myostatin protein to determine if reduction of myostatin protein will help ameliorate the bone phenotype of +/oim mice. Additional mice are needed to complete this study and those mice will be ready for analyses in the next few months. Following completion of these two aims, we will proceed to aims 4 and 5 and begin breeding the G610C OI model to mice haploinsufficient for the myostatin protein and subjecting these mice, as well as +/oim mice also haploinsufficient for myostatin, to treadmill exercise.

Lastly, toward Aim 3, we are currently analyzing femora collected from Wt, +/oim, +/mstn and +/mstn +/oim mice born to mothers which were also haploinsufficient for the myostatin protein. The data from these femora is being compared to data from femora collected from Wt, +/oim, +/mstn and +/mstn +/oim mice born to mothers which were not haploinsufficient for the myostatin protein to determine if a maternal myostatin allele status impacts the muscle mass and bone strength in the offspring as has been reported in the literature.

J Bone Miner Res. 2009:24(Suppl 1). Available at http://www.asbmr.org/Meetings/AnnualMeeting/AbstractDetail.aspx?aid=c7b434f9-56bd-4bba-8857-70f80726570b , Sep-2009

Myostatin is a member of the transforming growth factor ß (TGF-ß) superfamily and is a negative regulator of skeletal muscle growth. When myostatin is missing or non-functional, the result is uncontrolled muscle growth with a concomitant increase in bone strength. When mice lacking the myostatin gene (mstn/mstn) were subjected to running on a treadmill, their bone strength increased above non-exercised mstn/mstn mice. Exercise has also been shown to be beneficial for human patients at an increased risk for fracture (i.e. osteoporosis, osteogenesis imperfecta). Based on these previous reports, this project was designed to determine if a combination of myostatin inhibition and exercise could improve the bone strength in mice with compromised bone (G610C). The aims of this project are twofold: 1) Determine if a reduction and/or complete absence of the myostatin protein will ameliorate the bone phenotype of heterozygote and homozygote G610C mice and 2) Determine if a combination of exercise coupled with reduction and/or absence of myostatin will increase muscle mass and bone strength in heterozygote and homozygote G610C mice.

The impact of myostatin inhibition on the geometry and strength of bone has been previously reported. However, a comparison of the gender-specific response of the bone to the loss of the myostatin protein has not been reported and, therefore, was not originally a part of this proposal. However, as the long term goal of this project is to determine if myostatin inhibition is a viable candidate for a pharmacological therapy to improve bone strength, it is imperative to ascertain if a gender difference exists in the response of the bone to myostatin deficiency. Data from male and female mice were analyzed separately and interactions between gender and specific bone parameters were found. This data is the first time a gender-specific response to myostatin deficiency has been reported and adds another dimension to what is known about how the loss of myostatin impacts bone strength. It also necessitates that all future analyses take this gender difference into account and data generated from male mice be analyzed separately from data generated from female mice. Additionally, after the initiation of this project, it was reported that the myostatin genotype of the mother affects the muscle weight of the offspring. Wildtype pups born to mstn/+ mothers had increased muscle weights as compared to those born to wildtype mothers. The same was true for mstn/+ pups. At four months of age, femurs and tibiae from male and female wildtype (Wt), heterozygote (mstn/+) and homozygote (mstn/mstn) myostatin knock-out were removed along with several muscles. Total body mass was found to be greater in mstn/mstn mice as compared to Wt. Males were also larger than females in all genotypes. Muscle mass was larger in mstn/mstn mice than in Wt in both males and females. Gender also impacted muscle mass with males having larger muscles masses than females. Geometry was measured via CT analysis which demonstrated gender differences in all parameters measured, including polar moment of area, with males having longer and wider bones. Additionally, femurs from male mstn/mstn mice had a larger polar moment of area then femurs from either wt or mstn/+ mice. Torsional loading to failure also demonstrated gender differences in whole bone strength and stiffness with males having stronger and stiffer bones than females. Additionally, mstn/mstn males were also significantly stronger and stiffer than male wt or mstn/+ mice. Bone material properties were not different between the genotypes in either male or female mice. Taken together, these data suggest that myostatin inhibition impacts femoral geometry and increases whole bone strength and stiffness without impacting bone material in a gender-specific manner. Due to the complex nature of these analyses with the addition of gender differences and allele status of the mother, we have revised our original breeding scheme. We are currently breeding mstn/+ mice with G610C/+ mice to generate mice of four genotypes: wildtype, mstn/+ mice, G610C/+ mice and mstn/+ G610C/+ mice. Two types of breeding pairs have been set up: 1) mstn/+ mother and G610C/+ father and 2) G610C/+ mother and mstn/+ father. Offspring from these crosses will be divided into two groups, controls and exercisers. Both hindlimb muscles (soleus, plantaris, tibialis anterialis and gastrocnemius) and bones from both groups will be analyzed. Femora and tibiae will be analyzed via CT analysis and torsional loading to failure to determine bone geometry and strength. Muscles will be analyzed for contractile generating capacity, gross pathology and cross-sectional area. Collagen content will also be measured using the hydroxyproline assay in the femora from both groups. Data will be separated based on both gender and maternal myostatin allele status.

Additionally, after the initiation of this project, it was reported that the myostatin genotype of the mother affects the muscle weight of the offspring, although the impact on bone strength was not reported. Wildtype and mstn/+ pups born to mstn/+ mothers had increased muscle weights as compared to those born to wildtype mothers. In order to address this, we revised our breeding scheme to generate offspring that will allow us to 1) finish collecting baseline data on the myostatin mice, 2) collect baseline data on the G610C mice and 3) generate the double heterozygote mice (mstn/+ G610C/+) needed to test our hypothesis. The double heterozygote mice will be generated by crossing female mstn/+ with male G610C/+ mice as well as by crossing male mstn/+ with female G610C/+ mice. These crosses will allow us to determine if there is a maternal effect of the myostatin gene on the bone strength of the offspring. We are currently collecting muscles and bone from the offspring of these breeding pairs for analysis.

Matrix Biology 2008 Dec;27(Suppl 1):24. http://dx.doi.org/10.1016/j.matbio.2008.09.277 , Dec-2008

American Society of Human Genetics, November 2008. Available at the following URL: http://www.ashg.org/2008meeting/abstracts/fulltext/ , Nov-2008

Exercise is proposed to induce changes in bone via the mechanostat theory, which describes a mechanism by which bones respond to changes in loading (often by increased muscle mass) with alterations to bone architecture and strength. Changes in loading can be positive, as in the case of increased exercise, or negative, as in the case of astronauts returning from extended exposure to a weightless environment. If the increased loading is moderate and sustained, it can induce positive changes in bone mass or result in a stronger bone. The opposite effect occurs in the absence of gravity as seen in space, resulting in reduced stress on the bone, decreasing collagen secretion and ultimately producing a feedback loop reducing collagen synthesis and bone formation. The myostatin knock-out mouse carries a mutation that renders the myostatin gene, (a negative regulator of muscle growth) nonfunctional, resulting in increased muscle mass. These animals have been shown to have increased bone strength as compared to their wild-type littermates, presumably due to the increased mechanical stimuli put on the bone as a result of the increased musculature. When these animals were subjected to weight-bearing exercise, their bone strength increased substantially above the non-exercised knock-outs.

The G610C COL1A2 mouse model of osteogenesis imperfecta (OI) is based on a large human population identified through an osteoporosis screen. This model has reduced bone biomechanical integrity due to a glycine to cysteine substitution in the a2(I) chain of the type I collagen molecule leading to osteopenia and reduced bone biomechanical strength. Studies on this model are applicable to other diseases of bone loss, such as osteoporosis, and present a unique opportunity to study mechanisms to improve bone strength. We will cross the myostatin knock-out and the G610C COL1A2 mouse models to study the ability of increased muscle mass to stimulate bone formation in osteoporotic-like bone. Additionally, we will subject the myostatin/G610C animals to both non-weight-bearing (swimming) and weight-bearing exercise (running on a treadmill). Preliminary results of both exercise types in another, more severe mouse model of OI suggest that exercise positively impacts both muscle and bone. These animals also demonstrated improved bone strength as seen by increased torsional ultimate strength, tensile strength and energy to failure. Taken together, these studies indicate that exercise induces muscle hypertrophy and that larger muscles may be sufficient to heighten mechanical stimuli on the bone, thereby increasing bone biomechanical integrity.

We hypothesize that by exercising the myostatin knock-out/G610C mouse, we will substantially increase the muscle mass of these animals. The increased muscle mass should place added stress on the bones and, therefore, increase the biomechanical strength of the bones.