Portal de Periódicos da CAPES

Lim Mineralization Protein-1 Knockout Mice Have Reduced Spine Trabecular Bone Density on Microcomputed Tomography due to Decreased Bone Morphogenetic Protein Responsiveness

2014; Lippincott Williams & Wilkins; Volume: 61; Issue: Supplement 1 Linguagem: Inglês

10.1227/neu.0000000000000414

1524-4040

Matthew F. Gary, Manjula Viggeswarapu, Colleen Oliver, Mesfin Teklemariam, Sreedhara Sangadala, Louisa Titus, Scott D. Boden,

Bone Metabolism and Diseases

Bone morphogenetic protein (BMP)-2 has been shown to be a potent enhancer of spine fusion, but its role in osteoporosis has not yet been defined. The number of vertebral compression fractures related to osteoporosis is expected to double by 2025, and the total cost of osteoporotic fractures is expected to reach $228 billion for 2016 to 2025.1 An animal model with decreased global bone density and decreased responsiveness to BMPs could prove instrumental in identifying and testing novel therapeutics for this disease process. Because of BMP-2's critical role in development, attempted knockout (KO) of the BMP-2 gene has proven embryonically lethal. Thus, studying the loss of function of BMP-2 with the use of a transgenic approach will require a different strategy. Lim mineralization protein-1 (LMP) is an intracellular protein that enhances cellular responsiveness to the BMP pathway and plays a critical role in osteoblast differentiation.2,3,6-11 Canonical BMP-2 signaling is reliant on Smad proteins, which are the key intracellular signal transduction proteins. LMP exerts its effect by binding to and inhibiting the E3 ligase, Smurf1, blocking its normal function of facilitating degradation of Smads 1/5/8.3 A synthetic small molecule mimicking the active site of LMP has recently been shown to potentiate the transdifferentiation of mouse myoblasts toward the osteoblastic phenotype.4 Conversely, preliminary data from our laboratory demonstrated that inhibiting the LMP protein with small interfering (si) RNA blocked osteoblast differentiation (Figure 1).FIGURE 1: LMP-1 siRNA inhibited LMP-1 expression and blocked bone mineralization in MC3T3-E1 cells. The insets demonstrate reduced bone mineralization. LMP-1, Lim mineralization protein-1; si, small interfering.To model globally decreased BMP responsiveness, we generated an LMP KO mouse and hypothesized it would have decreased bone density. METHODS Knockout Mice Global Pdlim7 (mouse equivalent of LMP) KO embryonic stem cells were developed by the European Conditional Mouse Mutagenesis Program by using gene-trapping techniques. We purchased the Pdlim7 KO embryonic stem cells and introduced them into blastocysts. Twenty-six female mice were injected with these blastocysts, yielding 2 pregnant females. These 2 females gave birth to 9 viable chimeric pups. By utilizing well-established breeding techniques, a colony of Pdlim7 wild-type (WT), heterozygous (het), and KO mice was obtained. Genotype was confirmed by utilizing polymerase chain reaction of tail snips. The Veterans Affairs Institutional Animal Care and Use Committee approved all animal protocols. The mice were housed 4 to 5 mice/cage and maintained on a 12:12 h light-dark cycle with access to water and a standard rodent diet in a sterile ventilated cage. Once the mice reached 18 or 26 weeks of age, they were anesthetized with isoflurane for cardiac puncture to obtain serum and then euthanized in a CO2 chamber. Marrow-Derived Stromal Cells The left femur was resected en bloc, and all soft tissue was removed under sterile conditions. The femurs were then cut in half and the bone marrow collected via centrifugation. The subsequent bone marrow pellet was resuspended in 6 mL of culture medium (α-minimal essential medium + 10% horse serum +10% fetal bovine serum + antibiotics/antifungal). The total number of viable cells obtained from each specimen was counted utilizing trypan blue staining and a hemocytometer. We then seeded 24-well plates at 1 million cells per well. After 4 days of incubation, media were changed to osteogenic media (above culture media + ascorbic acid and beta-glycerophosphate). On day 21, the wells were treated with von Kossa silver staining, and a semiautomated computerized video image analysis system (Optomax 5; Optomax, Hollis, New Hampshire) was used to quantitate the area of mineralization in each well. Microcomputed Tomography The right femur and the entire spine were resected en bloc and stored in 70% ethanol. They were then imaged by using microcomputed tomography (μCT-40; Scanco Medical AG, Bassersdorf, Switzerland) with a resolution of 6 µm in all 3 spatial dimensions. The metaphysis of the femurs were scanned starting at the inferior growth plate and extending into the shaft proximally in 6-μm sections for a total of 238 slices. From this volume, 100 slices above the growth plate were selected for trabecular evaluations. The trabecular bone was manually contoured every 8 slices, and the intervening slices were contoured by utilizing the morph function. A second image was obtained at the mid-diaphysis to assess cortical bone structure. This scan was also performed at a resolution of 6 μm for a total of 99 slices. All 99 slices were utilized for the cortical evaluations. Trabecular bone density (BV/TV), trabecular thickness, and trabecular number were calculated with the direct distance transformation method.5 The L3 spinal vertebrae from each sample were scanned in 6-µm sections. The trabecular contours were automatically performed after assigning 2 borders (Figure 2).FIGURE 2: Microcomputed tomography of the L3 vertebral body. Note the manually drawn region of interest.Serum Chemistry The serum from 26-week WT (n = 12), het (n = 20), and KO (n = 6) mice were pooled in each group and sent to the Yerkes primate research laboratory for routine serum chemistry measurements. Statistical Analyses Data were analyzed using the Sigma Plot 12 software (Systat Software Inc, San Jose, California), and statistical significance was determined using 1-way/2-way analysis of variance. RESULTS Biometrics The 18-week LMP KO mice weighed less and were shorter in length than their WT counter parts, P = .001 (Figure 3). All of the mice had normal serum chemistries.FIGURE 3: Weight and height difference between LMP WT, het, and KO mice. *P < .05. LMP, Lim mineralization protein-1; WT, wild type; het, heterozygous; KO, knockout.Femur Analysis In the femur studies, 24 WT, 40 het, and 22 KO mice were analyzed at 26 weeks of age, while 25 WT, 32 het, and 19 KO mice were analyzed at 18 weeks of age. In the 18-week female KO mice, there was a significant decrease in trabecular bone volume of 25% (P = .04) and trabecular thickness of 11% (P = .02); there was a trend for decreased BV/TV of 23% (P = .06) and cortical volume of 6% (P = .06, Figure 4). In the 26-week female KO mice, there was a decrease in the trabecular bone volume of 45% (P = .001), BV/TV of 41% (P = .001), and cortical volume of 8% (P = .001, Figures 5 and 6). In the male KO mice, there was no statistically significant difference in any of the trabecular or cortical analyses at 18 or 26 weeks.FIGURE 4: Percent difference in femur trabecular bone and cortical bone structure of female 18-week-old KO mice. LMP, Lim mineralization protein-1; WT, wild type; KO, knockout; TV, total volume; BV, trabecular bone volume; BV/TV, trabecular bone density; Tb.N, trabecular number; Tb.Th, trabecular thickness; CV, cortical volume; CP, cortical porosity. <*P < .05, <**P < .1.FIGURE 5: Percent difference in femur trabecular bone and cortical bone structure of female 26-week-old KO mice. LMP, Lim mineralization protein-1; WT, wild type; KO, knockout; TV, total volume; BV, trabecular bone volume; BV/TV, trabecular bone density; Tb.N, trabecular number; Tb.Th, trabecular thickness; CV, cortical volume; CP, cortical porosity. <*P <.05.FIGURE 6: Note the decreased trabecular bone density in the LMP-deficient mice on microcomputed tomography. LMP, Lim mineralization protein-1; WT, wild type.Spine Analysis In the spine studies, 23 WT and 21 KO mice were analyzed at 26 weeks of age. Vertebral body trabecular bone density decreased 25% (P = .001) and trabecular thickness decreased 8% (P = .005) in the female KO mice (Figure 7). There was no significant difference in the male KO spine trabecular bone volumes.FIGURE 7: Percent difference in L3 vertebral body structure of female 26-week-old KO mice. LMP, Lim mineralization protein-1; WT, wild type; KO, knockout; TV, total volume; BV, trabecular bone volume; BV/TV, trabecular bone density; Tb.N, trabecular number; Tb.Th, trabecular thickness. *P < .05.Marrow Derived Stromal Cells The marrow stromal cells from the het and KO mice formed a reduced area of mineralization (Figure 8).FIGURE 8: Area of mineralization for MSCs obtained from 18-week mice. Cells were grown in osteogenic media for 21 days. LMP, Lim mineralization protein-1; WT, wild type; MSC, marrow stromal cells. *P < .05.DISCUSSION At the turn of the millennium, BMP-2 was projected to be a boon for patient care. Spine surgeons were excited to have a product that limited the need for autologous iliac crest and decreased the rate of pseudoarthrosis. Unfortunately, we have discovered that the massive doses required for consistent bone formation in humans also result in unintended side effects and make it prohibitively expensive. In addition, our understanding of BMPs to date has focused primarily on its role in local bone homeostasis. The question remains as to whether BMPs play a role in systemic bone homeostasis and the development of osteoporosis. The goals of this study were to (1) create an organism with global decreased responsiveness to BMPs to study novel therapeutics, (2) determine the global role of BMPs on bone homeostasis, and (3) develop an animal model for BMP-hyporesponsivity-induced osteoporosis. The utilization of BMP-2 during spine fusions has dropped precipitously because of its high cost and ambiguous side-effect profile. As an alternative, some surgeons are reverting back to iliac crest grafts, utilizing bone marrow aspirate, or not employing any osteoinductive agents. Certainly this will lead to increased rates of pseudoarthrosis. Novel osteoinductive agents with low cost, low side effects, and easy administration are desperately needed. One approach would be to manipulate the downstream regulators of BMPs, thereby altering cellular responsiveness. Such molecules are under development.4 The LMP KO model would be ideal for testing these novel therapeutics intended to increase BMP responsiveness. Bone is constantly turning over. The catabolism and anabolism of bone must be perfectly balanced between the bone-forming cells, osteoblasts, and the bone-resorbing cells, osteoclasts. Tipping the scale in favor of one over another could drastically alter this delicate balance. It is unclear what role, if any, BMP and its signaling cascade play in systemic bone homeostasis. By creating a mouse with global decreased BMP responsiveness, we can begin to elucidate this role. The trabecular bone density of the mice at 18 weeks was decreased by 28% in the KO, and by 26 weeks it was decreased by 45% in the KO. These data suggest that the LMP-deficient mice are losing more bone as they age than their WT counterparts. The scale has been tipped toward catabolism. The repercussions of this change on the organism as a whole will allow us to better understand diseases, such as HIV and sickle cell, which cause increased bone catabolism. CONCLUSION We present here the development of the first LMP KO mouse. We conclude that LMP is an important determinant of weight and length, female bone density, and marrow stromal cell propensity for mineralization. As the number of osteoporotic fractures is expected to double by 2025, we speculate that these mice will be useful in discovering and testing novel therapeutics to improve BMP-2 responsiveness and prevent vertebral fractures. Disclosures Emory has licensed rights to LMP and may derive royalties in the future of which a portion would go to Dr Scott Boden as a coinventor. The other authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. Acknowledgments This work was performed at the Atlanta VA Medical Center. This material was supported by the Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development and Emory University Department of Orthopaedics. We thank Maggie Bargouti for her supporting role.