ebook img

the wnt co-receptor lrp5 is essential for skeletal mechanotransduction, but not for the anabolic bone PDF

22 Pages·2006·1.41 MB·English
by  
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview the wnt co-receptor lrp5 is essential for skeletal mechanotransduction, but not for the anabolic bone

JBC Papers in Press. Published on June 20, 2006 as Manuscript M601000200 The latest version is at http://www.jbc.org/cgi/doi/10.1074/jbc.M601000200 THE WNT CO-RECEPTOR LRP5 IS ESSENTIAL FOR SKELETAL MECHANOTRANSDUCTION, BUT NOT FOR THE ANABOLIC BONE RESPONSE TO PARATHYROID HORMONE TREATMENT Kimihiko Sawakami1, Alexander G. Robling2, Minrong Ai3, Nathaniel D. Pitner1, Dawei Liu1, Stuart J. Warden1, Jiliang Li2, Peter Maye4, David W. Rowe4, Randall L. Duncan1, Matthew L. Warman3, and Charles H. Turner1 1Department of Orthopaedic Surgery, Biomechanics and Biomaterials Research Center, Indiana University School of Medicine, Indianapolis, IN, USA; 2Department of Anatomy & Cell Biology, Indiana University School of Medicine, Indianapolis, IN, USA; 3Howard Hughes Medical Institute and Department of Genetics and Center for Human Genetics, Case School of Medicine and University Hospitals of Cleveland, Cleveland, OH, USA; 4Department of Genetics and Developmental Biology, University of Connecticut Health Center, Farmington, CT, USA Running Title: Lrp5 in Bone Anabolic Pathways Keywords: LRP5, osteoporosis pseudoglioma, parathyroid hormone, anabolic treatments, mechanotransduction, bone mass, osteoblast Address Correspondence to: Charles H. Turner, PhD, Director of Orthopaedic Research, Indiana University School of Medicine, 1120 South Drive, FH 115, Indianapolis, IN 46202, USA; Tel: (317) 274- 3226; Fax: (317) 274-3702; email: [email protected] D The cell surface receptor, low-density lipoprotein Low-density lipoprotein receptor-related protein 5 ow n receptor-related protein 5 (LRP5) is a key regulator (LRP5) is a single-pass transmembrane protein lo a of bone mass. Loss-of-function mutations in LRP5 that functions as a co-receptor for the secreted de d cause the human skeletal disease osteoporosis- family of Wnt glycoproteins (1). Wnt signaling, fro pseudoglioma syndrome (OPPG), an autosomal m particularly through LRP5, is emerging as a key h rreedceuscseidv eb ondeis moradsesr ancdh asrtraecntegrtihz.e dW eb iyn vessetivgearteeldy pathway in the regulation of bone mass and ttp://w the role of LRP5 on bone strength using mice strength. For example, the autosomal recessive ww engineered with a loss-of-function mutation in the human disease osteoporosis-pseudoglioma .jb c gene. We then tested whether the osteogenic syndrome (OPPG) is caused by loss-of-function .o rg response to mechanical loading was affected by mutations in LRP5 (2). Patients with OPPG b/ y the loss of Lrp5 signaling. present with bone mineral densities (BMD) several g u Lrp5-null (Lrp5-/-) mice exhibited significantly standard deviations below the mean and are prone es lower bone mineral density and decreased to skeletal fracture and deformity. Interestingly, t on A slLotrarped5ni-n/g- gth m.o if c Tethh, ee youesltnt eaoo wgsteaensoi cbr leardesust cpeordenc sbreuy i tto8m 8me nteotc h9aa9nn%dic /oianrl hheatveer ozmygeoauns BcaMrrDiesr s ionf ltohses -oof-sftuenocptoiorno timc utraatniognes, pril 3, 2 indicating a dose-dependent effect of LRP5 01 activation at mechanically strained surfaces was 9 function (2). Conversely, single amino acid normal. Subsequent experiments demonstrated an inability of Lrp5-/-osteoblasts to synthesize the missense mutations in LRP5, that alter the bone matrix protein osteopontin after a mechanical receptor’s ability to be regulated by endogenous stimulus. inhibitors (3-5), segregate with an abnormally high We then tested whether Lrp5-/- mice increased bone mass phenotypes (HBM) in an autosomal bone formation in response to intermittent dominant manner (6-8). Affected individuals have parathyroid hormone (PTH), a known anabolic BMD values that are several standard deviations treatment. A 4-wk course of intermittent PTH (40 above the mean and have increased bone strength. µg/kg/day; 5 days/week) enhanced skeletal mass In addition to studies in humans, mice have equally in Lrp5-/- and Lrp5+/+ mice, suggesting that been created with loss-of-function mutations in the the anabolic effects of PTH do not require Lrp5 mouse ortholog of LRP5, called Lrp5 (9-11). These signaling. We conclude that Lrp5 is critical for mice recapitulate the clinical features observed in mechanotransduction in osteoblasts. Lrp5 is a mediator of mature osteoblast function following OPPG patients, suggesting that the mouse is a loading. Our data suggest an important component useful animal model for delineating the role of Lrp5 of the skeletal fragility phenotype in individuals in the mammalian skeleton (9-11). Additionally, affected with OPPG is inadequate processing of transgenic mice that overexpress wild-type Lrp5 or signals derived from mechanical stimulation and a HBM-causing missense allele of LRP5 (G171V) that PTH might be an effective treatment for under control of the type I collagen promoter, have improving bone mass in these patients. 1 Copyright 2006 by The American Society for Biochemistry and Molecular Biology, Inc. increased bone mass and skeletal strength (12). mechanotransduction signaling cascade, such as Taken together, these data indicate that LRP5 has matrix production, rather than the early sensing of an important role in determining skeletal mass, mechanical strain. Our data indicate that Lrp5 strength, and function. signaling is not needed to respond to PTH treatment, but that Lrp5 is important for the Though loss-of-function mutations in LRP5 skeleton’s ability to respond to mechanical loading. impart clear deficiencies on the skeleton, it is unclear how LRP5 participates in the modulation of bone mass. The striking similarity between the Experimental Procedures OPPG skeletal phenotype and a mechanical Animals – Lrp5-deficient mice (Lrp5-/-) were disuse phenotype—whereby exposure of the generated as reported previously (20). Briefly, the skeleton to normal mechanical stresses and mice were created on a 129S/J background strain strains is limited—suggests that LRP5 might be by disrupting exons 7 and 8 with a Lac-Z/Neo gene involved in mechanotransduction signaling. For cassette. When correctly targeted, this allele does example, disuse induces a loss of trabecular bone not express Lrp5 mRNA (Fig. 1), and produces no volume (13,14), a reduction in periosteal bone functional Lrp5 receptor, or receptor fragments. apposition (particularly when disuse occurs during We interbred mice that were heterozygous carriers growth) (15), and an increase in endocortical bone of this mutation and obtained wild type (Lrp5+/+), loss (particularly when disuse occurs during heterozygous (Lrp5+/-), and homozygous mutant adulthood) (16). Biopsies and radiographic (Lrp5-/-) offspring in the expected Mendelian genetic findings from patients with OPPG reveal reduced frequencies. To study osteoblast recruitment to D o trabecular bone volume and reduced periosteal w mechanically strained surfaces after loading (see n expansion (2,17). The reduced trabecular bone lo below), we bred the Lrp5 mutant allele onto the a d vOoPluPmGe pisa tmieanitnst,a winheidc ha ti sa cnoomrmmaol ntulyr noobvseer rvraetde iinn transgenic pOBCol3.6GFP mouse strain, in which ed fro green fluorescent protein (GFP) expression is m long-term disuse conditions (15,16). Moreover, h driven by the 3.6 kb rat type I collagen promoter ttp mechanical stimulation of cultured osteoblasts fragment. The pOBCol3.6GFP mice exhibit strong ://w causes translocation of beta-catenin to the w expression of GFP in preosteoblasts and w nucleus (18) and activation of a T cell factor (TCF) osteoblasts, with minimal to no GFP expression in .jb c responsive promoter (19), suggesting that .o other cell types (21). All procedures performed in rg mechanical loading activates canonical Wnt the experiments were approved by the Institutional by/ signaling. g Animal Care and Use Committee guidelines where u e s We undertook an investigation of the role of the animals were raised and studied. t o n Lrp5 in mechanical signaling. We hypothesized A Longitudinal in vivo peripheral dual-energy X-ray p that Lrp5 modulates bone mass, size, and absorptiometry (pixiMUS): Bone mineral content ril 3 strength, and that one of the modes of action , 2 (BMC) of the whole body, spine, and femoral 0 1 through which Lrp5 exerts its effects is 9 diaphysis were evaluated in vivo using peripheral mechanotransduction. To test our hypothesis, we dual-energy X-ray absorptiometry (pDXA; PIXImus generated knockout mice in which Lrp5 was II; GE-Lunar Corp., Madison, WI, USA). Mice were inactivated, and we subjected adult Lrp5-/- mice to anesthetized via inhalation of 2.5% isoflurane axial loading of the right ulna to stimulate (IsoFlo; Abbott Laboratories, North Chicago, IL, osteogenesis in vivo. A nearly complete USA) mixed with O (1.5 L/min) for total ~8 min, 2 obliteration of an osteogenic response to in vivo including both induction and scanning. The mice loading was found in Lrp5-/- mice. We then were placed in prone position on a specimen tray investigated the role of Lrp5 in modulating the and scanned. The head was excluded from total anabolic response to parathyroid hormone (PTH) body scans. The region of interest for the spine treatment and found that PTH remained equally included from the first lumbar vertebra (L1) to fifth effective in enhancing bone mass in Lrp5-/- and lumbar vertebra (L5). The region of interest for the Lrp5+/+ mice. The defect in mechanically-induced femur included the central 50% of the whole femur. osteogenesis was further investigated using an We also analyzed hindlimb properties by early osteoblast reporter mouse strain, which positioning the region of interest box to include all demonstrated that in vivo periosteal osteoblast skeletal tissue distal to the acetabulum. Scans recruitment/activation was not affected by Lrp5 were performed at 4, 8, 12, and 16 wks of age (for deficiency. Studies of Lrp5-/- and Lrp5+/+ primary the baseline phenotype characterization) or weekly osteoblasts ex vivo also indicated that loss of Lrp5 from12 to 16 wks of age (for mice in PTH signaling appears to affect later stages in the 2 experiments; see below). All BMC measures Each femur was positioned in a plastic tube filled were normalized by body weight (BW) to eliminate with 70% ethanol and centered in the gantry of a the confounding effects on changing body size Norland Stratec XCT Research SA+ pQCT (Stratec and weight during growth. Electronics, Pforzheim, Germany). Using a collimation of 0.26 mm and a voxel size of 0.07 Microcomputed tomography (µCT): Geometric mm, three slices through the distal femur (15%, properties of femoral mid-diaphysis, trabecular 17.5%, and 20% of the femoral length measured bone volume fraction and microarchitecture in the from the distal end of the femur) were recorded and femoral distal metaphysis and L5 were evaluated averaged to obtain bone mineral measurements using high-resolution desktop microcomputed from a trabecular site. A single slice through the tomography imaging systems (µCT-20; Scanco midshaft was also collected to monitor a cortical Medical AG, Basserdorf, Switzerland). For site. geometric properties of femoral shaft, a single Biomechanical testing: Mechanical properties of transverse slice through the mid-diaphysis was the femur and L5 vertebra were tested as taken at 9-µm resolution. Each mid-diaphysis slice previously described (22). Briefly, femurs were was imported into Scion Image v4.0.2 (Scion brought to room temperature slowly (~2 h) in a Corporation, Frederick, MD, USA), in which the saline bath and tested at the mid-diaphysis by geometric properties were calculated using three-point bending using a microforce materials standard and customized macros. Geometric properties included cortical area (mm2), and the testing machine (Vitrodyne V1000; Liveco, Inc., maximum (I , mm4) and minimum (I , mm4) Burlington, VT, USA). Load was applied in the MAX MIN D anteroposterior (AP) direction midway between two o cross-sectional moments of inertia (CSMI). In w addition, we calculated polar moment of inertia I supports positioned 9 mm apart. Tests were nlo athse thcea psaucmit yo fo IfM AaX abneda mIM IN(i.n T hthei sC ScaMsIe ,e sati mbaotneesP cdounridnugc tewd haicth a cfororcses heaandd spedeisdp laocf e0m.2e nmt mw/seerce, aded fro m diaphysis) to resist torsion and bending. For recorded at 0.025 sec intervals. Load- h displacement curves were generated, from which ttp evaluation of the trabecular envelope at the ultimate force (F ; N), stiffness (S; N/mm), and ://w femoral distal metaphysis and L5, each specimen U w was scanned with a slice increment of 9 µm. CT work to failure (U; mJ) were calculated. FU w.jb images were reconstructed, filtered (σ= 0.8 and represents the strength of the bone whereas U is a c.o support = 1.0), and thresholded (22% of maximum measure of the energy required to break the bone brg/ possible gray scale value) as previously described (23). y g u e (22). Scanning for the femur was started at 15% For L5, the endplates of the vertebral bodies st o of the total femur length measured from the tip of were removed via parallel cuts on a diamond n A p femoral condyle and extended proximally for 100 wafering saw (Isomet, Buehler, Lake Bluff, IL, ril 3 slices. Scanning for the L5 vertebral body USA). After removing the neural arch by clipping , 2 0 comprised 65% of the total vertebral body height, through the pedicles, the vertebral bodies were 1 9 requiring ~200 slices through the central region. submerged in a saline bath (~2 h) at room The area for trabecular analysis was outlined temperature and tested in axial compression at a within the trabecular compartment, excluding the crosshead speed of 0.05 mm/sec. F , S, and U U cortical and subcortical bone. Every 10 sections were calculated from the resulting load- were outlined, and the intermediate sections were displacement curves. interpolated with the contouring algorithm to For in situ forearm mechanical testing, five mice create a volume of interest. Parameters of at 16 weeks of age from each genotype were microarchitecture for both skeletal sites included chosen at random. They were killed by cervical bone volume (BV, mm3) and bone volume fraction dislocation under isoflurane-induced anesthesia. (BV/TV, %), as well as trabecular number (Tb.N, The right forearm was stored in refrigerated 70% mm-1), trabecular thickness (Tb.Th, µm), and ethanol for later strain measurement. The left trabecular separation (Tb.Sp, µm). In addition, we forearm was fixed between the cup-shaped platens computed the connectivity density (Conn.D, mm-3) and loaded to failure in compression at a and structure model index (SMI), which indicates crosshead speed of 2 mm/sec. F was calculated U the platelike (SMI = 0) or rodlike (SMI = 3) nature from the resulting force vs. displacement curves. of the underlying cancellous architecture. The mean ultimate force calculated for each Peripheral quantitative computed tomography genotype was used to set three peak load (pQCT): The right femurs of each mouse was magnitudes for the in vivo ulna loading experiments analyzed for volumetric BMD (vBMD) using pQCT. (see below). 3 In vivo parathyroid hormone (PTH) treatment: At measurements and mechanical strain data, strain 12 weeks of age, male and female Lrp5-/- and was estimated for the lateral periosteal surface of Lrp5+/+ mice were administered subcutaneous the histological sections from animals loaded in injections of human PTH 1-34 (40 µg/kg) or vivo as previously described (26). vehicle (99.7% normal saline, 0.2% bovine serum Strains on the medial surface of the endocortex albumin, and 0.1% HCl) 5 day/wk for 4 wks. All were calculated from periosteal strains and animals in the PTH study were scanned using the section properties, based on a scale factor pixiMUS densitometer at baseline (12 wks) and at calculated for each genotype and sex as follows: weekly intervals during the treatment (under isofluorane-induced anesthesia). Animals were sacrificed at 16 wks of age, one day after their last Where ε is the peak endocortical strain on the EM PTH or vehicle injection. medial surface and CT is the mean thickness of In vivo ulna loading: Male and female mice (16 the medial cortex. weeks old) in each of the three genotypes were Histomorphometry: Both right and left ulnae were divided randomly into three load magnitude fixed in 10% neutral-buffered formalin for 48h, groups for in vivo loading (n = 6-8/group) using dehydrated in graded alcohols, cleared in xylene, the ulna loading protocol described by Torrance and embedded in methylmethacrylate (Aldrich et al (24). Under isoflurane-induced anesthesia, Chemical Co., Inc., Milwaukee, WI, USA). the right forearm of each mouse was loaded at 60 Transverse thick sections (~80 µm) were cut at the cycles/day for 3 consecutive days using a 2-Hz mid-point using a diamond-embedded wire saw D haversine waveform. Loading was applied using o (Histosaw; Delaware Diamond Knives, Wilmington, w a previously described piezoelectric mechanical nlo DE, USA). Sections were ground to final thickness a stimulator (25). The non-loaded left forearms de of ~20 µm, and then mounted unstained on d served as an internal control. All mice were standard microscope slides. One section per ulna from asellosswieodns .n Ionrtmraaple rcitaognee aal cintijveictyti obnes twofe ecna lcleoiand (in20g was viewed at ×160 magnification on a Leitz http mg/kg body mass; Sigma Chemical, St. Louis, DMRXE microscope (Leica Mikroskopie und ://w MO, USA) and alizarin (30 mg/kg body mass; Scaypsttuerme dG umsibnHg, aW SePtzOlaTr, dGigeitraml acnaym) earnad ( Dthiaeg inmoasgtiec ww.jb Sfirisgtm lao)a wdienrge addamy. in isAtenrimeda l4s awnedr e9 dsaaycsr ifaicfteedr th16e Ifnosllotrwuminegn tpsr,i mInacr.y, Sdtaetralin gw eHreei gchotsll,e cMteI,d U SfroAm). Tthhee bc.org/ days after the first loading day. y g periosteal and endocortical surfaces using Image u e Ulna strain measurements: The left forearms Pro Plus v.4.1 software (Media Cybernetics, Silver st o from mice used in the forearm testing group Spring, MD, USA): total perimeter (B.Pm); single n A p (described above) were brought to room label perimeter (sL.Pm); double label perimeter ril 3 temperature slowly (~2 h) in a saline bath and measured along the first label (dL.Pm) and double , 2 0 minimally dissected to expose the lateral surface label area (dL.Ar). From these primary data, the 19 of the mid-shaft ulna. A single element strain following derived quantities were calculated (27): gauge (EA-06-015DJ-120; Measurements mineralizing surface (MS/BS = [1/2sL.Pm + Group, Raleigh, NC, USA) was bonded to the dL.Pm]/B.Pm; %), mineral apposition rate (MAR = exposed medial ulnar surface at the midpoint of dL.Ar/dL.Pm/5 days; µm/day), and bone formation the ulna (n = 5-6/group). The forearm was loaded rate (BFR/BS = MAR × MS/BS × 3.65; in cyclic axial compression, using the loading µm3/µm2/year). device to be used for in vivo ulna loading. Using In vivo GFP monitoring of load-induced a 2-Hz haversine waveform, the forearms were osteoblast recruitment: Sixteen to eighteen-week- loaded at 1.0, 1.2, 1.4, 1.6, and 1.8 N, during old Lrp5+/+, Lrp5+/-, and Lrp5-/- littermates that were which peak-to-peak voltage output from the also heterozygous for the pOBCol3.6GFP strain gauge was measured on a digital transgene on a mixed CD1/129Sv background, oscilloscope. Voltage measurements were were sedated with isoflurane and given a single converted to strain as previously described (26). bout of ulnar loading (90 cycles; 2 Hz; 1600 µε). The strain gauged ulnae were scanned through Five Lrp5+/+,(4 males, 1 female) 7 Lrp5+/- (3 males, the middle of the strain gauge on a µCT at 9-µm 4 females), and 10 Lrp5-/- (4 males, 6 females) resolution. The mid-shaft slices were imported mice were studied. The mice were sacrificed five into Scion Image, wherein I and the maximum MIN days after the loading bout and the right (loaded) section diameter in the I plane (Se.Dm; mm) MIN and left (unloaded) ulnae were removed, were calculated. From the geometric 4 decalcified, embedded in OCT (Tissue Tek, and the medium was aerated with 95% air-5% CO 2 Elkhart, IN), and sectioned transversely at 6 µm during the experiment. thickness using a cryostat (Leica). Sections from Measurement of adenosine triphosphate (ATP): the mid-shaft, as judged by the shape of the ulna Media samples were drawn from the reservoir of cross section, were stained with DAPI, visualized the flow loop one minute after beginning fluid shear by fluorescence microscopy, and photographed. stress. An ATP bioluminescence assay containing A 300 - 400 µm arc along the medio-lateral luciferin/ luciferase reagent was used to detect ATP periosteal surface (depending on the size of the (ATP Bioluminescence Assay kit HS II, Roche, bone) was used to determine the total number of Indianapolis, IN) in the media samples. This assay periosteal cells and the number of GFP-positive utilizes the conversion of D-luciferin by luciferase cells. Since Lrp5+/- and Lrp5+/+ mice had similar into oxyluciferin and light that requires ATP as a co- periosteal bone growth following mechanical factor. The resultant luminescence, measured using loading, data from these two genotypes were a Monolight 3010 (BD Biosciences Pharmingen, combined and then compared to that from the San Diego, CA), reflects ATP concentration. Lrp5-/- mice. Prostaglandin E Assay: After 60 minutes of 2 FSS, the cells were removed from the flow Cell culture studies chamber and 2 mL of 0.2% FBS media was added Isolation of bone cells: Calvarial osteoblasts onto the monolayer of cells and incubated for 30 were isolated from 3-5 day-old neonatal calvariae minutes at 37ºC in the incubator. After 30 minutes, from Lrp5+/+ and Lrp5-/- mice. Calvariae from the the conditioned media was collected for PGE2 Do w same genotype were grouped and subjected to measurement using a commercially available, n lo five sequential 15-minute digestions with enzyme competitive binding enzyme immunoassay kit ad e mixture of 1.5 u/ml collagenase P (Roche (BioTrak, Amersham Pharmaceuticals, Piscataway, d fro Molecular Biochemicals, Penzberg, Germany) in NJ). PGE2 release into the conditioned media was hm 0.05% trypsin/1mM EDTA (Gibco, Carlsbad, CA, normalized to total cell protein. The protein assay ttp USA) on a rocking platform. The first digest was was accomplished by Amido Black method (29). ://w w discarded and the second to the fifth digests were Western blotting: Cells were washed with cold w .jb passed through a 40 µm cell strainer (Falcon, PBS (1X), lysed with 2X SDS sample buffer (5mM c.o Becton Dickinson, Franklin Lakes, NJ, USA) and HEPES pH 7.9, 150mM NaCl, 26% glycerol, 1.5mM brg/ pooled. Cells were collected after centrifugation at MgCl2, 0.2mM EDTA, 0.5mM dTT and 0.5mM y g u 2,500 rpm for 8 minutes and plated in α -modified PMSF) on ice and immediately boiled for 5 e s essential medium (α-MEM) with 10% fetal bovine minutes. The protein samples were centrifuged at t on A serum (FBS), 100 U/ml of penicillin, and 100 µg/ml 14,000g for 10 minutes at room temperature and p of streptomycin in a T25 culture flask (Costar, the supernatant was retained and electrophoresed ril 3 Corning, NY, USA) and grown to confluence. After on a 10% SDS-polyacrylamide gel at 20 µg protein , 20 1 reaching confluence, cells were trypsinized and per lane. The proteins were electrotransferred to a 9 plated in a T75 culture flask (passage 1). nitrocellulose membrane and blocked in Tris- Confluent passage 1 cells were trypsinized and buffered saline containing 5% nonfat dry milk and seeded on collagen-coated glass slides for 0.1% Tween 20 (TBST). After blocking, the subsequent fluid flow experiment. membranes were incubated with an anti- Fluid flow experiments: Primary cells were osteopontin antibody (1:100, AssayDesign, Inc), seeded at a density of 2,000/cm2 and grown on 75 anti-vinculin antibody (1:500, Sigma-Aldrich, Inc.), x 38 mm2 glass slides (Fisher Scientific, or anti-extracellular signal-regulated kinase (ERK) Pittsburgh, PA) coated with 10µg/cm2 type I 1/2 antibody (1:500; Cell Signaling, Inc.) and anti- collagen (BD Biosciences, Bedford, MA). Upon phospho ERK1/2 antibody (1:500; Cell Signaling, reaching 90% confluence (2-3 days), the cells Inc.) overnight at 4°C. The membranes were were serum starved for 24 hours in 0.2% FBS washed 3 times in TBST, incubated with supplemented culture media (the same media appropriate HRP-conjugated secondary antibodies, used for flow). During experiment, fluid flow was washed an additional 3 times in TBST, and applied to the cell monolayer in a parallel plate developed using the enhanced chemi- flow chamber using a closed flow loop (28). This luminescence (ECL) method. system subjected the cells to a steady laminar flow Statistical Analysis: Data with multiple time profile producing a 12 dynes/cm2 fluid shear stress points, such as BMC/BW were first compared by (FSS). The apparatus was maintained at 37°C, repeated measures analysis of variance (ANOVA). 5 When analysis detected significant differences study (10), we observed that Lrp5-/- mice have a between genotypes, then significance at each time low bone mass phenotype. The difference in total point were assessed by Tukey-Kramer post hoc body BMC per unit body mass (BMC/BM) between test. Other phenotype values among genotype for Lrp5-/- and Lrp5+/+ mice became significant at 4 a given sex were compared by one-way ANOVA. weeks of age and persisted through adulthood, Gender comparisons were performed using a two- reaching a 23% deficit by 16 weeks of age in both way ANOVA with sex and genotype as male and female Lrp5-/- mice (Fig. 2A). independent variables. Periosteal dose responses Interestingly, male Lrp5+/- mice had significantly to different mechanical strains within genotypes less bone mass than Lrp5+/+ controls, but this was were tested for significance with least-squares not the case for the female Lrp5+/- mice. regression. Differences in slope and x-intercept µCT analysis in adult mice revealed that the (bone formation vs. mechanical strain) among midshaft femur cross-sectional area and the polar genotypes were tested for significance by analysis moment of inertia (an estimate of a structure’s of covariance (ANCOVA). When ANCOVA ability to resist torsional loading (31) were detected significant differences among slopes, significantly lower (by 25-50%) in Lrp5-/- mice Tukey’s HSD tests were performed to test compared to Lrp5+/+ controls (Table 1). Genotype- pairwise comparisons of slopes and x-intercepts related disparities were also detected in the (osteogenic threshold). Strains on the trabecular envelope of the distal femur and spine endocortical surface were much lower than (Fig 2B). For example, BV/TV was 81% lower in periosteal strains; consequently, a large portion of male Lrp5-/- femora than in their Lrp5+/+ controls. D the loaded bones were below the osteogenic o The Lrp5 deficiency appeared to have a greater w threshold and exhibited no osteogenic response to impact in male mice compared to females, and the nlo a loading. Therefore, the endocortical dose skeletal effects were greater at the distal femur ded response was analyzed using the segmented compared to spine. Other measures of cancellous fro m regression model PROC NLIN in Jmp v.4 (SAS architecture followed similar trends to those h Institute Inc., Cary, NC) in which a quadratic ttp equation was fit to each data series. Break points reported above for BV/TV, with the exception of ://w trabecular spacing (Tb.Sp), which, as expected, w (x0), where the segments join, were solved increased significantly in Lrp5-/- mice. w.jb iteratively according to the routine y = a+bx+cx2 if c.o mx a<xxi0m, aaln d(3 y0 )=. p ifT hx e> nxu0 mubnetilr st hoef cpoerrreiolasttieoanl wcaesll awshTsooolec i fafuetremthdoe rwra i tcahhn adLr ra5pct5ht elduriemzfeibc aietrh nvece yrs,t ekweblerea teda ilsf rspoehmce tn1eo6dt ywopkue-t by gurg/ nuclei and GFP+ periosteal cells in loaded versus es unloaded limbs, and in Lrp5+/+/Lrp5+/- and Lrp5-/- old Lrp5+/+, Lrp5+/-, and Lrp5-/- mice, and subjected t on them to biomechanical testing. The disparities A mice, were compared using student’s t-test. p observed in bone mass among the Lrp5+/+, Lrp5+/-, ril 3 For all tests, statistical significance was set at and Lrp5-/- mice were reflected in their , 2 0 α=0.05. 1 biomechanical properties. Under 3-point bending 9 conditions, the femoral shafts from Lrp5-/- were significantly more compliant, absorbed significantly Results less energy before failing, and failed at a Lrp5-null mice exhibit deficient bone mass, significantly lower force than Lrp5+/+ control femora geometry, structure, and strength, despite (Table 1). The difference between Lrp5+/+, and normal body weight and limb length Lrp5-/- properties ranged from 30-36% for all of the To begin assessing the role of Lrp5 in the parameters examined. The Lrp5+/- femora skeletal phenotype, we collected animal weights, exhibited an intermediate phenotype, but the measured femur lengths, and assessed bone difference between Lrp5+/+ and Lrp5+/- mechanical mineral content using pDEXA in wild type properties was significant only for ultimate force. (Lrp5+/+), heterozygous (Lrp5+/-), and Lrp5-null Vertebral compressive strength (FU) was reduced in (Lrp5-/-) mice. Within both sexes, the average male and female Lrp5-/- mice by 51% and 29%, body weight was statistically similar among respectively (Table 1). As in the femur, vertebral Lrp5+/+, Lrp5+/-, and Lrp5-/- mice (Table 1). Femur stiffness and work to failure followed similar trends lengths were similar in female Lrp5+/+, Lrp5+/-, and as FU. Lrp5-/- mice, but male Lrp5-/- mice exhibited significantly shorter femora at 18 wks of age Lrp5-/- mice exhibit almost no osteogenic response (Table 1). However, consistent with a previous to mechanical loading in vivo 6 To directly investigate the role of Lrp5 in bone 10.2% and 9.5% in Lrp5+/+ and Lrp5-/- mice, mechano-responsiveness in vivo, we subjected respectively (Fig. 4B). Among females, 4 wks of Lrp5+/+, Lrp5+/-, and Lrp5-/- mice to in vivo PTH treatment increased distal femur total BMD by mechanical loading and measured the bone 12.2% and 9.0% in Lrp5+/+ and Lrp5-/- mice, formation response histomorphometrically. Cross respectively. While the PTH-induced increase in sections of loaded ulnae from Lrp5+/+ mice BMD was significant for both sexes and genotypes, revealed new periosteal and endocortical lamellar the degree of enhancement was not statistically bone formation (bone between fluorochrome different between genotypes of the same sex. labels), mostly on the medial and lateral quadrants (Fig 3A). In contrast, Lrp5-/- mice ulnae showed Osteoblast recruitment to mechanically loaded very limited double labeling in the loaded ulnae. bone surfaces is normal in Lrp5-/- mice: The suppression of load-induced bone formation was evident in both male and female Lrp5-/- mice, After discovering a nearly complete absence of but females exhibited the most dramatic lack of the osteogenic response to mechanical loading in response. The relative periosteal bone formation Lrp5-/- mice, we sought to understand whether the rate (BFR/BS in the loaded ulna minus BFR/BS in deficit was related to inadequate recruitment/ the nonloaded ulna) was roughly 99 percent lower activation of osteoblasts at bone surfaces after in female Lrp5-/- mice compared to female Lrp5+/+ mechanical loading. To this end, we bred the Lrp5-/- mice (Fig. 3B). The heterozygous mice alleles onto the pOBCol3.6GFP mouse strain, responded to a lesser extent than Lrp5+/+ animals, which expresses green fluorescent protein (GFP) in D but statistical difference from Lrp5+/+ was reached early osteoblasts. Normally, osteoblasts begin to o w only for the males. The endocortical surface appear at strained surfaces 3-5 days after a nlo a showed similar trends as the periosteal surface. mechanical loading bout (32,33). We sacrificed de d Lrp5-/- mice failed to exhibit a bone formation Lrp5-/- pOBCol3.6GFP, Lrp5+/- pOBCol3.6GFP, and fro response on the endocortical surface, as revealed Lrp5+/+ pOBCol3.6GFP mice 5 days after a single hm by a strain vs. rBFR/BS regression that was not ulnar loading session, and counted the total ttp significantly different from zero (Fig 3C). number of cells and the number of GFP+ cells on ://w w w the lateral ulnar periosteum. .jb c We observed a difference in the number of GFP+ .o Parathyroid hormone is equally anabolic in the rg Lrp5-/- and Lrp5+/+ mouse skeleton cells, but not in the total number of periosteal cells by/ in the loaded versus non-loaded ulnae of all g u To assess the requirement of Lrp5 receptor e signaling in response to a well established bone genotypes (Figure 5). This result suggests that at st on 5 days following the ulnar stimulation with A anabolic agent, we treated 12-week-old mice for 4 p wks with daily injections of PTH 1-34 (40 µg/kg) or mechanical load, the periosteum had responded to ril 3 mechanical stimulation by activating cells already , 2 vehicle, and monitored whole body and hindlimb 0 at the periosteal surface, and not by recruiting new 1 9 bone mass changes by baseline (day 0) and final cells or stimulating periosteal cell division. Despite (day 28) pDEXA scans, and by femoral pQCT loaded ulnae having significantly more GFP+ cells scans at sacrifice. Percent change in bone than non-loaded ulnae, there were no differences in mineral content (baseline BMC vs final BMC for between Lrp5-/- and Lrp5+/+/Lrp5+/- mice (Figure 5). each animal) showed no significant sex effect, so This suggests that the absence of Lrp5 function male and female data were pooled. PTH does not affect the early stages of the osteogenic improved whole body BMC by 10.7% in Lrp5-/- response to mechanical stimulation. mice and by 7.2% in Lrp+/+ mice. ANOVA revealed significant overall genotype and treatment effects (p < 0.05), but a significant Primary osteoblasts from Lrp5-/- mice do not exhibit interaction between genotype and treatment was deficiencies in early mechano-transduction not detected, suggesting that PTH enhanced signaling in vitro, but later stage markers are hindlimb BMC significantly and equally in Lrp5-/- affected and Lrp5+/+ mice (Fig 4A). To further study the role of Lrp5 in osteoblast pQCT measurements of the mineral content in mechanotransduction, we cultured primary the distal femur revealed significant sex effects; osteoblasts from Lrp5+/+ and Lrp5-/- neonatal mouse consequently, male and female data were treated calvariae, sub-cultured them on to collagen-coated separately. Among males, 4 wks of PTH glass slides, and subjected them to fluid shear treatment increased distal femur total BMD by stress (12 dyn/cm2) to simulate mechanical loading 7 in vitro (Fig 6A). As has been shown by others intermittent PTH treatment. Thus, our data suggest (34,35), fluid shear stress (FSS) induces that one of the main mechanisms through which osteoblasts to release ATP into the culture media Lrp5 signaling exerts its effects on the skeleton is in after one minute. Primary osteoblasts cultured responding to mechanical signals. from Lrp5+/+ and Lrp5-/- mice released ATP in The results from our PTH experiments, indicating response to FSS, and Lrp5-/- cells were actually that intermittent PTH treatment was fully anabolic in more responsive than Lrp5+/+ cells (Fig 6B). We Lrp5-/- mice, were surprising. We had anticipated also measured PGE , an important mediator of 2 that Lrp5 inactivation would inhibit the anabolic mechanical loading in osteoblasts (36-38). Sixty response to PTH based on a report in the literature minutes of FSS induced a significant release of (46). Most data support an integral role for LRP5 in PGE into the culture media (4.5 to 6 fold over 2 the “canonical” Wnt signaling pathway, where static culture cells) in both Lrp5+/+ and Lrp5-/- cells, activation by a family of secreted glycoproteins— but no genotype difference in FSS-induced PGE 2 Wnts—at the cell surface ultimately leads to the release was detected (Fig. 6C). Next, we nuclear accumulation of β-catenin. Nuclear β- investigated the ability of fluid shear stress to catenin complexes with members of the TCF/LEF activate the mitogen-activated protein (MAP) transcription factor family to regulate gene kinase cascade in Lrp5+/+ and Lrp5-/- cells, since transcription. In vitro, UMR-106 cells treated with MAP kinase signaling is essential for the PTH exhibit an accumulation of β-catenin and expression of several genes linked to enhanced TCF-mediated transcriptional activity, mechanotransduction (39,40). In both Lrp5+/+ and two hallmarks of canonical Wnt signaling. D Lrp5-/- osteoblasts, 30 min of shear stress induced o Furthermore, continuous PTH exposure was w significant ERK1/2 activation, compared to reported to suppress Dkk1 expression in the rat nlo a genotype-matched static cells not subjected to femur, which in turn should increase signaling ded flow (Fig. 6D). No difference in shear-induced through Lrp5 (46). fro ERK1/2 activation was detected between Lrp5+/+ hm and Lrp5-/- cells. Thus, MAP kinase signaling Our findings could have resulted from PTH ttp appears intact in Lrp5-deficient cells. utilizing the canonical Wnt signaling cascade by ://w w acting through the closely related receptor Lrp6, w Because we observed no deficiencies in Lrp5-/- which is not altered in the Lrp5-null mice. Like .jb c osteoblasts with respect to early cellular .o Lrp5, Lrp6 has been shown to influence bone mass rg responses to FSS, we looked to a later marker of in mice (47) and Lrp5 and Lrp6 have overlapping by/ mechanotransduction. Osteopontin expression g patterns of expression, including in skeletal tissues, u e has been shown to correlate with mechanically- and are both inhibited by Dkk1. However, st o induced bone formation in vivo (41,42) and is intermittent PTH treatment has been shown to n A p upregulated by fluid shear stress in vitro (40,43- increase bone formation in mice overexpressing ril 3 45). Eight hours after initiation of fluid shear Dkk1, which argues against PTH action requiring , 2 stress of Lrp5+/+ and Lrp5-/- osteoblasts, we 01 functioning Lrp5 or Lrp6 receptors (48). 9 measured osteopontin protein expression by Furthermore, PTH is known to affect other signaling Western blotting. Osteopontin production was pathways, such as IGF-I receptor signaling (49). significantly enhanced in Lrp5+/+ but not Lrp5-/- Our finding that PTH can promote bone anabolism cells (Fig 7B). These data suggest that the later in Lrp5 mutant mice has been independently responses to mechanical stimulation, such as supported by a study (published in abstract form) matrix protein production, are impaired by the loss which found that 6 wks of intermittent PTH of Lrp5 signaling. treatment (80 µg/kg/d) increased osteoblast surface and femoral cortical thickness in Lrp5-/- mice to the Discussion same degree as in wildtype mice (11). Since the anabolic effects of intermittent PTH do not require a The main objective of our study was to functioning Lrp5 receptor in mice, it is possible that understand the role of Lrp5 in the skeletal intermittent PTH therapy in patients with OPPG or response to mechanical loading. Lrp5 null mice in heterozygous LRP5 mutation carriers may help clearly demonstrate that Lrp5 has an important increase bone mass and prevent fractures. Lrp5 role in normal bone acquisition during growth and mutant mice that were given a daily dose of lithium development. Loss of Lrp5 signaling results in a chloride also exhibited an increase in bone mass smaller, structurally inferior skeleton that is (20). LiCl is postulated to bypass the Lrp5 severely compromised in its ability to respond to deficiency by activating canonical Wnt signaling mechanical stimuli, but responds normally to downstream of the Wnt/Lrp5/Frizzled receptor 8 complex. Our studies do not preclude PTH by ex vivo studies in which we compared the exerting similar anabolic effects by modulating responsiveness of Lrp5-/- and Lrp5+/+ primary canonical Wnt signaling downstream of the osteoblasts to fluid shear stress. Lrp5 deficiency Wnt/Lrp5/Frizzled receptor complex. has little effect on early mediators of mechanical signaling, such as ATP and PGE release or In contrast to PTH responsiveness, skeletal 2 ERK1/2 activation, which is detectable within mechano-responsiveness was severely suppres- sed in the Lrp5-/- mice. The reproducible bone seconds or minutes of mechanical stimulation. However, Lrp5-/- osteoblasts responded differently formation response that is elicited by in vivo ulnar than Lrp5+/+ cells with respect to upregulation of loading was nearly lost in Lrp5-null mice. This osteopontin expression, which normally occurs 8 to was especially true for female mice, which 24 hrs after a mechanical stimulus (43-45, 51-53). exhibited a ~99% reduction in relative periosteal bone formation rate compared to Lrp5+/+ controls. Some early mediators of mechanical signaling, such as prostacyclin (54) and nitric oxide (55),were Our data complement mechanotransduction not measured so we cannot be certain that all early studies conducted in the high bone mass (HBM) mediators are unaffected by Lrp5 deficiency. mutation transgenic mice. Mice expressing a transgene for the high bone mass mutation in Lrp5 Mechanical loading (e.g., exercise) is an (G171V) exhibit a greater osteogenic response to effective means to increase bone mass, prevent mechanical loading in vivo than normal, non- bone loss, and perhaps most importantly, reduce transgenic mice (50). Interestingly, the G171V fracture susceptibility. The exercise-induced mutant mouse lost less bone than WT in response increase in bone strength occurs because new D o to disuse, but the bone loss occurring from a bone formation is targeted to skeletal surfaces w n nonmechanical challenge (ovariectomy) was the experiencing large mechanical strains, i.e., bone is lo a d same in G171V and wild type mice. Collectively, added where it is most needed. This targeting ed these observations highlight the key role of Lrp5 mechanism allows for the addition of a small fro m signaling in skeletal mechanotransduction. amount of bone to confer large increases in h ttp Interestingly, Lrp5-/- mice do not appear to have mechanical properties of the tissue by creating a ://w more mechanically favorable geometry. For w a deficiency in recruiting/activating osteoblasts to w the ulnar surface after a loading bout. Normally, example, increasing bone mineral content of a limb .jb c bone by only 5-10% via mechanical loading is .o after a mechanical loading session, osteoblasts rg appear at bone surfaces subjected to high associated with a 65% improvement in ultimate by/ force, a 90% improvement in fracture energy, and g mechanical strains 3-5 days after loading, and u e then begin synthesizing matrix which will a >100-fold increase in fatigue life (56-58). st o n ultimately become mineralized (32,33). A We have shown that Lrp5 function is required for A p deficiency in osteoblast recruitment/activation at the osteogenic response to mechanical loading. ril 3 loaded surfaces is one potential mechanism by We do not yet know whether agents such as LiCl, , 2 0 1 which impaired mechanotransduction can be that can activate canonical Wnt signaling 9 manifest. However, use of the pOBCol3.6GFP downstream of the receptor, can restore mouse allowed us to track the load-induced mechanoresponsiveness in patients with OPPG. osteoblast appearance in the otherwise quiescent Nor do we know whether other approaches to ulnar periosteal surface. In tissue sections taken increase Lrp5 signaling in healthy individuals can from these mice, we found equal numbers of enhance the skeleton’s response to exercise. periosteal cell nuclei in loaded and unloaded Overexpression of the wild type LRP5 receptor in ulnas, suggesting that the appearance of mice (using a collagen I promoter fragment) did osteoblasts five days after loading represents improve mechanical properties of bone, but it activation of previously quiescent cells, rather than remains to be determined whether overexpression cell division or recruitment of new cells. mediated its effect by enhancing load-induced Importantly, Lrp5-/- and wild-type mice had bone gain (12). Responding to mechano- comparable increases in the numbers of GFP+ transduction is unlikely to be the sole function of early osteoblasts at the periosteal surface Lrp5 in the skeleton. Canonical Wnt signaling has following mechanical loading. These data indicate recently been demonstrated to enable marrow that the Lrp5-/- osteoblasts are able to be activated stromal cells to differentiate toward the osteoblast by mechanical loading and imply that the defective lineage, rather than other cell fates (59-61) and to periosteal bone apposition response caused by alter the expression in mature osteoblasts of Lrp5 deficiency is in the further maturation of secreted regulators of osteoclast differentiation. these osteoblasts. This conjecture is supported Consistent with these additional functions, the 9 administration of LiCl to Lrp5 mutant mice modulating that interaction pharmacologically could decreases marrow adiposity, while increasing prove to be a useful target for OPPG patients. trabecular bone volume (20). In conclusion, Lrp5 appears to be a potent We observed sex-related differences in the regulator of bone mass, size, and strength. One of severity of the phenotype in Lrp5 mutant mice. the main mechanisms of action for the receptor in Lrp5-/- females exhibited a more mild phenotype in bone is through mechanical signaling, but not PTH terms of bone mass (BMC), structure (I , BV/TV) signaling. Our data indicate that Lrp5 is a late- P and mechanical properties (F , stiffness, work to acting mediator in the osteogenic response to U failure) than their male counterparts. For some mechanical loading, suggesting that skeletal measurements, the phenotype was twice as fragility in individuals afflicted with OPPG might be severe in males. The reason for these disparities related to inadequate processing of signals derived between the sexes is unclear, particularly in light from mechanical stimulation. In addition, of OPPG’s known autosomal mode of intermittent PTH might be an effective treatment for transmission. One potential explanation receiving improving bone mass in these patients. increasing attention is the link between signaling in the estrogen and Wnt pathways, which appear Acknowledgements to participate in considerable cross-talk. For We wish to thank Andrea Paris for animal care, example, transgenic mice expressing Wnt 10b are Jean Bao and Erin Hatt for assistance with uCT resistant to ovariectomy induced bone loss (59). scanning, Dr. Shigeo Tanaka for technical Furthermore, estrogen receptor α (62) and assistance in loading device, and Dr. Keith Condon D o androgen receptors (63,64) interact functionally for assistance with tissue processing. This work w n with β-catenin in vivo. If enhanced estrogen was supported by the National Institutes of Health loa d signaling in females is providing some osteo- (AR046530 CHT and AR053237 AGR). ed protective effects from the loss of Lrp5 signaling, fro m h ttp ://w w w .jb c .o rg b/ y g u e s t o n A p ril 3 , 2 0 1 9 10

Description:
1Department of Orthopaedic Surgery, Biomechanics and Biomaterials Research to: Charles H. Turner, PhD, Director of Orthopaedic Research, Indiana Warden, S. J., Hurst, J. A., Sanders, M. S., Turner, C. H., Burr, D. B., and Li,
See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.