Repression of osteocyte Wnt/catenin signaling ... - Wiley Online Library

ORIGINAL ARTICLE

JBMR

Repression of Osteocyte Wnt/b-Catenin Signaling Is an Early Event in the Progression of Renal Osteodystrophy Yves Sabbagh , 1 * Fabiana Giorgeti Graciolli , 2 * Stephen O’Brien , 1 Wen Tang , 1 Luciene Machado dos Reis , 2 Susan Ryan , 1 Lucy Phillips , 1 Joseph Boulanger , 1 Wenping Song , 1 Christina Bracken , 1 Shiguang Liu , 1 Steven Ledbetter , 1 Paul Dechow , 3 Maria Eugenia F Canziani , 4 Aluizio B Carvalho , 4 Vanda Jorgetti , 2 Rosa MA Moyses , 2 and Susan C Schiavi 1 1

The Sanofi-Genzyme R&D Center, Genzyme, A Sanofi Company, Framingham, MA, USA Department of Internal Medicine, Division of Nephrology, University of Saõ Paulo, Saõ Paulo, Brazil 3 Department of Biomedical Sciences, Baylor College of Dentistry, Texas A&M Health Science Center, Dallas, TX, USA 4 Department of Internal Medicine, Division of Nephrology, Federal University of Saõ Paulo, Saõ Paulo, Brazil 2

ABSTRACT Chronic kidney disease–mineral bone disorder (CKD-MBD) is defined by abnormalities in mineral and hormone metabolism, bone histomorphometric changes, and/or the presence of soft-tissue calcification. Emerging evidence suggests that features of CKD-MBD may occur early in disease progression and are associated with changes in osteocyte function. To identify early changes in bone, we utilized the jck mouse, a genetic model of polycystic kidney disease that exhibits progressive renal disease. At 6 weeks of age, jck mice have normal renal function and no evidence of bone disease but exhibit continual decline in renal function and death by 20 weeks of age, when approximately 40% to 60% of them have vascular calcification. Temporal changes in serum parameters were identified in jck relative to wild-type mice from 6 through 18 weeks of age and were subsequently shown to largely mirror serum changes commonly associated with clinical CKD-MBD. Bone histomorphometry revealed progressive changes associated with increased osteoclast activity and elevated bone formation relative to wild-type mice. To capture the early molecular and cellular events in the progression of CKD-MBD we examined cell-specific pathways associated with bone remodeling at the protein and/or gene expression level. Importantly, a steady increase in the number of cells expressing phosphor-Ser33/37-b-catenin was observed both in mouse and human bones. Overall repression of Wnt/b-catenin signaling within osteocytes occurred in conjunction with increased expression of Wnt antagonists (SOST and sFRP4) and genes associated with osteoclast activity, including receptor activator of NF-kB ligand (RANKL). The resulting increase in the RANKL/osteoprotegerin (OPG) ratio correlated with increased osteoclast activity. In late-stage disease, an apparent repression of genes associated with osteoblast function was observed. These data confirm that jck mice develop progressive biochemical changes in CKD-MBD and suggest that repression of the Wnt/b-catenin pathway is involved in the pathogenesis of renal osteodystrophy. ß 2012 American Society for Bone and Mineral Research. KEY WORDS: RENAL OSTEODYSTROPHY; OSTEOCYTE; WNT/b-CATENIN

Introduction

C

hronic kidney disease (CKD) is a growing worldwide health concern with greater than 13.4 million individuals in stages 3 or beyond in just the United States alone.(1,2) Patients continue to die from comorbidities associated with renal dysfunction and the risk of dying far exceeds the risk of entering dialysis.(3) Vascular calcification and left ventricular hypertrophy (LVH), two pathological findings associated with cardiovascular disease and

sudden cardiac death, are detected in early CKD and continue to increase in severity as renal failure advances.(4) Evidence suggests that vascular disease is at least partially related to poor bone health.(5–8) Importantly, deviation from the normal range of serum bone biomarkers predicts an increased risk of cardiovascular events,(9) and impaired bone remodeling is associated with increased vascular calcification.(10) In the general population, individuals with osteoporosis have increased atherosclerosis and coronary artery calcification(11–14) and there

Received in original form October 4, 2011; revised form March 8, 2012; accepted March 27, 2012. Published online April 10, 2012. Address correspondence to: Susan C Schiavi, PhD, Genzyme Science Center, 49 New York Ave., Framingham, MA 01701, USA. E-mail: [email protected] Additional Supporting Information may be found in the online version of this article. *YS and FGG contributed equally to this work. Journal of Bone and Mineral Research, Vol. 27, No. 8, August 2012, pp 1757–1772 DOI: 10.1002/jbmr.1630 ß 2012 American Society for Bone and Mineral Research

1757

is an inverse relationship between coronary artery calcification and bone mineral density (BMD) in both normal and CKD patients.(15) Agents that enhance bone quality also reduce cardiovascular calcification.(16–23) Additionally, soft-tissue calcification including vascular calcification is a common feature of genetic mouse and rat models harboring defects associated with bone mineralization.(24–26) Clinical improvements in bone health, regardless of the initial bone defect, ameliorate progression of cardiovascular disease.(23) Taken together, these observations support a mechanistic link between bone health and cardiovascular disease and have prompted the Kidney Disease: Improving Global Outcomes (KDIGO) organization to describe characteristics in mineral dysregulation as CKD–mineral bone disorder (CKD-MBD).(27) Hence, CKD-MBD is associated with characteristic laboratory findings, progressive alterations in bone remodeling, and soft-tissue calcification. Delineation of early changes in CKD leading to bone and cardiovascular disease will provide the foundation for development of improved therapeutic strategies. As a first step in understanding how bone alterations may contribute to the progression of cardiovascular disease, we assessed the early pathogenesis of renal bone disease by characterizing progressive changes in jck mice, a genetic model of CKD.(28) jck mice develop concurrent serum biochemical changes characteristic of human CKD-MBD,(29) with approximately 40% to 60% of jck mice also developing vascular calcification and initial stages of LVH. Histomorphometric analysis revealed the presence of high-turnover (HTO) bone disease prior to a significant elevation of serum parathyroid hormone (PTH). To explore molecular pathways associated with these changes, we monitored expression of key pathways known to be associated with bone turnover. In particular, new evidence suggests that the osteocyte is a prominent source of bone remodeling factors, including receptor activator of NF-kB ligand (RANKL), osteoprotegerin (OPG), and sclerostin, because these influence osteoclast and osteoblast activity.(30–34) Furthermore, modulation of the WNT/b-catenin pathway has recently been shown to regulate osteoclast activity by altering the osteocyte RANKL/OPG ratio.(34) Our data demonstrate that repression of the Wnt/b-catenin pathway within osteocytes occurs in parallel with an increase in the RANKL/OPG ratio and osteoclast activity. Decreased expression of genes associated with osteoblast function is observed late in disease progression. Similar changes in b-catenin signaling are observed in human CKD biopsies representing individuals with either low-turnover (LTO) or HTO bone disease. These data suggest that changes in b-catenin signaling may be a common early event in the pathogenesis of renal osteodystrophy.

Subjects and Methods Mice All studies were approved by the institutional animal care committee. Wild-type (WT) (C57BL/6J) and jck mice were originally obtained from Jackson Laboratories (Bar Harbor, ME, USA). Characterization of these mice has been described.(28) Mice were maintained in a virus- and parasite-free barrier facility and

1758

SABBAGH ET AL.

exposed to a 12-hour light/dark cycle. WT animals were maintained on standard rodent chow diet (PicoLab Rodent Diet 20, #5053; LabDiet, St. Louis, MO, USA) containing 0.63% phosphate, 0.81% calcium, with 2.2 IU/g Vitamin D3. To promote the development of hyperphosphatemia, jck mice were maintained on a casein-based diet containing 0.4% calcium and 1.0% phosphate (Diet #D08112306; Research Diets, Inc., New Brunswick, NJ, USA) from 4 weeks of age. Note that parallel studies were performed using WT mice fed the casein diet (0.4% calcium, 1.0% phosphate) and jck fed the standard rodent chow diet (Supplemental Fig. 2). There were no differences in serum or bone histomorphometry values between WT mice fed either diet. Therefore, the data presented are from WT mice fed a standard chow unless otherwise mentioned in the text or figure legends. When maintained on a standard chow, jck mice developed hyperparathyroidism and hyperphosphatemia but induction was inconsistent or delayed relative to findings using the specialized diet. As such, all data presented in this work are from jck mice fed the casein-based diet containing 0.4% calcium and 1.0% phosphate unless otherwise mentioned in the text or figure legends. Female mice were used because the progression of renal disease occurs slower relative to male jck mice, providing a broader window to capture early events.(28)

Serum and urine analysis Whole blood was collected under isoflurane anesthesia via retroorbital bleed, incubated for 20 minutes at room temperature, and then centrifuged at 48C. Serum was aliquoted and frozen at 808C for subsequent analysis. For urine collection, animals were placed in metabolic cages for a 24-hour period. Urine was centrifuged to remove particulates and resultant supernatant volume recorded and analyzed. All serum and urine phosphate (Pi), calcium (Ca), blood urea nitrogen (BUN), and creatinine (enzymatic method) were measured on an Integra 400 bioanalyzer (Roche Diagnostics, Indianapolis, IN, USA). Intact PTH (Immutopics, San Clemente, CA, USA), 1,25-dihydroxyvitamin D (1,25(OH)2D3) (IDS, Fountain Hills, AZ, USA), and intact fibroblast growth factor 23 (FGF23) (Kainos, Tokyo, Japan) ELISAs were performed according to the manufacturer’s instructions.

Tissue harvest and histology Tissues were harvested, fixed in 10% formalin for 20 hours, and then placed in 70% ethanol. Processed tissues were embedded in paraffin and sectioned. Sections were dewaxed in xylene, rehydrated by ethanol gradient, and stained in hematoxylin and eosin. Aortic sections were stained by the von Kossa method to visualize calcium phosphate deposition. For bone immunostaining, femurs were decalcified, processed, and embedded in paraffin and sectioned. For dynamic bone histomorphometric analysis, mice were given intraperitoneal (IP) injections of calcein (20 mg/kg) 8 days before euthanasia, and xylene orange (30 mg/kg) 3 days before sacrifice. Femurs were fixed in 40% alcohol, processed, and embedded in methyl methacrylate. Each femur was sectioned in the ventral/dorsal plane. Two levels per block separated by 100 mm were selected. Goldner’s Trichrome and von Kossa sections were analyzed for static measurements. Journal of Bone and Mineral Research

Unstained serial sections were used for dynamic measurements using the Osteo II Bioquant system (BIOQUANT, Nashville, TN, USA). Nomenclature is in agreement with recommendations by Parfitt and colleagues.(35)

Assessment of vascular calcification Vascular calcification was assessed using Osteosense-680 (Perkin Elmer, Bedford, MA, USA), an imaging agent consisting of a fluorescently labeled bisphosphonate, which was injected into the tail vein 24 hours before isolation of the aortic tissue. The isolated tissue was then imaged using the Odyssey infrared imaging system (LI-COR Biosciences, Lincoln, NE, USA). Quantitative biochemical analysis of aortic calcium was measured on dissected aortic tissue that were frozen, lyophilized, and decalcified with 0.6 N HCl at 378C for 24 hours. After centrifugation, the calcium content of the resulting supernatant was determined colorimetrically with the o-cresolphthalein complexone reactant using the Total Calcium LiquiColor kit (Stanbio Laboratories, Boerne, TX, USA). Aortic calcium content was normalized to the dry weight of the tissue and expressed as micrograms of calcium per milligram dry weight.

Micro–computed tomography bone imaging Tibias were imaged by micro–computed tomography (mCT) using isotopic voxels at a resolution of 3.5 mm (AG Micro-CT 35; Scanco Medical, Bru¨ttisellen, Switzerland). Fifty slices of middiaphyseal cortical bone (175 mm total thickness) were analyzed with a threshold of 565 mg/cc. The following variables were calculated using Scanco software routines and compared between groups: bone volume fraction (BV/TV), apparent density (density of cortical bone including resorption spaces and voids) (data not shown), and BMD (threshold set for cortical bone density only).

Strength testing Tibias were tested in three-point bending, and a strength parameter (maximum load at failure) was assessed with a Test Resources (Shakopee, MN, USA) DDL200 axial loading machine outfitted with an Interface SMT1-22 force transducer. Cross-head displacement rate was 0.1 mm/s. Tests were conducted on the mid-diaphyses with the bones resting on two supports 5 mm apart and the tibial anterior margins facing upward toward the actuator.

Gene expression analysis Colony formation unit assay Bone marrow was collected from six, 6-week-old female WT or jck animals and cultured in a modified essential medium (a-MEM) containing 15% heat-inactivated fetal bovine serum (HI-FBS), and penicillin-streptomycin (Invitrogen, Carlsbad, CA, USA). Media was replaced 3 days postadherence, and refreshed 7 days thereafter. After 13 days, adherent cells were plated at 5000 cells/ cm2 in triplicate six-well plates and cultured in differentiation media, a-MEM, 50 mg/mL ascorbic acid, 10% HI-FBS, and penicillin-streptomycin. Media was replaced every 4 days. To monitor the formation of colony formation units (CFUs), at 7, 15, or 21 days, cells were fixed and stained for alkaline phosphatase activity with 10 mg naphthol AS-MX, 20 mg Fast Blue BB salt, 1 mL N,N-dimethylformamide, in 19 mL of 0.1 M Tris (pH 9.2) (all from Sigma, St. Louis, MO, USA) to detect CFU-osteoblasts (CFU-OB). Colonies greater than 1 mm in diameter were counted. CFU-fibroblasts (CFU-F) were detected by staining the same plates with 0.2% crystal violet in 2% ethanol for 1 hour. Plates were washed with water and then air-dried. All colonies greater than 1 mm in diameter were counted to yield the percentage of CFU-O/CFU-F.

Quantitative in vitro mineralization Bone marrow was cultured according to the CFU assay. On day 13, in quadruplicate, adherent cells were plated 50,000 cells/cm2 in a 96-well plate in mineralization media (a-MEM, 50 mg/mL ascorbic acid, 5 mM b-glycerol phosphate) (Sigma), 10% HI-FBS, penicillinstreptomycin, plus 1 mg/mL calcein (Sigma). On days 3, 7, and 10, cells were washed free of unbound calcein and scanned for fluorescence (Excitation at 485 nm, Emission at 520 nm). Calcein standards were also applied to calculate absolute values of calcein incorporation. Journal of Bone and Mineral Research

Soft tissues were harvested and snap frozen in Trizol (Sigma). Bone shafts were collected, epiphyses removed, bone marrow displaced via centrifugation, and the shafts placed in Trizol (Sigma). RNA was extracted using the chloroform and isopropanol precipitation method. The extracted RNA was treated with DNase, purified on a Qiagen (Valencia, CA, USA) column, and eluted in RNAse-free water. A reverse transcriptase reaction was performed. The generated cDNA was used in single Taqman assays or Taqman low-density arrays (Applied Biosystems, Carlsbad, CA, USA) containing genes of interest and assayed according to the manufacturer’s protocol. The difference in expression was calculated using 18S as the control gene.

Immunostaining of mouse bones and aortas Five-micrometer paraffin sections were used for immunohistochemistry. Briefly, sections were dewaxed, rehydrated, and incubated with DeCal Retrieval Solution (BioGenex, Fremont, CA, USA) for bone sections only, at room temperature for 30 minutes followed by PBS washes. The sections were then treated with 3% H2O2 in methanol for 10 minutes and blocked with 1.5% blocking serum from the species in which the secondary antibody was made. Then, primary antibodies (rabbit polyclonal anti-b-catenin [1:1000; Abcam, Cambridge, MA, USA]; rabbit polyclonal antiphosphor-Ser33/37-b-catenin [1:100; Abcam]; goat-anti-mouse sclerostin [1:100; R&D Systems, Minneapolis, MN, USA]; goat antimouse SM22 [1:400; Abcam]; and goat anti-mouse osteopontin [1:200; R&D Systems]) were incubated overnight at 48C, washed in PBS, and detected by diluted biotinylated secondary antibodies. After a PBS wash, slides were incubated in Vector ABC Reagent (Vector Laboratories, Burlingame, CA, USA), developed with diaminobenzidine substrate (Vector Laboratories), and counterstained with Mayer’s hematoxylin. The slides were then washed in tap water for 5 minutes, dehydrated,

REPRESSION OF OSTEOCYTE WNT/b-CATENIN SIGNALING IN RENAL OSTEODYSTROPHY

1759

cleared, and mounted with Cytoseal XYL (Thermo, Rockford, IL. USA). Images were visualized on a Zeiss Axiovert 200 (Thornwood, NY, USA). For quantitation, slides were scanned using an Aperio scanner (Vista, CA, USA). Osteocytes were selected using Imagescope software (Aperio, Vista, CA, USA) and an algorithm was developed to quantitate the amount of positive osteocytes for each section.

Immunohistochemistry of human bone biopsies Immunohistochemistry was performed on 30 human biopsies from similar-aged individuals. Normal bones were from 10 postmortem donors with no known bone disease at the time of autopsy. Additional biopsies were selected from previous clinical studies representing 10 CKD and 10 hemodialysis (HD) patients in which five of each category were diagnosed with HTO or LTO bone disease (Table 1).(36–38) CKD biopsies were from individuals with 24-hour creatinine clearance rates between 15 and 90 mL/m2 who had not received phosphate binders, vitamin D analogues, or corticosteroids.(36) HD patients (stage 5) were 18 years or older, and had been on hemodialysis three times weekly for at least 3 months.(37) The technique for immunohistochemical detection of bone proteins from clinical samples was adapted from a previously reported method.(39) In brief, two adjacent 5-mm sections of bone tissue were placed side by side on each slide. Bone sections were deacrylated in a 1:1 mixture of xylene and chloroform for 30 minutes, rehydrated in graded alcohol solutions, submitted to a quick decalcification with 1% acetic acid for 10 minutes, and rinsed twice with distilled water. Endogenous peroxidase activity was inhibited by a mixture of 3% hydrogen peroxide in methanol for 30 minutes, followed by two water washes. The samples were incubated with Protein Block (DAKO Cytomation, Carpinteria, CA, USA) to block nonspecific bindings. Sections were incubated overnight at 48C in a humidified chamber using primary human monoclonal antibody anti-sclerostin (dilution 1:100 [R&D Systems], b-catenin [dilution 1:100; Abcam], and phospho-Ser33/37-b-catenin [diluTable 1. Patient Characteristics Corresponding to Bone Biopsies Used in Fig. 9 Normal

CKDa

HDa

Patients, n 10 10 10 Age 49.5 1.55 50 3.5 53 4.8 Gender (M/F) 5/5 8/2 5/5 Turnover (H/L) NA 5/5 5/5 Creatinine clearance ND 48 5.5 0 (mL/min) iCa (mmol/L) ND 1.31 0.01 1.22 0.03 P (mg/dL) ND 3.6 0.2 6.7 0.3 PTH (pg/mL) ND 94 14 400 70 FGF23 (pg/mL) ND 65 14 32,764 12,738 Values are presented as average SEM. CKD ¼ chronic kidney disease; HD ¼ hemodialysis; M/F ¼ number of males/females; H/L ¼ number with high or low bone turnover; NA ¼ not not applicable; ND ¼ not determined; iCa ¼ ionized; P ¼ phosphate; PTH ¼ parathyroid hormone; FGF23 ¼ fibroblast growth factor 23. a CKD represents patients in CKD stages 2-4, HD stage 5. p < 0.05 versus CKD 2–4.

1760

SABBAGH ET AL.

tion 1:100; Abcam]) in Antibody Diluent (DAKO Cytomation). After incubation with the primary antibody, sections were rinsed in Wash Buffer (DAKO Cytomation) and subsequently incubated with Labeled Streptavidin Biotin (LSAB) System–horseradish peroxidase (HRP) (DAKO Cytomation) according to manufacturers’ instructions. Antigen–antibody complexes were visualized using a 3-amino-9-ethylcarbazole substrate chromogen (AEC) (Sigma). The sections were rinsed in distilled water and counterstained with Mayer’s hematoxylin solution (Merck, Darmstadt, Germany). Negative controls were performed for each bone section by omitting the primary antibody. Reproducibility was ensured by repeating the immunohistochemical analysis on all specimens. The entire trabecular bone was assessed in each section by bone histomorphometric diagnosis and by serum biochemical values. In order to quantify the bone expression of sclerostin, b-catenin, and phosphor-Ser33/37-bcatenin—the presence of which in the trabecular bone was limited to the osteocyte cell—staining was classified as ‘‘present’’ or ‘‘absent’’ in any given osteocyte, and the total number of trabecular osteocytes with positive staining was counted and normalized to bone area.

Statistics Student’s t test was used to analyze significance between two groups. Values of p < 0.05 were considered significant.

Results Progressive changes in jck mouse serum biochemistries are characteristic of CKD-MBD To determine if the jck mouse, a well-characterized progressive genetic model of polycystic kidney disease (PKD), exhibits biochemical features of CKD-MBD, we examined temporal changes in serum chemistries and hormones (Fig. 1).(28) Serum BUN and creatinine levels were monitored over a period ranging from 6 to 18 weeks of age in female jck and WT controls. As expected, WT mice maintain their serum BUN within the normal range (22.50 0.43 mg/dL), whereas jck mice exhibit a gradual increase in both serum BUN (Fig. 1A) and creatinine (Supplemental Fig. 1), which continue to rise until death at about 20 weeks of age. Total serum calcium concentrations remain similar between jck and age-matched WT mice (Fig. 1B). A statistically significant increase in serum phosphate levels was observed in jck mice relative to WT mice from 6 weeks (9.85 0.57 mg/dL versus 8.45 0.19). Importantly, severe hyperphosphatemia was observed in jck mice relative to baseline beginning at approximately 15 weeks of age (8.8 0.39 mg/dL versus 14.41 0.33 mg/dL) when BUN levels were approximately 95 mg/dL (an approximately fourfold relative increase to WT mice) that continue to accumulate with declining kidney function (Fig. 1C). Progressive accumulation of serum PTH and FGF23 was observed in jck (Fig. 1D, E). Serum FGF23 levels were clearly induced prior to the elevation of serum PTH and reached concentrations exceeding 1000-fold greater than WT animals at 18 weeks (Fig. 1E) (26.01 7.70 pg/mL versus 0.14 0.02 pg/mL). Although serum 1,25(OH)2D3 concentrations are elevated at the Journal of Bone and Mineral Research

Fig. 1. Serum biochemistry values during progression of chronic kidney disease in jck (*) and WT (*) mice are characteristic of CKD-MBD. (A) Renal dysfunction, as measured by accumulation of BUN, progressively declines in jck mice whereas no change is observed in WT mice. (B) Total serum calcium levels are similar between jck and WT mice and do not change significantly over time. (C) Serum phosphorus is elevated in jck mice relative to WT with greatest increases observed at 15 and 18 weeks of age. (D) PTH values are similar between WT and jck mice until 15 weeks of age when levels begin to increase in jck serum. (E) FGF23 levels are elevated in jck relative to WT serum. (F) Serum 1,25(OH)2D3 concentrations are elevated in jck serum relative to WT. Levels in jck serum decline over time relative to 6-week-old baseline values. n ¼ 3–16 mice/group. Results are expressed as mean SEM. p < 0.05 for jck mice versus WT age-matched mice.

start of the study relative to WT animals, a decline from baseline was observed in jck mice with advanced kidney dysfunction (Fig. 1F).

jck mice develop cardiovascular abnormalities The prevalence of vascular calcification was monitored in jck mice at different ages to assess the degree of soft-tissue calcification. Von Kossa staining detected medial calcification in a subset of aortas isolated from 15 weeks of age. By 18 weeks of age, vascular calcification and myocardial calcification were observed in about 40% of the jck mice, whereas staining was not observed in any WT animals (Fig. 2A). Using the more sensitive method of Osteosense labeling, approximately 60% of jck mice had detectable calcification at 18 weeks of age (Fig. 2B). These results correlated with a significant increase in total calcium content of aortas at 18 weeks of age (Fig. 2C). A major mechanism associated with vascular calcification in humans is induction of an osteochondrogenic differentiation of smooth muscle cells. As shown in Fig. 2A, expression of the smooth muscle cell marker, SM22 alpha, is markedly decreased in the calcified areas of aortas from jck mice as compared to those from WT mice. In contrast, staining of the osteoblast marker protein, osteopontin (OPN), is absent in WT mice and strongly immunoreactive in the calcified areas of jck aortas. Interestingly, wall thickness was significantly reduced in calcified areas of aorta relative to WT controls (Fig. 2A). An additional hallmark of vascular calcification is the development of aortic stiffness that contributes to LVH. As shown in Fig. 2D and E, mRNA levels of two LVH markers, atrial natriuretic factor (ANP) and brain natriuretic peptide (BNP) were significantly increased in 19-week-old jck mice relative to WT mice, suggesting that these mice were beginning to develop LVH. Journal of Bone and Mineral Research

Histomorphometric analysis reveals early bone changes in mice The jck mouse harbors a mutation in the Nek8 gene, encoding a member of the never in mitosis A (NIMA)-related kinase family protein that leads to PKD. Before assessing whether jck bones develop progressive renal osteodystrophy, we first asked whether the Nek8 mutation directly contributes to bone changes before the onset of kidney disease by examining osteogenic differentiation of mesenchymal stem cells (MSCs). MSCs were isolated from WT and jck mice at 6 weeks of age and differentiated ex vivo. As shown in Fig. 3A, the percentage of alkaline phosphatase–positive colony forming units (CFU-OB) derived from WT and jck mice were identical. Additionally, measurement of net mineralization using in vitro calcein incorporation demonstrated that there was no difference between the two cell populations’ innate ability to undergo terminal differentiation (Fig. 3B). There were no significant differences in bone strength as determined by three-point bending in WT versus jck mice at 6 weeks of age before the onset of renal dysfunction (Fig. 3C). Finally, mCT analysis failed to reveal any differences in BMD as defined by material density (Fig. 3D) (or apparent density, not shown) or BV/TV (Fig. 3E). To confirm the presence and define the progressive changes associated with the pathogenesis of renal osteodystrophy in mice, we performed bone histomorphometric analysis on jck bones isolated at various ages. The three primary parameters suggested by KDIGO guidelines (turnover, mineralization, and volume)(27) were used to characterize the bone phenotype (Fig. 4). jck mice exhibit classic features of HTO starting early in disease (9 weeks old) with bone formation rates (BFRs) increasing approximately twofold (0.57 0.04 jck at 9 weeks old versus 0.35 0.03 WT at 6 weeks old, and 0.32 0.03


1761

Fig. 2. Vascular calcification is present in jck mice. (A) Calcification is observed by von Kossa staining in smooth muscle layers in jck but not WT aortas. Immunodetection of the smooth muscle marker, SM22 alpha is evident in WT and significantly reduced in jck aortas. Inversely, the osteoblast marker, osteopontin (OPN), is not detected by immunohistochemistry in WT aortas but is present in calcified areas in jck aortas. (B) Upper panel: the vascular tree whole mount from WT and jck mice. Lower panel: calcification is evident in jck aortas but absent in WT aortas using the fluorescently tagged bisphosphonate, Osteosense (Red). (C) Total calcium content is significantly elevated in aortas of jck mice relative to WT. Relative levels of (D) ANP and (E) BNP mRNA, markers of left ventricular hypertrophy (LVH), are elevated more than threefold in jck mice. 20 images. Results are expressed as mean SEM. p < 0.05 for jck mice versus age-matched WT.

Fig. 3. Mutations in the Nek8 gene induce PKD but have no significant influence on bone health in jck mice. (A) In vitro differentiation potential of MSCs isolated from WT (*) and jck (*) mice is similar. (B) Osteoblast function as assessed by in vitro mineralization is similar between differentiated MSCs isolated from WT and jck. (C) Bone strength assessed by three-point bending was similar in bones from 6-week-old jck mice with normal renal function and aged-matched WT controls. (D) BMD and (E) BV/TV were also similar in jck and WT mice with normal renal function. Results are expressed as mean SEM. p < 0.05 for jck mice versus age-matched WT.

1762

SABBAGH ET AL.

Journal of Bone and Mineral Research

Fig. 4. Turnover, mineralization, and volume parameters during progression of renal disease in jck mice. (A–C) Histomorphometric assessment of trabecular bone from age-matched jck (*) and WT (*) mice. (A) Bone formation rate (BFR) as assessed using double labeling was increased in jck mice relative to WT from 9 weeks of age. (B) Mineralization lag time (MLT) declines with age in both jck and WT mice with no observed differences. (C) Bone volume (BV/TV) is variable in both WT and jck mice. Volume is increased at 12 and 15 weeks of age in jck compared to WT. n ¼ 5–9 mice/group. Results are expressed as mean SEM. p < 0.05 for jck mice versus age-matched WT.

jck at 6 weeks old; values are BFR/bone surface (BS) in mm3/mm2/ year) due to a significant increase in mineralizing surface/ bone surface (MS/BS) (Fig. 4A and Table 2) and the absence of a mineralization defect (Fig. 4B). The increased BFR was also evident by increased serum osteocalcin levels (47.83 1.73 versus 83.84 5.25 ng/mL; p < 0.05) and Ctelopeptide (15.37 0.49 versus 18.52 0.65 ng/mL, p < 0.05) in jck mice relative to WT mice at 10 weeks of age. BV/TV was increased at 15 weeks of age (6.38% 0.26%) relative to aged-matched WT mice (5.21% 0.42%) but not in comparison to 6-week-old WT (7.00% 0.50%) or jck mice (7.64% 0.48%) (Fig. 4C). In addition, trends toward increased erosion surface (ES/BS) and osteoclast surface (Oc.S/BS) were also evident in jck mouse bones, reflecting increased osteoclast activity in association with disease progression (Table 2). Histological analysis of WT and jck bone sections from 15-week-old mice revealed the presence of peritrabecular fibrosis and increased cortical resorption consistent with features of osteitis fibrosa (Fig. 5).

Gene expression analysis supports increased osteoclast function but also reveals repression in genes associated with osteoblast function Gene expression analysis was performed on bone shafts from WT and jck mice to assess whether increased osteoclast and osteoblast function occurred as expected for a high bone remodeling state. Expression of genes whose protein products are involved in osteoclast differentiation and function (RANK, RANK ligand, cathepsin K, and tartrate resistant acid phosphatase 5 [TRAP5]) were significantly elevated by 9 weeks of age in the jck mice relative to WT mice and remained elevated throughout disease progression (Fig. 6A–D). In contrast, a general repression of genes that are involved in osteoblast differentiation and bone mineralization (tissue nonspecific alkaline phosphatase [TNAP], osteocalcin, collagen type1-a1, and the phosphoethanolamine/phosphocholine phosphatase [Phospho1]) were significantly decreased in late stages of the disease (Fig. 7A–D). Importantly, expression of the RANKL antagonist, OPG was not significantly decreased relative to WT age-matched controls (Fig. 7E). Journal of Bone and Mineral Research

The role of osteocyte function in development of renal osteodystrophy Recent evidence suggests that inactivation of osteocyte-specific b-catenin signaling may promote increased osteoclast activity and repression of genes associated with osteoblast function.(32–34) To explore the potential role of b-catenin signaling in these gene changes, we assessed the temporal expression of total and phosphorylated b-catenin, and sclerostin, a well-characterized inhibitor of the pathway, on mouse and human bone sections. Immunohistochemical analysis of jck mouse bones demonstrates that the number of osteocytes expressing phosphorylated-Ser 33/37-b-catenin relative to total b-catenin is significantly increased in jck relative to WT mice and its expression is maintained throughout disease (Fig. 8A). As shown in Fig. 8B, the number of osteocytes expressing sclerostin is significantly elevated in bone sections from jck relative to WT mice and declines late in disease (Fig. 8B). To further explore the continuous repression of b-catenin in the face of declining sclerostin expression, we examined the mRNA expression of two additional inhibitors of Wnt/b-catenin signaling, Dkk1 and sFRP4. Dkk1 mRNA levels were unchanged (data not shown), whereas sFRP4 mRNA expression began to rise by 9 weeks of age and continued to accumulate paralleling the expression of phosphorylated b-catenin (Fig. 8C). Immunostaining was performed on human bone biopsies obtained from individuals with or without renal disease to investigate the potential relevance of these findings to human disease. Relative to normal bones, an increased number of osteocytes expressing phosphorylated-Ser 33/37-b-catenin was observed in biopsies from CKD or HD patients with highest expression observed in HD patients (Fig. 9A). Sclerostin expression in normal human biopsies was also limited to a few osteocytes, with high expression observed in individuals with CKD. The number of positive osteocytes was substantially decreased in HD patients relative to CKD but were elevated relative to normal (Fig. 9B).

Discussion Mortality remains high at all stages of CKD, with the primary cause of death attributed to cardiovascular events.(15,40–47)


1763

Table 2. Mouse Bone Static and Dynamic Histomorphometry Bone parameters

WT

BV/TV (%) 6 7.00 0.50 9 8.16 0.44 12 6.78 0.52 15 5.21 0.42 OV/BV (%) 6 3.91 0.35 9 3.02 0.52 12 1.88 0.29 15 2.13 0.39 ES/BS (%) 6 3.38 0.77 9 3.17 0.68 12 3.75 1.54 15 3.98 0.61 Ob.S/BS (%) 6 22.93 1.57 9 21.30 1.69 12 28.52 2.62 15 21.07 2.44 Oc.S/BS (%) 6 3.36 0.68 9 3.25 0.88 12 4.09 1.09 15 4.58 0.60 MS/BS (%) 6 12.60 1.38 9 15.12 0.93 12 17.28 1.49 15 28.11 2.49 MAR (mm/day) 6 2.44 0.18 9 2.51 0.16 12 2.48 0.20 15 2.39 0.09 BFR/BS (mm3/mm2/yr) 6 0.35 0.03 9 0.38 0.04 12 0.43 0.06 15 0.67 0.05 MLT (day) 6 2.10 0.41 9 1.71 0.22 12 1.17 0.40 15 0.95 0.19

jck

t test

7.64 0.48 8.06 0.30 8.90 0.40 6.38 0.28

n.s. n.s.

3.48 0.41 3.31 0.32 2.77 0.59 1.25 0.48

n.s. n.s. n.s. n.s.

2.49 0.33 3.81 0.71 4.53 0.96 5.77 0.84

n.s. n.s. n.s. n.s.

18.44 1.29 27.07 1.85 25.39 2.60 23.54 3.05

n.s.

2.15 0.35 3.92 0.92 3.95 0.93 6.16 0.91

n.s. n.s. n.s. n.s.

11.67 0.93 19.43 1.32 19.95 1.97 27.96 1.75

n.s.

2.75 0.15 2.92 0.11 3.38 0.15 2.40 0.19

n.s. n.s.

0.32 0.03 0.57 0.04 0.68 0.08 0.70 0.06

n.s.

1.34 0.14 0.84 0.14 0.73 0.06 0.65 0.10

n.s. n.s.

n.s. n.s.

n.s.

n.s. n.s. n.s.

n.s. n.s.

Data expressed as mean SEM from histomorphometric analysis of isolated from animals at 6, 9, 12, or 15 weeks of age. WT ¼ wild-type; BV/TV ¼ bone volume/tissue volume; n.s. ¼ not significant; OV/BV ¼ osteoid volume/bone volume; ES/BS ¼ eroded surface/bone surface; Ob.S/BS ¼ osteoblast surface/bone surface; Oc.S/BS ¼ osteoclast osteoclast surface/bone surface; MS/BS ¼ mineralizing surface/bone surface; MAR ¼ mineral apposition rate; BFR/BS ¼ bone formation rate/bone surface; MLT ¼ mineralization lag time. p < 0.05.

1764

SABBAGH ET AL.

Treating disease at an earlier stage may have greater impact on slowing cardiovascular damage, but development of novel therapeutics is limited by an incomplete understanding of how renal dysfunction leads to cardiovascular disease. Serum phosphate has been shown to strongly correlate with these cardiovascular events, with emerging evidence that poor bone health may also contribute to phosphate imbalance.(5,7) Individuals with CKD-MBD exhibit a range of renal osteodystrophies with defects ranging from LTO to HTO bone disease.(27,48,49) Classic radiolabel studies performed in the mid-1990s suggest that at either extreme of bone turnover, there is a loss of buffering capacity such that bone cannot accept or retain the increased burden of calcium and presumably phosphate that results from reduced bone mineralization.(50,51) In the presence of failing kidneys, excess mineral not retained in bone may deposit into soft tissue. This hypothesis is supported by the large body of epidemiologic studies demonstrating that calcification can be observed early in CKD and smaller observational studies identifying early bone defects.(36,52) However, these data cannot distinguish preexisting bone disease from early changes associated with de novo pathogenesis of renal osteodystrophy. Another current challenge is the absence of clinical data from a single study that describes the progressive changes in the entire spectrum of CKD-MBD parameters. This limitation is particularly striking in regard to bone, because longitudinal clinical studies are impractical. An alternative strategy is to delineate mechanisms in a preclinical model that adequately reflects human CKD-MBD. To date, the vast number of studies in animals have used the well-characterized injury models such as the 5/6 nephrectomized or adenine-treated mouse or rat. Due to the acute and severe changes imposed on the kidney in these models and given that the average bone remodeling time in rodents is 3 weeks, it is difficult to delineate early events during the progression of CKD. Recent reports have illustrated the capacity of using progressive genetic models of kidney disease to study CKD-MBD parameters.(26,53,54) We have extended these studies by characterizing the natural progression of CKD-MBD in the jck mouse, a genetic model of PKD,(28) as a first step toward defining the mechanistic link between renal dysfunction, declining bone, and cardiovascular disease. Importantly, we confirm that changes in jck serums closely mirror those observed in clinical serum samples as reported by Craver and colleagues,(29) albeit with some differences (Fig. 1). For example, significant changes in total serum calcium content are not observed in the jck mice, whereas a decline in serum calcium is observed in stage 4 and 5 clinical samples (Fig. 1B).(29) As expected, hyperphosphatemia is a late event with significant changes relative to baseline observed in 15-week-old jck mice similar to those observed in humans (Fig. 1C). However, serum phosphate levels in mice are substantially greater than those typically observed in the clinic, because mice are not undergoing treatment with phosphate binders or dialysis. Changes in expression of the key hormones regulating phosphate and calcium are well characterized in the CKD population. FGF23 and PTH levels are also elevated in jck (Fig. 1D, E) with the induction of FGF23 occurring prior to the elevation of serum PTH as observed in human studies.(55,56) Disruption in the Journal of Bone and Mineral Research

Fig. 5. H&E and Goldner’s Trichrome stain of trabecular and cortical bone of WT and jck mice. Histologic assessment of 15-week-old jck of distal femur reveals features consistent with osteitis fibrosa, including peritrabecular fibrosis (arrows) and cortical resorption (arrowheads). There is no difference in apparent mineralization (red staining) between WT and jck as viewed by trichrome staining. n ¼ 10 mice/group. 40 images.

Fig. 6. Expression of genes associated with osteoclast function is elevated in jck mice. Real-time RT-PCR of bone shafts isolated from jck (*) versus WT (*). mRNA expression of (A) RANK, (B) RANK ligand (RANKL), (C) tartrate-resistant acid phosphatase 5 (Trap5), and (D) cathepsin K are increased relative to agematched controls starting at 9 weeks of age. n ¼ 5–16 mice/group. Results are expressed as the ratio of 1/DCt as mean SEM. p < 0.05 for jck mice versus age-matched WT.



1765

Fig. 7. Expression of genes associated with osteoblast differentiation/function is repressed late in disease progression. Real-time RT-PCR of bone shafts isolated from jck (*) versus WT (*). mRNA expression of (A) tissue nonspecific alkaline phosphatase (TNAP), (B) osteocalcin, (C) collagen1a, and (D) phospho1 are decreased, relative to age-matched controls, by 15 weeks of age. and (E) osteoprotegerin (OPG) expression remains normal. n ¼ 5–16 mice/group. Results are expressed as the ratio of 1/DCt as mean SEM. p < 0.05 for jck mice versus age-matched WT.

Fig. 8. Inactivation of b-catenin is associated with increased sclerostin expression in jck mice. (A) Immunodetection of the number of phospho-Ser33/37b-catenin–positive osteocytes was higher in jck (*) mice from 6 through 12 weeks of age relative to WT (*). Representative stain is from 9-week-old bones. Quantitative analysis reveals that the number of positive phospho-Ser33/37-b-catenin–expressing osteocytes is significantly greater in jck relative to WT. No difference in total b-catenin was observed (data not shown). (B) Immunodetection of sclerostin-positive osteocytes reveals increased staining in jck mice. Representative stain is from 9-week-old bones. The number of positive sclerostin-expressing osteocytes is significantly greater in sections from 6- and 9-week-old jck relative to WT. (C) mRNA expression of sFRP4 was significantly increased starting at 9 weeks of age in jck mice. 40 images. Results are expressed as mean SEM. p < 0.05 for jck mice versus age-matched WT.

1766

SABBAGH ET AL.


Fig. 9. Inactivation of b-catenin is associated with increased sclerostin expression in clinical bone biopsies. (A) Immunodetection of the number of phospho-Ser33/37-b-catenin osteocytes suggests b-catenin signaling decreases with disease progression. (B) Immunodetection of sclerostin-positive osteocytes reveals increased staining in CKD and HD bones relative to control bones. Expression is greatest in CKD bones. CKD represents patients in CKD stages 2 to 4; HD stage 5. Results are expressed as mean SEM. p < 0.05 for CKD or HD versus normal.

synthesis of active vitamin D relative to baseline also occurs in jck mice as observed in human serum from CKD stages 3 and 4 (Fig. 1F).(29) The observation that basal jck serum 1,25(OH)2D3 levels are elevated at the start of the study relative to WT animals likely reflects the reduced dietary calcium and elevated phosphate levels in the jck mouse diet compared to the normal chow fed to WT mice. A similar rise in serum 1,25(OH)2D3 was observed in WT mice when placed on this diet without causing bone abnormalities or vascular calcification (Supplemental Fig. 2A, C). Taken together these data demonstrate that the progressive biochemical and hormonal dysregulation in the jck mice mirror clinical changes reflecting CKD-MBD. Approximately two-thirds of the jck mice develop vascular calcification as measured by Osteosense staining and total aortic calcium content (Fig. 2B, C). Characteristic induction of the osteoblast-associated protein, osteopontin, with corresponding loss of the smooth muscle marker SM22, was also observed, demonstrating that osteogenic differentiation(57) is also contributing to the development of vascular calcification in jck (Fig. 2A). A simultaneous increase in markers of LVH, as measured by induction of ANP and BNP mRNA was also noted in concert with changes associated with cardiovascular calcification (Fig. 2E). Thus, the prominent pathological changes related to cardiac function are also disrupted in jck mice and demonstrate that these mice develop the critical endpoint associated with mortality within the CKD population. One of the complexities of mineral bone disease is the spectrum of pathologies that range from LTO bone disease (adynamic bone disease and osteomalacia) to HTO bone disease (osteitis fibrosa and mixed-uremic osteodystrophy). Histological assessment of bones from 15-week-old jck mice revealed findings consistent with osteitis fibrosa, including a relative Journal of Bone and Mineral Research

increase in osteoclast activity, the presence of cortical resorption, and peritrabecular fibrosis (Fig. 5). Longitudinal histomorphometric analysis confirmed that bone formation rates and static parameters increased as expected of HTO bone disease beginning by 9 weeks of age (Fig. 4A and Table 2). These data are consistent with previously published data, in which bone disease has been observed in both rat and human bones even in the presence of significant residual renal function.(52,53,58) 1,25(OH)2D3 is an important physiological activator of osteoclast activity.(59) We therefore tested whether the difference in basal 1,25(OH)2D3 levels found in jck and WT mice was responsible for increased osteoclast activity by performing bone histomorphometry on WT and jck bones fed either diet. In both WT and jck mice fed the lower calcium, high phosphate diet, 1,25(OH)2D3 levels were twofold greater compared to mice fed standard chow (Supplemental Fig. 2A, C). Despite equivalent elevation of 1,25(OH)2D3 levels on this specialized diet, increased osteoclast surface was observed only in jck mice (Supplemental Fig. 2D). These data indicate that the increase in osteoclast activity in jck mice cannot be solely explained by the elevated 1,25(OH)2D3 levels. Our data suggesting that increased BFRs occurred prior to detectable changes in serum PTH is in apparent conflict with the current dogma (Figs. 1C and 4A). It is well accepted that elevations in PTH are primarily responsible for the high bone turnover in the majority of individuals with late-stage CKD, and PTH measurements are used to predict HTO renal osteodystrophy.(58,60–64) Although we cannot exclude the possibility that assay sensitivity may have limited our ability to detect the earliest rise in PTH, several recent clinical observations support our preclinical result because PTH levels do not uniformly predict bone turnover when diagnosed via a


1767

bone biopsy.(65–67) These findings raise the possibility that PTH may not be the sole determinant of elevated bone turnover. Gene expression analysis was used to identify other pathways that may contribute to increased bone formation in early stages of the disease. Unexpectedly, repression of genes associated with osteoblast differentiation/function was observed at 15 weeks of age despite the presence of high bone formation rates (Fig. 7). Increased expression of genes associated with osteoclast activity occurred in association with the onset of elevated bone formation and in concert with the increased cortical resorption at 9 weeks of age (Fig. 5). Importantly, mRNA expression of RANKL, the primary regulator of osteoclast activity, was induced prior to observable changes in serum PTH in line with increased BFRs. Although we did not identify the specific cell source impacted by renal dysfunction, recent evidence has found that the osteocyte is the primary source of RANKL and its negative regulator OPG, and hence plays a major role in supporting osteoclastogenesis.(31,68) Osteocyte-specific deletion of RANKL in mice causes an osteopetrotic phenotype resulting from a significant decrease in osteoclast number and surface.(31) These data extend previous reports demonstrating that osteocyte-specific deletion of b-catenin is associated with an increase in the RANKL/OPG ratio, thereby supporting a role of this signaling pathway in balancing the expression of these opposing factors.(34) In vitro experiments using primary and established osteocyte cells have demonstrated that the Wnt inhibitor, sclerostin, stimulates RANKL expression, leading to an increase in the RANKL/OPG mRNA ratio favoring the formation of TRAPþ multinucleated cells.(68) These studies are also in agreement with our data demonstrating that the increased sclerostin expression observed early in CKD-MBD leads to a concomitant increase in RANKL expression that may drive osteoclastogenesis, which was associated with increased bone formation rates in the jck mouse. Recent evidence has also linked loss of b-catenin signaling in osteocytes with corresponding decreases in genes associated with osteoblast function.(32–34) Furthermore, serum levels of the Wnt/b-catenin antagonists, sclerostin (SOST) and Dickkopft-1 (DKK1), were recently reported to be elevated in individuals on dialysis compared to non-CKD individuals, raising the possibility that repression of the b-catenin signaling pathway may also occur in the setting of CKD.(69) The importance of b-catenin signaling on bone mass and remodeling has been well studied. In addition to regulation of osteoclast activity, b-catenin signaling promotes osteogenesis of mesenchymal stem cells, increased osteoblast proliferation, mineralization, and decreased osteoblast apoptosis.(70) In the Wnt canonical pathway, decrease in Wnt activity leads to phosphorylation and degradation of b-catenin. To assess the relative impact of the pathway on osteoclast activation, we examined the number of osteocytes that stained for total versus phosphorylated b-catenin in both jck mouse bones and clinical bone biopsies. These data were consistent with the gene expression data, confirming that phosphorylated b-catenin but not total b-catenin is elevated, supporting a progressive inhibition of signaling during the course of CKD (Figs. 8B and 9B). Consistent with these findings, the number of cells expressing sclerostin was also induced at early stages of CKD

1768

SABBAGH ET AL.

(Figs. 8A and 9A). However, this finding could not be totally explained by changes in sclerostin expression, because we found the highest phosphorylated b-catenin expression in late CKD, whereas the peak sclerostin expression was detected in early CKD in both mouse and human bones (Fig. 8). Importantly, declining sclerostin was also correlated with a simultaneous increase in mRNA expression of an additional WNT/b-catenin antagonist, sFRP4, but not Dkk1 in jck mice. These data suggest that the progressive pathology of CKD maintains pressure on bone to reduce osteocyte-specific b-catenin signaling, increase RANKL expression and subsequent osteoclast activity. The limited analysis in human biopsies raises the possibility that repression of osteocyte b-catenin signaling may be implicated irrespective of the type of renal osteodystrophy. In comparison to observations in 10 normal human bone samples, increases in sclerostin and phosphorylated b-catenin–expressing cells were found across 10 CKD and 10 HD biopsies that included equivalent numbers of HTO and LTO biopsies. Subgroup analysis revealed a similar increase in the expression of sclerostin and phosphorylated b-catenin in osteocytes between biopsies from patients with HTO versus LTO disease relative to normal controls. However, due to the limited number of biopsies evaluated, this precludes a definitive conclusion. One potentially relevant observation is that increased osteoclast activity was a common finding irrespective of the type of renal osteodystrophy.(36,71) Additional studies will be required to confirm this hypothesis. The finding that sclerostin is induced early is also consistent with the observation that bone formation may be initiated prior to PTH induction. As PTH is a known inhibitor of SOST transcription, sclerostin would not be expected to be induced if PTH was the initial driver for elevated bone formation.(72–74) Indeed, subsequent decline in sclerostin later in disease progression may be attributed to the rising levels of PTH (Figs. 1D and 8B). Intermittent but not continuous PTH has also been shown to induce b-catenin signaling by directly binding to LRP6/frizzled complex.(75) It is assumed that hyperparathyroidism is analogous to continuous PTH administration such that these findings using intermittent PTH would not be relevant in the setting of CKD.(75) In support of this, transgenic mice expressing constitutively active PTH receptors have a bone phenotype similar to osteitis fibrosa.(74) However, it is also possible that elevations in sclerostin may contribute to repression of PTH signaling in CKD bones because PTH responses are blunted in transgenic mice overexpressing sclerostin.(76) Finally, the underlying defect in jck mice is due to activating mutations in Nek8, a gene encoding a serine protease kinase that promotes protein trafficking of two cilia-associated proteins, polycystin 1 and 2 (PC1 and PC2).(77) Activating Nek8 mutations lead to increased primary cilia length in the kidney,(28,77) whereas loss of function mutations in PKD1 induce PKD, skeletal defects in vivo, and decreased osteoblast differentiation in vitro.(78,79) As primary cilia may be part of the mechanosensing machinery that influences bone remodeling via b-catenin signaling,(80) NEK8 mutations in jck could theoretically influence the onset of HTO bone disease. However, tibia and femur length and total body weight were not different between 6-week-old WT and jck mice before the onset of renal dysfunction indicating that the mutation had no influence on normal growth (data not shown). Journal of Bone and Mineral Research

Fig. 10. Unified model demonstrating that early in disease, sclerostin inhibits b-catenin pathway and increases osteoclast activity. Late in disease, elevated PTH levels contribute to the high turnover and the decrease in sclerostin levels. However, sFRP4 levels rise with disease progression leading to continuous b-catenin repression and possibly affecting osteoblast function.

Second, differentiation of bone marrow stromal cells from 6-week-old mice was identical (Fig. 3A, B). Finally, BMD and overall strength were identical in 6-week-old WT and jck mice (Fig. 3C–E). These observations suggest that observed differences in bone likely reflect changes associated with renal dysfunction and do not support a role of cilia or NEK8 in the development of renal osteodystrophy in the jck mouse. Therefore, the jck mouse provides a reasonable model that replicates salient features of human CKD-MBD with associated HTO renal osteodystrophy. A unifying model to explain our findings in jck mice is presented in Fig. 10. One of the earliest changes in CKD-MBD is the induction of the Wnt inhibitor, SOST, with corresponding repression of b-catenin signaling, which promotes an increase in the ratio of RANKL to OPG and subsequent activation of osteoclasts. Osteoblast differentiation may temporarily increase in response to coupling with osteoclast activity. In late stages, increased PTH represses SOST but additional Wnt antagonists such as sFRP4 are induced, sustaining repression of b-catenin and maintaining the high osteoclast activity. Our data confirms that in jck mice, HTO renal osteodystrophy occurs early in the disease progression and suggests that there is a repression of the b-catenin pathway in osteocytes that is maintained even in the presence of high BFRs. The data further raises the possibility that PTH elevation may not be the only contributor to HTO because induction of BFRs occurs even before significant changes in PTH are observed through b-catenin modulation of RANKL/OPG ratio and increased osteoclast activity. These data demonstrate the complexity of changes associated with bone during the progression of CKD and underscores the need for further investigation to optimize treatment strategies for CKD-MBD.

Disclosures YS, SOB, WT, SR, LP, JB, WS, CB, SL, SL, and SCS are employees of Genzyme, a Sanofi Company. FGG, LMR, VJ, and RMM have received research funding from Genzyme. Journal of Bone and Mineral Research

Acknowledgments We thank Drs. Stephen Harris and Roland Baron for helpful discussions related to the work presented in this manuscript. We also thank, from Genzyme, William Weber for imaging analysis, Leah Curtin for colony management, and Michael Phipps and Matthew DeRiso for genotyping. RMAM, VJ, ABC, and MEC were supported by CNPQ, Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico. FGG was supported by Fundaçaõ de Amparo a` Pesquisa do Estado de Saõ Paulo (FAPESP). Authors’ roles: RMAM, VJ, SLe, and SCS contributed to the project conception. YS, FGG, SOB, WT, RMAM, and SCS developed specific experimental strategies. YS, FGG, SOB, WT, LMD, SR, LP, JB, WS, CB, SLiu, PD, MEC, ABC, and RMAM contributed to the acquisition and analysis of data. YS, WT, SLiu, FGG, PD, RMAM, VJ, and SCS contributed to the interpretation of the data. All authors participated in revising the manuscript and approved manuscript submission.

References 1. Coresh J, Selvin E, Stevens LA, Manzi J, Kusek JW, Eggers P, Van Lente F, Levey AS. Prevalence of chronic kidney disease in the United States. JAMA. 2007;298:2038–47. 2. Eknoyan G, Lameire N, Barsoum R, Eckardt KU, Levin A, Levin N, Locatelli F, MacLeod A, Vanholder R, Walker R, Wang H. The burden of kidney disease: improving global outcomes. Kidney Int. 2004; 66:1310–4. 3. Keith DS, Nichols GA, Gullion CM, Brown JB, Smith DH. Longitudinal follow-up and outcomes among a population with chronic kidney disease in a large managed care organization. Arch Intern Med. 2004;164:659–63. 4. Watanabe R, Lemos MM, Manfredi SR, Draibe SA, Canziani ME. Impact of cardiovascular calcification in nondialyzed patients after 24 months of follow-up. Clin J Am Soc Nephrol. 2010;5: 189–94.


1769

5. Moe S, Drueke T, Cunningham J, Goodman W, Martin K, Olgaard K, Ott S, Sprague S, Lameire N, Eknoyan G. Definition, evaluation, and classification of renal osteodystrophy: a position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int. 2006;69:1945–53.

23. Barreto DV, Barreto Fde C, Carvalho AB, Cuppari L, Draibe SA, Dalboni MA, Moyses RM, Neves KR, Jorgetti V, Miname M, Santos RD, Canziani ME. Association of changes in bone remodeling and coronary calcification in hemodialysis patients: a prospective study. Am J Kidney Dis. 2008;52:1139–50.

6. Moe SM. Definition and classification of renal osteodystrophy and chronic kidney disease-mineral bone disorder (CKD-MBD). In: Olgaard K, Salusky IB, Silver J, editors. The spectrum of mineral and bone disorders in chronic kidney disease. New York: Oxford University; 2010. p 1–14.

24. Giachelli CM. Inducers and inhibitors of biomineralization: lessons from pathological calcification. Orthod Craniofac Res. 2005;8:229–31.

7. Malluche HH, Mawad H, Monier-Faugere MC. The importance of bone health in end-stage renal disease: out of the frying pan, into the fire? Nephrol Dial Transplant. 2004;19(Suppl 1):i9–13.

26. Moe SM, Chen NX, Seifert MF, Sinders RM, Duan D, Chen X, Liang Y, Radcliff JS, White KE, Gattone VH 2nd. A rat model of chronic kidney disease-mineral bone disorder. Kidney Int. 2009;75:176– 84.

8. Hruska KA, Mathew S, Lund RJ, Memon I, Saab G. The pathogenesis of vascular calcification in the chronic kidney disease mineral bone disorder: the links between bone and the vasculature. Semin Nephrol. 2009;29:156–65. 9. Fahrleitner-Pammer A, Herberth J, Browning SR, Obermayer-Pietsch B, Wirnsberger G, Holzer H, Dobnig H, Malluche HH. Bone markers predict cardiovascular events in chronic kidney disease. J Bone Miner Res. 2008;23:1850–8. 10. London GM, Marty C, Marchais SJ, Guerin AP, Metivier F, de Vernejoul MC. Arterial calcifications and bone histomorphometry in end-stage renal disease. J Am Soc Nephrol. 2004;15:1943–51. 11. Boukhris R, Becker KL. Calcification of the aorta and osteoporosis. A roentgenographic study. JAMA. 1972;219:1307–11. 12. Banks LM, Lees B, MacSweeney JE, Stevenson JC. Effect of degenerative spinal and aortic calcification on bone density measurements in post-menopausal women: links between osteoporosis and cardiovascular disease? Eur J Clin Invest. 1994;24:813–7. 13. Hak AE, Pols HA, van Hemert AM, Hofman A, Witteman JC. Progression of aortic calcification is associated with metacarpal bone loss during menopause: a population-based longitudinal study. Arterioscler Thromb Vasc Biol. 2000;20:1926–31. 14. Kiel DP, Kauppila LI, Cupples LA, Hannan MT, O’Donnell CJ, Wilson PW. Bone loss and the progression of abdominal aortic calcification over a 25 year period: the Framingham Heart Study. Calcif Tissue Int. 2001;68:271–6. 15. Braun J, Oldendorf M, Moshage W, Heidler R, Zeitler E, Luft FC. Electron beam computed tomography in the evaluation of cardiac calcification in chronic dialysis patients. Am J Kidney Dis. 1996;27: 394–401. 16. Ho LT, Sprague SM. Renal osteodystrophy in chronic renal failure. Semin Nephrol. 2002;22:488–93. 17. Price PA, Buckley JR, Williamson MK. The amino bisphosphonate ibandronate prevents vitamin D toxicity and inhibits vitamin Dinduced calcification of arteries, cartilage, lungs and kidneys in rats. J Nutr. 2001;131:2910–5. 18. Price PA, Faus SA, Williamson MK. Bisphosphonates alendronate and ibandronate inhibit artery calcification at doses comparable to those that inhibit bone resorption. Arterioscler Thromb Vasc Biol. 2001; 21:817–24. 19. Price PA, June HH, Buckley JR, Williamson MK. Osteoprotegerin inhibits artery calcification induced by warfarin and by vitamin D. Arterioscler Thromb Vasc Biol. 2001;21:1610–6. 20. Lomashvili KA, Monier-Faugere MC, Wang X, Malluche HH, O’Neill WC. Effect of bisphosphonates on vascular calcification and bone metabolism in experimental renal failure. Kidney Int. 2009;75:617–25. 21. Gonzalez EA, Lund RJ, Martin KJ, McCartney JE, Tondravi MM, Sampath TK, Hruska KA. Treatment of a murine model of highturnover renal osteodystrophy by exogenous BMP-7. Kidney Int. 2002;61:1322–31. 22. Lund RJ, Davies MR, Brown AJ, Hruska KA. Successful treatment of an adynamic bone disorder with bone morphogenetic protein-7 in a renal ablation model. J Am Soc Nephrol. 2004;15:359–69.

1770

SABBAGH ET AL.

25. Wallin R, Wajih N, Greenwood GT, Sane DC. Arterial calcification: a review of mechanisms, animal models, and the prospects for therapy. Med Res Rev. 2001;21:274–301.

27. Kidney Disease: Improving Global Outcomes (KDIGO) CKD-MBD Work Group. KDIGO clinical practice guideline for the diagnosis, evaluation, prevention, and treatment of Chronic Kidney Disease-Mineral and Bone Disorder (CKD-MBD). Kidney Int Suppl. 2009; (113): S1–130. 28. Smith LA, Bukanov NO, Husson H, Russo RJ, Barry TC, Taylor AL, Beier DR, Ibraghimov-Beskrovnaya O. Development of polycystic kidney disease in juvenile cystic kidney mice: insights into pathogenesis, ciliary abnormalities, and common features with human disease. J Am Soc Nephrol. 2006;17:2821–31. 29. Craver L, Marco MP, Martıńez I, Rue M, Borra`s M, Martıń ML, Sarro´ F, Valdivielso JM, Fernańdez E. Mineral metabolism parameters throughout chronic kidney disease stages 1-5–achievement of K/DOQI target ranges. Nephrol Dial Transplant. 2007;22:1171–6. 30. Bonewald LF. The amazing osteocyte. J Bone Miner Res. 2011;26: 229–38. 31. Nakashima T, Hayashi M, Fukunaga T, Kurata K, Oh-Hora M, Feng JQ, Bonewald LF, Kodama T, Wutz A, Wagner EF, Penninger JM, Takayanagi H. Evidence for osteocyte regulation of bone homeostasis through RANKL expression. Nat Med. 2011;17:1231–4. 32. Glass DA 2nd, Bialek P, Ahn JD, Starbuck M, Patel MS, Clevers H, Taketo MM, Long F, McMahon AP, Lang RA, Karsenty G. Canonical Wnt signaling in differentiated osteoblasts controls osteoclast differentiation. Dev Cell. 2005;8:751–64. 33. Holmen SL, Zylstra CR, Mukherjee A, Sigler RE, Faugere MC, Bouxsein ML, Deng L, Clemens TL, Williams BO. Essential role of beta-catenin in postnatal bone acquisition. J Biol Chem. 2005;280:21162–8. 34. Kramer I, Halleux C, Keller H, Pegurri M, Gooi JH, Weber PB, Feng JQ, Bonewald LF, Kneissel M. Osteocyte Wnt/beta-catenin signaling is required for normal bone homeostasis. Mol Cell Biol. 2010;30: 3071–85. 35. Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, Ott SM, Recker RR. Bone histomorphometry: standardization of nomenclature, symbols, and units. Report of the ASBMR Histomorphometry Nomenclature Committee. J Bone Miner Res. 1987;2:595– 610. 36. Tomiyama C, Carvalho AB, Higa A, Jorgetti V, Draibe SA, Canziani ME. Coronary calcification is associated with lower bone formation rate in CKD patients not yet in dialysis treatment. J Bone Miner Res. 2010;25:499–504. 37. Barreto DV, Barreto FC, Carvalho AB, Cuppari L, Cendoroglo M, Draibe SA, Moyses RM, Neves KR, Jorgetti V, Blair A, Guiberteau R, Fernandes Canziani ME. Coronary calcification in hemodialysis patients: the contribution of traditional and uremia-related risk factors. Kidney Int. 2005;67:1576–82. 38. Barreto DV, Barreto Fde C, de Carvalho AB, Cuppari L, Draibe SA, Dalboni MA, Moyses RM, Neves KR, Jorgetti V, Miname M, Santos RD, Canziani ME. Phosphate binder impact on bone remodeling and coronary calcification–results from the BRiC study. Nephron Clin Pract. 2008;110:c273–83.


39. Gomes SA, dos Reis LM, de Oliveira IB, Noronha IL, Jorgetti V, Heilberg IP. Usefulness of a quick decalcification of bone sections embedded in methyl methacrylate[corrected]: an improved method for immunohistochemistry. J Bone Miner Metab. 2008;26:110–3. 40. Mehrotra R, Adler S. Coronary artery calcification in nondialyzed patients with chronic kidney diseases. Am J Kidney Dis. 2005;45:963. 41. Mehrotra R, Budoff M, Christenson P, Ipp E, Takasu J, Gupta A, Norris K, Adler S. Determinants of coronary artery calcification in diabetics with and without nephropathy. Kidney Int. 2004;66: 2022–31. 42. Menon V, Sarnak MJ. The epidemiology of chronic kidney disease stages 1 to 4 and cardiovascular disease: a high-risk combination. Am J Kidney Dis. 2005;45:223–32. 43. Sarnak MJ, Levey AS, Schoolwerth AC, Coresh J, Culleton B, Hamm LL, McCullough PA, Kasiske BL, Kelepouris E, Klag MJ, Parfrey P, Pfeffer M, Raij L, Spinosa DJ, Wilson PW. Kidney disease as a risk factor for development of cardiovascular disease: a statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation. 2003;108:2154–69. 44. Foley RN, Parfrey PS, Sarnak MJ. Epidemiology of cardiovascular disease in chronic renal disease. J Am Soc Nephrol. 1998;9:S16–23. 45. Go AS, Chertow GM, Fan D, McCulloch CE, Hsu CY. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med. 2004;351:1296–305. 46. Moe SM, Chen NX. Pathophysiology of vascular calcification in chronic kidney disease. Circ Res. 2004;95:560–7. 47. Weiner DE, Tighiouart H, Stark PC, Amin MG, MacLeod B, Griffith JL, Salem DN, Levey AS, Sarnak MJ. Kidney disease as a risk factor for recurrent cardiovascular disease and mortality. Am J Kidney Dis. 2004;44:198–206. 48. Parfitt AM. Renal bone disease: a new conceptual framework for the interpretation of bone histomorphometry. Curr Opin Nephrol Hypertens. 2003;12:387–403. 49. Felsenfeld A, Torres A. Bone histomorphometry in renal osteodystrophy. In: Olgaard K, Silver J, Salusky IB, editors. The spectrum of mineral and bone disorders in chronic kidney disease. 2nd edition. New York: Oxford University Press; 2010. 50. Kurz P, Tsobanelis T, Brunkhorst R, Roth P, Werner E, Schoeppe W, Vlachojannis J. Calcium kinetic studies in patients on CAPD: improvement of secondary hyperparathyroidism without concomitant improvement of calcium turnover. Perit Dial Int. 1997;17: 59–65. 51. Parfitt AM. Misconceptions (3): calcium leaves bone only by resorption and enters only by formation. Bone. 2003;33:259–63. 52. Malluche HH, Ritz E, Lange HP, Kutschera L, Hodgson M, Seiffert U, Schoeppe W. Bone histology in incipient and advanced renal failure. Kidney Int. 1976;9:355–62. 53. Moe SM, Radcliffe JS, White KE, Gattone VH, Seifert MF, Chen X, Aldridge B, Chen NX. The pathophysiology of early stage chronic kidney disease-mineral bone disorder (CKDMBD) and response to phosphate binders in the rat. J Bone Miner Res. 2011 Nov; 26(11): 2672–81. 54. Stubbs JR, He N, Idiculla A, Gillihan R, Liu S, David V, Hong Y, Quarles LD. Longitudinal evaluation of FGF23 changes and mineral metabolism abnormalities in a mouse model of chronic kidney disease. J Bone Miner Res. Epub 2011 Sep 28. DOI: 10.1002/ jbmr.516. 55. Gutierrez O, Isakova T, Rhee E, Shah A, Holmes J, Collerone G, Juppner H, Wolf M. Fibroblast growth factor-23 mitigates hyperphosphatemia but accentuates calcitriol deficiency in chronic kidney disease. J Am Soc Nephrol. 2005;16:2205–15. 56. Pavik I, Jaeger P, Kistler AD, Poster D, Krauer F, Cavelti-Weder C, Rentsch KM, Wuthrich RP, Serra AL. Patients with autosomal


dominant polycystic kidney disease have elevated fibroblast growth factor 23 levels and a renal leak of phosphate. Kidney Int. 2011; 79:234–40. 57. Giachelli CM, Speer MY, Li X, Rajachar RM, Yang H. Regulation of vascular calcification: roles of phosphate and osteopontin. Circ Res. 2005;96:717–22. 58. Martin KJ, Olgaard K, Coburn JW, Coen GM, Fukagawa M, Langman C, Malluche HH, McCarthy JT, Massry SG, Mehls O, Salusky IB, Silver JM, Smogorzewski MT, Slatopolsky EM, McCann L. Diagnosis, assessment, and treatment of bone turnover abnormalities in renal osteodystrophy. Am J Kidney Dis. 2004;43:558–65. 59. Dusso AS, Brown AJ, Slatopolsky E. Vitamin D. Am J Physiol Renal Physiol. 2005;289:F8–28. 60. Solal ME, Sebert JL, Boudailliez B, Marie A, Moriniere P, Gueris J, Bouillon R, Fournier A. Comparison of intact, midregion, and carboxy terminal assays of parathyroid hormone for the diagnosis of bone disease in hemodialyzed patients. J Clin Endocrinol Metab. 1991; 73:516–24. 61. Andress DL. Bone and mineral guidelines for patients with chronic kidney disease: a call for revision. Clin J Am Soc Nephrol. 2008;3: 179–83. 62. Wang M, Hercz G, Sherrard DJ, Maloney NA, Segre GV, Pei Y. Relationship between intact 1-84 parathyroid hormone and bone histomorphometric parameters in dialysis patients without aluminum toxicity. Am J Kidney Dis. 1995;26:836–44. 63. Qi Q, Monier-Faugere MC, Geng Z, Malluche HH. Predictive value of serum parathyroid hormone levels for bone turnover in patients on chronic maintenance dialysis. Am J Kidney Dis. 1995;26:622–31. 64. Torres A, Lorenzo V, Hernandez D, Rodriguez JC, Concepcion MT, Rodriguez AP, Hernandez A, de Bonis E, Darias E, Gonzalez-Posada JM, et al. Bone disease in predialysis, hemodialysis, and CAPD patients: evidence of a better bone response to PTH. Kidney Int. 1995;47:1434–42. 65. Barreto FC, Barreto DV, Moyses RM, Neves KR, Canziani ME, Draibe SA, Jorgetti V, Carvalho AB. K/DOQI-recommended intact PTH levels do not prevent low-turnover bone disease in hemodialysis patients. Kidney Int. 2008;73:771–7. 66. Ferreira A, Frazao JM, Monier-Faugere MC, Gil C, Galvao J, Oliveira C, Baldaia J, Rodrigues I, Santos C, Ribeiro S, Hoenger RM, Duggal A, Malluche HH. Effects of sevelamer hydrochloride and calcium carbonate on renal osteodystrophy in hemodialysis patients. J Am Soc Nephrol. 2008;19:405–12. 67. Drueke TB. Is parathyroid hormone measurement useful for the diagnosis of renal bone disease? Kidney Int. 2008;73:674–6. 68. Wijenayaka AR, Kogawa M, Lim HP, Bonewald LF, Findlay DM, Atkins GJ. Sclerostin stimulates osteocyte support of osteoclast activity by a RANKL-dependent pathway. PLoS One. 2011;6:e25900. 69. Cejka D, Herberth J, Branscum AJ, Fardo DW, Monier-Faugere MC, Diarra D, Haas M, Malluche HH. Sclerostin and Dickkopf-1 in renal osteodystrophy. Clin J Am Soc Nephrol. 2011;6:877–82. 70. Case N, Rubin J. Beta-catenin–a supporting role in the skeleton. J Cell Biochem. 2010;110:545–53. 71. Rocha LA, Higa A, Barreto FC, dos Reis LM, Jorgetti V, Draibe SA, Carvalho AB. Variant of adynamic bone disease in hemodialysis patients: fact or fiction? Am J Kidney Dis. 2006;48:430–6. 72. Bellido T, Ali AA, Gubrij I, Plotkin LI, Fu Q, O’Brien CA, Manolagas SC, Jilka RL. Chronic elevation of parathyroid hormone in mice reduces expression of sclerostin by osteocytes: a novel mechanism for hormonal control of osteoblastogenesis. Endocrinology. 2005;146: 4577–83. 73. Keller H, Kneissel M. SOST is a target gene for PTH in bone. Bone. 2005;37:148–58. 74. O’Brien CA, Plotkin LI, Galli C, Goellner JJ, Gortazar AR, Allen MR, Robling AG, Bouxsein M, Schipani E, Turner CH, Jilka RL, Weinstein RS,


1771

Manolagas SC, Bellido T. Control of bone mass and remodeling by PTH receptor signaling in osteocytes. PLoS One. 2008;3: e2942. 75. Wan M, Yang C, Li J, Wu X, Yuan H, Ma H, He X, Nie S, Chang C, Cao X. Parathyroid hormone signaling through low-density lipoproteinrelated protein 6. Genes Dev. 2008;22:2968–79.

78. Xiao Z, Zhang S, Magenheimer BS, Luo J, Quarles LD. Polycystin-1 regulates skeletogenesis through stimulation of the osteoblastspecific transcription factor RUNX2-II. J Biol Chem. 2008;283: 12624–34.

76. Kramer I, Loots GG, Studer A, Keller H, Kneissel M. Parathyroid hormone (PTH)-induced bone gain is blunted in SOST overexpressing and deficient mice. J Bone Miner Res. 2010;25:178–89.

79. Xiao Z, Zhang S, Mahlios J, Zhou G, Magenheimer BS, Guo D, Dallas SL, Maser R, Calvet JP, Bonewald L, Quarles LD. Cilia-like structures and polycystin-1 in osteoblasts/osteocytes and associated abnormalities in skeletogenesis and Runx2 expression. J Biol Chem. 2006; 281:30884–95.

77. Sohara E, Luo Y, Zhang J, Manning DK, Beier DR, Zhou J. Nek8 regulates the expression and localization of polycystin-1 and polycystin-2. J Am Soc Nephrol. 2008;19:469–76.

80. Wallingford JB, Mitchell B. Strange as it may seem: the many links between Wnt signalling, planar cell polarity, and cilia. Genes Dev. 2011;25:201–13.

1772

SABBAGH ET AL.
