SEARCH

SEARCH BY CITATION

Keywords:

  • Telomerase;
  • Telomeres;
  • Mesenchymal Stem Cells;
  • Osteoblasts;
  • Bone;
  • Aging;
  • Osteoporosis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Telomere shortening owing to telomerase deficiency leads to accelerated senescence of human skeletal (mesenchymal) stem cells (MSCs) in vitro, whereas overexpression leads to telomere elongation, extended life span, and enhanced bone formation. To study the role of telomere shortening in vivo, we studied the phenotype of telomerase-deficient mice (Terc−/−). Terc−/− mice exhibited accelerated age-related bone loss starting at 3 months of age and during 12 months of follow-up revealed by dual-energy X-ray absorptiometric (DXA) scanning and by micro–computed tomography (µCT). Bone histomorphometry revealed decreased mineralized surface and bone-formation rate as well as increased osteoclast number and size in Terc−/− mice. Also, serum total deoxypyridinoline (tDPD) was increased in Terc−/− mice. MSCs and osteoprogenitors isolated from Terc−/− mice exhibited intrinsic defects with reduced proliferating cell number and impaired osteogenic differentiation capacity. In addition, the Terc−/−-MSC cultures accumulated a larger proportion of senescent β-galactosidase+ cells and cells exhibiting DNA damage. Microarray analysis of Terc−/− bone revealed significant overexpression of a large number of proinflammatory genes involved in osteoclast (OC) differentiation. Consistently, serum obtained from Terc−/− mice enhanced OC formation of wild-type bone marrow cultures. Our data demonstrate two mechanisms for age-related bone loss caused by telomerase deficiency: intrinsic osteoblastic defects and creation of a proinflammatory osteoclast-activating microenvironment. Thus telomerization of MSCs may provide a novel approach for abolishing age-related bone loss. © 2011 American Society for Bone and Mineral Research.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Osteoporosis is a multifactorial and complex disorder characterized by bone fragility and increased risk for fractures.1 One of the principal pathophysiologic mechanisms underlying osteoporosis is the impairment of osteoblastic bone formation during bone remodeling.2 Bone formation is carried out by osteoblastic cells recruited to the bone-remodeling sites from progenitor and stem cells in the bone marrow, where they undergo several rounds of proliferation and differentiation to create the “osteoblastic team” needed for bone formation. Dynamic histomorphometric studies suggest that both the number and activities of the osteoblastic team are affected in aged osteoporotic patients with fragility fractures.3 Several possible cellular and molecular mechanisms can mediate the defective osteoblastic functions, including cell-autonomous intrinsic defects4 and defective microenvironment owing to age-related changes in the humoral milieu in bone microenvironment.5

One of the principal mechanisms responsible for age-related defective cellular functions, known collectively as cellular senescence, is telomere shortening caused by incomplete replication of linear chromosomes by DNA polymerase.6 Telomere length is maintained by telomerase, a ribonuclear protein complex consisting of an integral RNA (hTR) that serves as the telomeric template, a catalytic subunit (hTERT) that has reverse-ranscriptase activity, and associated protein components.7, 8 The absence of telomerase activity in somatic cells leading to telomere shortening has been implicated in the pathophysiology of several age-related diseases and premature aging syndromes.9

We have previously performed a number of in vitro studies to demonstrate the role of telomere shortening in bone biology. We have reported that skeletal (stromal, mesenchymal) stem cells (MSCs) lack telomerase activity10 and exhibit telomere shortening in association with a replicative senescence phenotype in long-term culture.4 Also, MSC cultures established from elderly donors exhibited impaired cell proliferation, accumulation of senescent cells, and shorter replicative life span in vitro.5 We also have demonstrated that retelomerization of MSCs through overexpression of hTERT leads to elongation of telomeres of MSCs and extended in vitro lifespan. In addition, telomerized cells maintain their “stemness” characteristics, and their bone-forming abilities are enhanced based on in vitro and in vivo criteria.10 These data suggest telomere shortening and telomerase activity are important factors in controlling the biologic functions of MSCs and osteoblastic cells. However, the in vivo consequences of defective telomerase activity and shortened telomeres on bone stem cells and bone biology are poorly studied.

Several animal models are currently available for studying the link between telomere length, tissue turnover, homeostasis, and diverse pathology resulting from telomere dysfunction.11–14 One of the most characterized models is the Terc-deficient (Terc−/−) mouse, which has been generated by elimination of the murine Terc gene.15 These mice are viable, but only a limited number of generations can be derived before loss of viability is observed.16 The late generation Terc−/− mice possess shortened telomeres and exhibit an accelerated aging phenotype, including immune senescence, impaired hematopoiesis, intestinal atrophy, and infertility.13, 14 In this study, we investigated the postnatal bone phenotype of Terc−/− mice and the effects of telomere shortening on bone stem cell populations in vivo. Our findings demonstrate that Terc−/− mice exhibit an age-related bone-loss phenotype owing to intrinsic defects in MSCs and osteoblastic (OB) cell compartments as well as enhanced osteoclastogenesis owing to the presence of a proinflammatory microenvironment within the skeleton.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Mice breeding, genotyping, and handling

Terc-deficient mice (Terc−/−; strain 004132) were purchased from Jackson Laboratories (Bar Harbor, ME, USA) and kept in a pathogen-free environment on standard chow. Terc−/− mice were intercrossed to generate third-generation Terc−/− (Terc−/−-G3) mice and were maintained in a C57BL/6J background. Wild-type Terc+/+ mice we employed as controls. Genotyping was performed according to the protocol recommended by Jackson Laboratories. NOD/MrkBomTac-Prkdcscid mice (NOD/SCID mice) were purchased from Taconic M&B (Ry, Denmark). Mouse experiments were approved by the Danish Animal Ethical Committee.

Skeletal staining and X-ray analysis

Embryos (17.5 days old) were deskinned and eviscerated, fixed in 100% ethanol for 4 days, and then transferred into acetone for 1 to 2 days. Staining was performed in a solution of 90% ethanol, 5% acetic acid, and 5% H2O supplemented with 0.005% alizarin red S (Sigma, St Louis, MO, USA) and 0.015% alcian blue 8GS (Sigma) for 3 days at 37°C. Samples were rinsed in water and cleared for 3 days in 1% potassium hydroxide followed by clearing in 0.8% KOH–20% glycerol for 1 week. Samples then were transferred into 100% glycerol for long-term storage. For X-ray analysis, 32-week-old mice, both wild type and Terc−/−, were anesthetized, and images were taken and developed.

Body mass and body composition measurements

Total body mass (grams), bone mineral content (milligrams), and bone mineral density (milligrams per square centimeter) were measured using a dual-energy X-ray absorptiometric (DXA) scanner. Mice were weighed before DXA scanning, were given gas anesthesia, and were scanned with Lunar PIXImus (Version 1.44; Lunar, Copenhagen, Denmark) calibrated using a phantom model with 0.058 g/cm2 bone mineral density (BMD).

Histomorphometry

All histomorphometric analyses were performed according to a protocol described previously.17 Briefly, skeletons were fixed in 3.7% PBS–buffered formaldehyde for 18 hours at 4°C. After 24 hours of incubation in 70% ethanol, the lumbar vertebral bodies (L3–L5) and one tibia of each mouse were dehydrated in ascending alcohol concentrations and embedded in methyl methacrylate. Sections of 5 µm were cut in the sagittal plane on a Microtec rotation microtome (Techno-Med, Munich, Germany). These sections were stained by toluidine blue and by the van Gieson/von Kossa procedure. Nonstained sections of 12 µm were used to determine the bone-formation rate. Parameters of static and dynamic histomorphometry were quantified on toluidine blue–stained undecalcified proximal tibia and lumbar vertebral sections of 5 µm. Analysis of bone volume, trabecular number, trabecular spacing, and trabecular thickness and determination of osteoblast and osteoclast numbers and surface were carried out according to standardized protocols using the Osteo-Measure Histomorphometry System (Osteometrics, Atlanta, GA, USA). Fluorochrome measurements for determination of the bone-formation rate were performed on two nonconsecutive 12-µm sections for each animal. Statistical differences between the groups (n = 5) were assessed by the Student's t test.

Cell culture

Isolation and expansion of primary bone marrow mMSCs

Bone marrow cells were harvested according to the protocol described by Peister18 with some in-house modifications.24 The medium was changed for every third day; subsequently, after 1 to 2 weeks, cells were dissociated using trypsin/EDTA for 4 minutes at 37°C and plated according to the experimental setup.

Neonatal calvaria cell isolation and culture

Neonatal calvaria cells were isolated from 3-day-old pups according to the protocol used in our laboratory.20 Briefly, calvaria were isolated under aseptic conditions and submitted to sequential collagenase II digestion at 37°C. Dulbecco's modified Eagle's medium (DMEM; Cat. No. 31966, Gibco BRL, Carlsbad, CA, USA). Cells were cultured in DMEM with 20% fetal bovine serum (FBS) and 100 µg/mL of streptomycin (Gibco) and 100 U/mL of penicillin (Gibco) and switched to osteogenic medium for differentiation.

Differentiation assays

Osteoblast differentiation

For osteogenic differentiation of MSCs, cells were plated at high densities (20 × 103 cells/cm2) in 24-well plates containing complete expansion medium (CEM), Iscove modified Dulbecco medium (IMDM; Cat. No. 21980, Gibco) containing 12% FBS (Gibco), 100 U/mL of penicillin (Gibco), 100 µg/mL of streptomycin (Gibco), and 12 µM L-glutamine (Cat. No. 25030, Gibco), supplemented with osteogenic cocktail [10 nM dexamethasone (Sigma), 10 mM β-glycerol phosphate (Sigma), and 50 µg/mL of vitamin C (Sigma)]. The medium was changed every third day until day 16, and then cells were stained for alkaline phosphatase (ALP) and for in vitro matrix mineralization.

For osteogenic differentiation of neonatal calvaria, cells were plated at 10 × 103 cells/cm2 in 4-well plates for staining and 6-well plates for ALP protein quantitation and gene expression profiling, including different time points (control, 5, 10, and 15 minutes). Differentiation was performed in α-minimum essential medium (α-MEM; Cat. No. 32561, Gibco) containing 10% FBS, 100 U/mL of penicillin, 100 µg/mL of streptomycin, 50 µg/mL of vitamin C (Sigma), and 10 mM β-glycerol-phosphate (Sigma). The medium was changed every third day and stained for ALP on day 5, 10, and 15, along with matrix mineralization on days 10 and 15.

Adipocyte differentiation

For adipogenic differentiation of MSCs, cells were plated at high densities (25 × 103 cells/cm2) in 24-well plates in CEM, Iscove modified Dulbecco medium (IMDM; Cat. No. 21980, Gibco) containing 12% FBS (Gibco), and 12 µM L-glutamine (Cat. No. 25030, Gibco). To induce adipocytes, the medium was supplemented with 5 µg/mL of insulin (Sigma), 10 nM dexamethasone (Sigma), and 50 µM indomethacin (Sigma). The medium was changed every third day, and on day 16, cells were visualized for adipocytes by oil red O staining.

Osteoclast culture and differentiation

Bone marrow cells (BMCs) derived from 8-week-old mice were plated in 96-well plates in osteoclastic medium (OCM) containing α-MEM (Gibco) supplemented with 10% FBS (Gibco), 100 U/mL of penicillin (Gibco), 100 µg/mL of streptomycin (Gibco), 25 ng/mL of recombinant human macrophage colony-stimulating factor (rhM-CSF; R&D Systems, Minneapolis, MN, USA), and 25 ng/mL of recombinant human receptor activator of NF-κB (rhRANKL; Pepro-Tech, Rocky Hill, NJ, USA). The presence of osteoclastic cells was verified by staining for tartrate-resistant acid phosphatise (TRACP) performed 5 days after culture, and TRACP+ multinucleated cells with more than 4 nuclei were scored as osteoclastic cells (OCs).

Assays for cell proliferation, senescence, and DNA damage

Cell proliferation was assessed using the BrdU Detection Kit (Cat. No. 93-3943, Invitrogen) according to the manufacturer's instructions. Short-term proliferation was carried out in calvarial OB cultures by plating 2000 cells/cm2, followed by counting the cells on days 1, 3, 6, and 9 using a Nucleocounter (Chemo-Metec, Allerød, Denmark).

Cellular senescence was assessed by using a β-galactosidase-associated staining, as described previously.21 Senescent cells appear greenish blue and were identified by light microscopy.

DNA damage was assessed by γ-H2AX staining after acid ethanol fixation and using anti-phosphohistone H2AX (ser-139) antibody (clone JBW301) according to the manufacturer´s instructions (Upstate Cell Signaling Solutions, Billerica, MA, USA).22

In vivo cell implantation

Cells (5 × 105) mixed with hydroxyl-apatite/tricalcium phosphate ceramic powder (HA/TCP, 40 mg; Zimmer Scandinavia, Albertslund, Denmark) were implanted subcutaneously in the dorsal surface of 2-month-old female NOD/SCID mice. Implants were recovered after 8 weeks, fixed in 4% formalin, and stored in PBS. Implants were dehydrated, embedded, sectioned (7.5 µm thick), and stained with hematoxylin and eosin. The amount of bone formed per implant was quantified as described previously.23

Illumina microarray bead chip analysis

RNA from long bones (tibias, n = 3) of Terc−/− and wild-type mice were subjected to microarray analysis using Illumina MouseWG-6 v2 Expression Bead Chips (Gene Logic, Gaithersburg, MD, USA), RNA was isolated as described in the Supplemental Text, and the gene arrays were performed in triplicate. Analysis of differentially expressed genes was performed by Ingenuity Pathway analysis (Ingenuity Systems, Redwood, CA, USA; www.ingenuity.com). Differentially expressed molecules were categorized in different groups according to biologic function and/or disease in Ingenuity's Knowledge Base. Data were classified into disease and disorder, molecular and cellular functions, physiologic system development and function, and canonical pathways. Genes known to play an important role in osteoblast and osteoclast commitment, proliferation, differentiation, and function were identified among genes that were differentially up- or downregulated (minimum value = 1.8 fold, p = .00001). Data were deposited in Minimum Information about Microarray Experiment Compliant in Gene Expression Omnibus (GEO Accession No. GSE21523).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Terc−/− mice exhibit an age-related bone-loss phenotype

We have employed third-generation (G3) Terc−/− mice in these studies because they exhibit the telomere-shortening-related accelerated-aging phenotype.13, 14, 24 Terc−/− mice were smaller than wild-type mice throughout their life span (Fig. 1B, a) and exhibited reduced body weight by 25% at 20 weeks of age and a progressive decline in body weight reaching 42% at 32 weeks of age (Fig. 1B, b). Terc−/− mice exhibited mild patterning defects with delayed ossification in occipital bone, calvarial sutures, sternubra, and metatarsal and caudal vertebrae when E17.5-day-old embryos were examined (Fig. 1A). Postnatal bone mass measurements employing DXA scanning throughout a 32-week observation period demonstrated the presence of an age-related bone loss in Terc−/− mice. No difference in total bone mineral content (BMC) or total bone mineral density (BMD) was observed between wild-type and Terc−/− mice up to 6 weeks of age (Fig. 1B, b). However, starting at 8 weeks, Terc−/− mice BMD and BMC were reduced by 10% and 26%, respectively compared to wild-type mice, and this difference was maintained throughout a 32-week observation period (Fig. 1B, b). Femur BMC and BMD exhibited similar changes throughout the same period, with 13% and 23% reductions, respectively, in Terc−/− mice compared with wild-type mice (Supplemental Fig. S1A). In addition, X-ray examination of 32-week-old Terc−/− mice revealed the severe thoracic kyphosis characteristics of the osteoporotic phenotype (Fig. 1B, a).

thumbnail image

Figure 1. Terc−/− mice showed a bone-loss phenotype. (A) Whole-mount staining of 17.5-day-old wild-type and Terc−/− embryos (n = 6). Arrow show mild modeling defects at different skeletal sites (eg, delay closure of calvaria sutures, delayed ossification of the sternubra, delayed ossification of the metatarsals). (B, a) X-ray examination of the skeleton (n = 3) of wild-type and Terc−/− mice demonstrating kyphosis. (B, b) DXA scanning follow-up on body weight, total BMC (grams), and BMD (g/cm2) of Terc−/− and wild-type mice starting at age 4 to 32 weeks. *p ≤ .05. (C) Bone histomorphometry of spinal region (L5). Bone volume per tissue volume (BV/TV, %), trabecular thickness (Tb.Th, µm), trabecular separation (Tb.Sp, µm), number of osteoblasts per bone surface (NOb/BPm, mm1), and osteoblast surface per bone surface (ObS/BS, %). Bone histomorphometry of tibia. Number of osteoclasts per bone surface (NOc/BPm, mm1) and osteoclast surface per bone surface (Oc.S/BS, %). (D) Dynamic bone histomorphometry (n = 5). Mineralized surface per bone surface (MS/BS, %), mineral apposition rate (MAR, µm/d), bone-formation rate per bone surface (BFR/BS, µm3/µm2/year), bone-formation rate per tissue volume (BFR/TV, %), and bone-formation rate per bone volume (BFR/BV, %/year). Data are means ± SD. *p ≤ .05.

Download figure to PowerPoint

In order to detect compartment-specific changes in bone mass, micro–computed tomographic (µCT) scanning was employed. Supplemental Fig. S1A, B demonstrates that reduced bone mass was more apparent in the trabecular bone with significantly decreased bone volume/total volume (BV/TV, 31%) in Terc−/− mice. In addition to decreased bone mass, Terc−/− mice exhibited deterioration of bone architecture with a significant decrease in trabecular thickness and cortical thickness by 18% and 28%, respectively. Furthermore, Terc−/− bone showed increased bone surface (BS)/bone volume (BV) of both cortical bone and trabecular bone by 54% and 17%, respectively (Supplemental Fig. S2A, B). In addition, other structural parameters, such as actual BMD of bone volume (aBMD; mg HA/cm3), were reduced in Terc−/− mice by 4% (Supplemental Fig. S2A).

Defective bone formation and enhanced osteoclastogenesis in Terc−/− mice

To study tissue-level cellular mechanisms of bone loss, bone histomorphometry was performed and revealed reduced BV/TV and trabecular thickness (Tb.Th) by 26% and 17%, respectively (Fig. 1C). In addition, osteoclast number and surface were increased in Terc−/− mice by 55% and 144%, respectively (Fig. 1C). Dynamic histomorphometry revealed defective osteoblastic function, as shown by reduced mineralized surface per bone surface (MS/BS) and tissue-level bone-formation rate (BFR) by 31% and 44%, respectively, in the Terc−/− mice compared with wild-type mice (Fig. 1D).

Skeletal stem cell populations exhibit intrinsic cell-autonomous defects in Terc−/− mice

Impaired bone formation and enhanced osteoclastogenesis revealed by bone histomorphometry can result from intrinsic cellular defects in bone cell populations or microenvironmental (extrinsic) changes. We studied both possibilities in a number of cell culture models.

To study the presence of intrinsic (cell-autonomous) defects, we evaluated several osteoblastic cell populations obtained from Terc−/− and wild-type mice. Ex vivo analysis of MSC (containing a population of osteoprogenitor and stem cells) frequency and differentiation capacity (ie, colony-forming unit assays) was performed. Both the number of colony-forming units fibroblasts (CFU-Fs) and the number of osteogenic ALP+ (CFU-ALP+) colonies (Fig. 2A) were reduced significantly in Terc−/− mice compared with wild-type controls. Additionally, the number of 5-bromodeoxyuridine-positive (BrdU+) proliferating cells (labeling fraction) in CFU-F and CFU-ALP+ colonies was reduced (Supplemental Fig. S3). Osteoblastic differentiation capacity of in vitro MSC cultures was shown to be significantly impaired in Terc−/− mice, as assessed by decreased ALP activity, matrix mineralization (Fig. 2B, C), and reduced expression of osteoblast-specific genes (ie, Runx2, Osx, Alpl, Col1a1, Opn, Ibsp, Sparc, and OC) compared with wild-type MSC cultures (Fig. 2D).

thumbnail image

Figure 2. Terc−/− MSCs displayed impaired osteoblast and adipocyte differentiation in vitro. (A) MSCs were cultured from bone marrow of wild-type and Terc−/− mice at low density, as described in “Materials and Methods.” Colony-forming unit fibroblasts (CFU-Fs) were counted after 5 days in culture. ALP staining activity was performed on CFU-Fs, and CFU-ALP+ colonies were represented as a percent of total number of CFU-Fs. (B) Osteoblast differentiation of bone marrow–derived MSCs from wild-type and Terc−/− mice. Cells were induced to differentiate into osteoblasts as described in “Materials and Methods.” ALP activity staining was performed after 15 days of induction. Matrix mineralization was verified by alizarin red staining after 15 days of induction. (C) ALP activity (U/g of protein) was measured on day 15 during osteoblast differentiation of MSCs. (D) Real-time PCR analysis of osteoblastic markers on day 15 during osteoblast differentiation of MSCs. Expression of each target gene was calculated as a relative expression to β-actin and represented as fold induction over control noninduced cells. (E) Adipocyte differentiation of bone marrow–derived MSCs from wild-type and Terc−/− mice. Cells were induced into adipocyte lineage for 15 days as described in “Materials and Methods.” Fat accumulations were determined by oil red O staining, and adipocyte expression markers were quantified by real-time PCR analysis. (F) Quantification of in vivo bone marrow fat content. Histomorphometric analysis was performed on tibial sections from 22-week-old female mice (n = 5). Black arrows indicate the adipocytes. Data are means ± SD. *p ≤ .05. **p ≤ .01. Bar (E, F) = 100 µm.

Download figure to PowerPoint

Impaired in vitro adipocyte differentiation in Terc−/− mice

MSCs can differentiate into osteoblasts and adipocytes,25 and in some models of bone loss, adipogenesis is enhanced.26 In Terc−/− mice, MSCs exhibited reduced adipocyte differentiation ability, as evidenced by reduced number of lipid-filled mature adipocytes as well as reduced gene expression levels of adipocytic markers (ie, Pparγ, Adipoq, and aP2; Fig. 2E). In vivo, bone marrow adipocyte tissue volume in Terc−/− mice was comparable with that of wild-type mice (Fig. 2F).

Osteoblastic cells display impaired differentiation owing to intrinsic cell autonomous defects in Terc−/− mice

As a model of committed osteoblastic cells, we studied OBs cultured from neonatal calvaria. With cytochemical staining for in vitro ALP, matrix mineralization and ALP enzymatic activity were reduced in Terc−/− cells (Fig. 3A). Similarly, expression analysis of osteoblastic markers was reduced significantly for Osx, Alpl, Col1a1, Ibsp, and OC but not Runx2 and Sparc in Terc−/− cells compared with wild-type controls (Fig. 3B). These changes suggest impairment in OB differentiation and function. In order to test the in vivo consequence of these changes and whether it was possible to rescue this phenotype using a “youthful” environment, we implanted neonatal calvarial OBs from Terc−/− and wild-type mice subcutaneously in young immune-deficient mice and quantified the amount of heterotopic bone formed. As shown in Fig. 3C, Terc−/− OBs formed a significantly reduced amount of ectopic bone compared with wild-type cells.

thumbnail image

Figure 3. Impaired in vitro and in vivo osteoblast differentiation of Terc−/− OB calvaria while exhibiting enhanced bone resorption. (A) Osteoprogenitor cells were isolated from neonatal calvaria of wild-type and Terc−/− pups as described in “Materials and Methods.” Cells were induced to osteoblast differentiation for 15 days. ALP activity was verified by cytochemical staining and quantified by biochemical assay. Matrix mineralization was verified by alizarin red staining after 15 days of induction. (B) Osteoblast gene expression analysis by real-time PCR on day 15 of induction. (C) In vivo ectopic bone formation by calvarial OBs. Cells were mixed with HA/TCP and implanted subcutaneously into NOD-SCID mice. Bar = 100 µm. (D) In vitro osteoclast differentiation of bone marrow cells. Whole bone marrow cells were isolated from wild-type and Terc−/− mice and cultured in M-CSF and RANKL to induce osteoclast differentiation. TRACP staining of multinuclear cells was performed on day 5. Red arrow shows enlarged osteoclasts in Terc−/− cells. Cells with more than five nuclei were counted as osteoclasts. Bar = 500 µm. (E) Total deoxypyridinoline cross-link (tDPD, nmol/L) measurements of serum samples obtained from wild-type and Terc−/− mice (n = 5). Data are means ± SD of three independent experiments. *p ≤ .05.

Download figure to PowerPoint

Terc−/− osteoclastic cells exhibit a normal phenotype in vitro

Since the histologic bone analysis of Terc−/− mice revealed a marked increase in OC number, we examined the OC differentiation of Terc−/− bone marrow cells in vitro. As shown in Fig. 3D, OC differentiation capacity of bone marrow cells was comparable between wild-type and Terc−/− mice, as assessed by tartrate-resistant acid phosphatase (TRACP) staining for multinucleated cells (Fig. 3D). The size of osteoclasts in Terc−/− cultures was bigger (red arrow) with increased numbers of nuclei compared with wild-type OC cultures (Fig. 3D). In contrast, and in concurrence with the histomorphometric data, serum of Terc−/− mice showed increased levels of total deoxypyridinoline (tDPD, nmol/L), a product of collagen degradation and a bone-resorption marker (Fig. 3E).

Impaired proliferation and accumulation of senescent cells in Terc−/− bone cell cultures

When tested for their proliferative capacity by BrdU incorporation and short-term proliferation assay, Terc−/− bone cells (MSCs and calvarial OBs) showed a clear defect in their proliferative ability (Fig. 4A, a, B, a). Moreover, Terc−/− MSC and calvarial OB cultures contained a larger proportion of β-gal+ cells (Fig. 4A, b, B, b) and γ-H2AX+ cells (Fig. 4A, c, B, c), which are markers of senescence and DNA damage, respectively, compared with wild-type cultures. In addition, Terc−/− bone cells exhibited a higher expression level of cell cycle inhibitors and senescence markers p21 and p16 than wild-type cells (Supplemental Fig. S4). In addition, Terc−/− calvarial (OB) cells did not reveal any enhanced apoptotic cell death at baseline or after treatment with dexamethasone (4 µM). Treatment with hydrogen peroxide (H2O2, 0.5 mM) resulted in enhanced apoptotic cell death (Supplemental Fig. S5A). Furthermore, no sign of increased apoptosis was observed during OB differentiation of Terc−/− MSCs compared with wild-type MSCs (Supplemental Fig. S5B).

thumbnail image

Figure 4. Telomerase-deficient mice (Terc−/−), OB calvaria, and MSCs exhibit cell intrinsic defects. (A) Skeletal (mesenchymal) stem cell (MSC) cell-autonomous defects. (a) MSCs were cultured from bone marrow of wild-type and Terc−/− mice and labeled with BrdU for 2 hours. Bar = 20 µm. (b) MSC senescence was identified by β-galactosidase staining. β-Gal+ cells were represented as a percent of the total number of cultured cells. Bars = 20 µm. (c) γ-H2AX immune staining for DNA damage in MSC cultures Bar = 100 µm. The graph shows the number of γ-H2AX+ nuclei as a percentage of total number of cells. (B) Osteoprogenitor (OB calvaria) intrinsic defects. (a) Short-term proliferation of osteoprogenitors and number of cells at different time points (1, 3, 6, and 9 day). (b) β-Gal staining (β-gal+) for senescent OB calvaria cells. Bars = 20 µm. (c) γ-H2AX immune staining for DNA damage in OB calvaria cultures (%). Bar = 100 µm. Data are means ± SD of three independent experiments. *p ≤ .05.

Download figure to PowerPoint

Gene expression profiling revealed a proinflammatory microenvironment in the Terc−/− mouse skeleton

To investigate the molecular mechanisms of bone loss in Terc−/− mice, RNA microarray analysis was carried out on long bones of both Terc−/− and wild-type mice (tibias, n = 3). Based on the Ingenuity pathway knowledge base, we found a significant overrepresentation of proinflammatory response genes of 21% of the total upregulated genes (≥1.8-fold; Supplemental Table S1), whereas skeletal and muscle function genes were overrepresented by 17% of the total number of downregulated genes (≤1.8-fold; Supplemental Table S1).

Among the significantly upregulated genes, we identified genes belonging to inflammatory response, response to infectious diseases, and immune response. For example, Toll-like receptor, interleukin 6, and nuclear factor κB (NF-κB) signaling were among the significantly overrepresented pathways (Supplemental Fig. S6B). A list of upregulated proinflammatory osteoclast supportive genes known to be involved in osteoclast differentiation is provided in Supplemental Table S1. Of note are genes Il1b, Fos, Junb, Socs3, Tlr2, PU.1, Ostf1, and Tnfα (Fig. 6A, Supplemental Table S1). Additionally, we have confirmed the upregulation of these genes by quantitative real-time polymerase chain reaction (PCR; Supplemental Fig. S7).

The significantly downregulated genes included the following categories: hematologic diseases, genetic diseases, and genes related to organismal injury. A number of genes known to be involved in osteoblast commitment, differentiation, and matrix mineralization were among the significantly downregulated genes (eg, Bglap, Col1a1, Col2a1, Sparc, Alpl, Phex, Ibsp, Gdf10, and Sp7; Fig. 5A, Supplemental Table S1). These results were further confirmed by real-time PCR (Supplemental Fig. S7). Moreover, Terc−/− mice showed significant downregulation of a number of oxidative stress-response and mitochondrial dysfunction genes (Supplemental Table S1).

thumbnail image

Figure 5. Microarray analysis (Ingenuity Systems; www.ingenuity.com) of Terc−/− bone showed a proinflammatory bone-resorbing microenvironment with reduced bone-formation markers. (A) Upregulated and downregulated genes in Terc−/− bone (≥1.8-fold, p = .00001) were annotated based on their biologic functions in bone. Fold changes of genes upregulated and downregulated in Terc−/− bone over wild-type controls and known to be involved in osteoclast or osteoblast differentiation. (B) Effect of mouse Terc−/− sera on the in vitro osteoclast differentiation of control wild-type bone marrow cells. Wild-type whole bone marrow cells were induced into OC differentiation in the presence of 5% serum obtained from either Terc−/− or wild-type mice. TRACP staining for MNCs was performed after 5 days. *p ≤ .05. Bar (D) = 200 µm.

Download figure to PowerPoint

Terc−/− serum promotes osteoclastogenesis of wild-type bone marrow cells

The microarray data suggest the presence of a proinflammatory microenvironment within the bone that may contribute to the enhanced bone resorption observed in Terc−/− mice. As a surrogate for bone microenvironment, the biologic effects of serum on an osteoclastic cell population were examined. Serum obtained from Terc−/− mice enhanced differentiation of wild-type bone marrow mononuclear cells into osteoclasts, as shown by increasing the number of TRACP+ multinucleated cells on day 5 (Fig. 5B).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

In this study, we demonstrated that loss of telomerase enzyme function and telomere shortening in Terc−/− mice leads to age-related bone loss that is caused by decreased bone formation and increased bone resorption. We found that decreased bone formation is caused mainly by intrinsic defects in the osteoblastic cell populations and that enhanced bone resorption is the result of the presence of a proinflammatory bone microenvironment (Fig. 6).

thumbnail image

Figure 6. A schematic model describing telomerase deficiency–induced bone loss in mice. Telomerase deficiency results in the shortening of telomeres in third-generation Terc−/− mice because mice have long telomeres per se. As a result, Terc−/− MSCs showed senescence-mediated alterations [ie, reduced proliferation by increased expression of cell cycle inhibitors (p21 and p16) and accumulation of DNA damage]. Consequently, two major shifts occurred vis-à-vis bone homeostasis. (a) DNA damage and reduced proliferation that culminate into compromised osteoblast differentiation on induction. (b) Creation of a proinflammatory microenvironment supportive of osteoclast-mediated bone resorption. Hence grouping of cell intrinsic and extrinsic (microenvironment) defects results into a bone-loss phenotype in Terc−/− mice.

Download figure to PowerPoint

To study the effects of telomerase deficiency and telomere shortening on bone biology in the postnatal period, we employed the Terc−/− mouse model.27 Mice, in contrast to humans, possess long telomeres. Terc−/− mice are deficient in the gene encoding the telomerase RNA component (Terc) and thus are telomerase-deficient.28 Terc−/− mice are normal at first generation (G1), but after successive mating of homozygous null intercrosses, the telomeres become shortened, and an accelerated-aging phenotype becomes apparent starting from the third generation (G3) onward with progressive infertility and abnormal hematology and immune responses.13, 14 Other pathologies also have been reported in late generations, including heart failure and atrophy of small intestine and skin, and impaired angiogenic responses.13 Our results demonstrate that Terc−/− mice additionally exhibit an age-related bone-loss phenotype starting at 8 weeks of age, and the low-bone-mass phenotype is already established in middle age, leading to an osteoporotic phenotype with kyphosis and deteriorated bone architecture. Interestingly, we observed that the decreased bone mass was more apparent in the trabecular bone than in the cortical bone compartment owing to higher bone turnover in Terc−/− mice. Since surface/volume ratio is higher in trabecular bone than in cortical bone, trabecular bone loss is usually observed in high-turnover states.29

Histomorphometric studies of tissue-level bone cell dynamics demonstrated the presence of impaired osteoblastic bone formation and enhanced osteoclastic bone resorption, which together with increased bone turnover resulted in age-related bone loss. Additionally, DNA microarray analysis of bones of Terc−/− mice demonstrated the downregulation of osteoblastic bone-forming genes and enhanced expression of osteoclastogenesis-promoting genes, corroborating the presence of both osteoblastic and osteoclastic defects. Thus the bone phenotype of Terc−/− mice is a good model for skeletal changes observed in a number of metabolic bone diseases in humans (eg, postmenopausal bone loss and glucocorticoid- and inflammation-induced bone loss).

Decreased in vivo bone formation in Terc−/− mice can result from several defects in OB cell populations, including impaired cell proliferation and differentiation. We found decreased numbers of CFU-Fs, which are surrogate markers for in vivo osteogenic stem cells,30 suggesting the presence of stem cell atrophy in Terc−/− mice. It is plausible that the generation of sufficient numbers of CFU-Fs needed for bone formation depends on the ability of stem cells to proliferate, and this ability becomes limiting in the context of shortened telomeres in Terc−/− mice. Impaired proliferation and differentiation of stem and progenitor cell populations have been observed in other cellular compartments of Terc−/− mice. For example, Terc−/− mice exhibit reduced proliferative potential of hematopoietic stem cells,31 decreased proliferation in testis as well as impaired sperm production,14 and decreased proliferative capacity of adult neural stem cells.32 There are several possible molecular mechanisms that link telomere shortening with OB dysfunction in Terc−/− mice. We found that cultured MSCs and OBs from Terc−/− mice accumulated a large proportion of senescent cells exhibiting DNA damage and upregulation of cell-cycle inhibitor genes. A similar phenomenon has been observed in detailed studies of liver regeneration, where senescent cells were the cells possessing critically shortened telomeres.24 It is thus plausible that accumulation of senescent cells within tissues leads to a decrease in the pool size of competent cells available for tissue regeneration. In addition, under situations of increased demand for bone regeneration, as in Terc−/− mice with high bone turnover, impaired bone formation is manifested. It also has been suggested that the senescent cells secrete a number of extracellular products that impair tissue homeostasis.6 We did not find clear evidence for increased apoptosis at baseline or during OB differentiation of Terc−/− OB-calvaria and MSCs. Thus senescence-associated intrinsic defects in the Terc−/− OB cell population are the dominant factor in causing impaired bone formation. A corollary of this finding is that ectopic implantation of Terc−/− OBs in a “youthful” environment did not rescue the bone-formation ability of the cells. Interestingly, we observed that Terc−/− cells exhibited increase susceptibility to oxidative damage/stress after hydrogen peroxide treatment (H2O2), corroborating previous findings of lower catalase activity in Terc−/− mouse embryonic fibroblasts (MEFs)33 and poor STAT5a phosphorylation in Terc−/− macrophages after H2O2 treatment.34

Terc−/− mice exhibited increased bone resorption and enhanced osteoclastogenesis in vivo. However, the OCs derived from Terc−/− mice are normal when cultured in vitro, suggesting that the enhanced osteoclastogenesis is secondary, owing to changes in bone microenvironment, and not owing to cell autonomous (intrinsic) defects. The ability of Terc−/− serum, compared with wild-type serum, to increase osteoclast formation further supports this hypothesis. Comparable with our findings, normal OC generation from bone marrow cells has been reported in Terc−/− mice with short and dysfunctional telomeres.35 OC cell generation and function are regulated by a large number of cytokines and systemic hormones.36–42 Microarray analysis of bones of Terc−/− mice provided evidence that the bone microenvironment contains high levels of several proinflammatory factors known to affect osteoclast generation and bone resorption (eg, interleukin 1β and tumor necrosis factor α). Interestingly, RANKL was upregulated (1.1-fold) and OPG was downregulated (1.4-fold), suggesting increased RANKL/OPG (2-fold). Additionally, several signaling pathways known to affect osteoclastogensis were upregulated (eg, NF-kB43 and interleukin 6 signaling).42 Similar to our findings, Ju and colleagues described the presence of microenvironmental changes in the bone marrow of Terc−/− mice that limited the engraftment of transplanted wild-type hematopoietic stem cells owing to increased production of a number of cytokines, including granulocyte colony-stimulating factor (G-CSF).44

Is the presence of a proinflammatory microenvironment a bone-specific or generalized phenomenon? Aging is accompanied by a chronic low-grade inflammation state caused by immune senescence resulting from chronic antigenic overload.45 The presence of chronic inflammation in Terc−/− mice has not been studied systemically, but some of the reported characteristics of Terc−/− mice suggest the presence of chronic inflammation. The mice exhibit an immune senescence phenotype with impaired T- and B-cell function46 and an atrophied spleen,13 and sera from Terc−/− mice contained high levels of several inflammatory molecules related to the innate immune response.47 Our data also suggest that these proinflammatory changes were more pronounced in the skeleton, which can be explained by the close proximity of the immune system and bone. It is also possible that changes in bone are caused by deficiency in specific hormones (eg, sex steroids) or other serum factors affecting bone turnover.5, 48 Identification of these factors and their contribution to the bone-loss phenotype in Terc−/− mice can lead to the development of novel strategies for reversing the bone-aging phenotype.

There is a growing body of findings that links the presence of an age-related impaired bone cell homeostasis with shortening and dysfunctional telomeres. Human OBs and MSCs exhibit an in vitro replicative senescence phenotype in association with telomere shortening.10 Overexpression of hTERT in human MSCs results in telomeres elongating and abolishing replicative senescence–associated impaired cellular function in vitro and in vivo.10 Patients with premature aging syndrome (eg, Werner syndrome or dyskeratosis congenital), having telomere dysfunction, exhibit an osteoporotic phenotype.49, 50 Finally, bone mass showed small but significant correlation with telomere length.51, 52 In conclusion, our findings implicate telomere shortening in the pathogenesis of age-related bone loss owing to multiple cellular and molecular mechanisms (Fig. 6). Approaches to transiently enhance telomerase activity and elongate the cellular telomeres through telomerase activators53 or approaches to protect the cells from the harmful effects of telomere shortening need to be identified and tested10 in order to provide novel therapeutic strategies to abolish age-related bone loss.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

We are grateful to Lone Christiansen and Bianca Jørgensen for excellent technical assistance. The study was supported by grants from the Lundbeck Foundation, the NovoNordisk Foundation, and the Danish Medical Research Council. HS has received a PhD fellowship from the NovoNordisk Foundation.

HS performed mouse and cell culture experiments and wrote the article. BMA designed the experiments and edited the article. ND performed the µCT examinations and the in vivo implantation. PC performed the histomorphomteric experiments, WQ edited the article. MA designed the experiments. MK designed the project and wrote the article.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Additional Supporting information can be found in the online version of this article.

FilenameFormatSizeDescription
JBMR_349_sm_SuppMat.doc5296KSupplemental Material

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.