TY - JOUR
AU1 - Wang,, Liping
AU2 - Roth,, Theresa
AU3 - Nakamura, Mary, C
AU4 - Nissenson, Robert, A
AB - Abstract Progranulin (PGRN) is best known as a glial protein for which deficiency leads to the most common inherited form of frontotemporal dementia. Recently, PGRN has been found to be an adipokine associated with diet-induced obesity and insulin resistance. Therefore, PGRN may have homeostatic effects on bone because PGRN is reported to promote the differentiation of bone-resorbing osteoclasts. We investigated the actions of PGRN on bone using PGRN gene (Grn) knockout (KO) mice and transgenic mice with PGRN mutation and surprisingly found that loss of PGRN prevented the bone loss in female mice induced by aging and estrogen deficiency, whereas it had no effect on male bones during aging. Strikingly, bone formation was increased in female (but not male) PGRN KO mice. We also found that loss of PGRN inhibited bone resorption and osteoclastogenesis in both male and female mice and promoted the production of osteogenic factors in osteoclast lineage cells. These results indicate that PGRN serves to uncouple bone turnover in female mice by promoting bone resorption and suppressing bone formation. Furthermore, we demonstrated that microglial cells/macrophages, but not adipocytes, are an important source of PGRN in producing negative skeletal effects in females. Targeting PGRN production by microglial cells/macrophage-lineage cells may provide a therapeutic approach for the treatment of osteoporosis in females. Adult bone is a dynamic organ. Bone cells, including osteoblasts and osteoclasts, forming bone remodeling units perceive osteotropic signals such as mechanical strains and biochemical cues and coordinately command both bone formation and resorption to maintain bone structural integrity and mineral homeostasis (1). A reduction in osteoblast function due to diseases or aging leads to a decrease in the bone volume formed by individual bone remodeling units that often causes negative remodeling, as the bone formed is less than that which is resorbed (2). This is precisely the case in postmenopausal osteoporosis, in which estrogen deficiency results in increased bone resorption accompanied by a failure of bone formation to increase correspondingly. This observation has heightened interest in identifying factors, potentially under estrogenic control, that promote bone resorption but repress bone formation because such factors would provide targets for treating osteoporosis. Progranulin (PGRN) is a cysteine-rich, secreted glycoprotein and has been shown to be a pluripotent growth factor that mediates cell-cycle progression, tumorigenesis, wound healing (3), and neurodegenerative disease such as frontotemporal dementia (FTD) (4). Increasing evidence also supports a possible role of PGRN in bone homeostasis. Firstly, serum PGRN concentrations have been found to be significantly higher in individuals with type 2 diabetes (T2D) and obese subjects (5). Animal studies have shown that PGRN deficiency ameliorates insulin resistance in PGRN knockout (KO) mice, whereas administration of recombinant PGRN induces glucose intolerance and insulin resistance in wild-type (WT) mice (6). Secondly, PGRN is significantly increased in both serum and bone marrow fluid of ovariectomized (Ovx) mice (7), a condition associated with a decreased bone mass (8). Thirdly, recent studies have indicated that PGRN is an important regulator of cartilage formation and function (9, 10) and stimulates osteoclastic differentiation and resorption (7). Finally, PGRN has recently been identified as a proinflammatory adipokine (6, 11). Adipose tissue may behave as an endocrine organ to secrete adipokines to impair skeletal integrity (12–14). Diabetes and osteoporotic fracture are major public health concerns because of the increasingly obese and aging population (15). Although bone mineral density can be increased in obese patients possibly due to the increased loading on their bones, fracture risk increases in patients who are obese or have T2D. However, the mechanism acting on bone metabolism in obesity or T2D is not clear. To this end, we studied the role of PGRN in bone homeostasis. Contemporary osteoporosis therapies fall into two main classes: antiresorptive drugs, which slow down bone resorption, and anabolic drugs, which simulate bone formation. Both aspects of the remodeling process are affected in the same direction by most current pharmaceutical therapies (16, 17). In this study, we characterized the bone phenotype of PGRN KO mice and found that loss of PGRN prevented age-related bone loss in female mice but not in male mice and resulted in both anabolic and antiresorptive effects in adult female mice. These results identify PGRN as a potential therapeutic target in the treatment of osteoporosis in obese and nonobese patients. Materials and Methods Animals All transgenic mouse studies were approved by and performed in accordance with the Institutional Animal Care and Use Committees at the San Francisco VA Medical Center and the University of California, San Francisco (UCSF). PGRN KO mice PGRN KO mice with global deletion of the progranulin gene (Grn) were generously provided by Dr. Robert V. Farese at UCSF, and their neural phenotype has been reported previously (18, 19). Briefly, PGRN KO mice were maintained in a C57BL/6 background, and the mice used for experiments were backcrossed with inbred C57BL/6 mice (The Jackson Laboratory, Bar Harbor, ME) for at least six generations. To generate littermate WT mice to serve as the controls for age-matched PGRN KO mice, we bred all PGRN KO mice by crossing hemizygous PGRN KO mice. PGRN mutant (R493X) mice The mutant R493X mice used in this study were generated in Dr. Robert V. Farese’s laboratory (UCSF). In brief, gene targeting in murine embryonic stem cells was used to generate the mice harboring a Grn nonsense mutation analogous to human R493X, a premature stop codon at arginine 493 (20). Grn mRNA levels in hemizygous R493X mice were ∼50% reduced in tissues (liver, brain, and kidney), and homozygous Grn R493X targeted mice have markedly reduced Grn mRNA (>90%) levels in the previously mentioned tissues and lack of PGRN protein in their plasma. The 6-month-old homozygous R493X and littermate WT mice used in this study were bred by crossing hemizygous R493X mice (20). Targeted deletion of PGRN To specifically delete PGRN in adipocytes, adiponectin Cre mice (The Jackson Laboratory) were bred to floxed Grn mice obtained from Gladstone Institute (San Francisco, CA). Both mouse lines were maintained in a C57 background. The bones of 3-month-old and 6-month-old conditional KO mice and their littermate control mice (floxed Grn mice) were studied. Similarly, to specifically delete PGRN in microglial cells/macrophages, Cx3cr1 Cre mice were bred with floxed Grn mice in Dr. Eric Huang’s laboratory at UCSF. Cx3cre1 Cre mice were generously provided by Dr. Clifford Lowell at UCSF. The bones of 16-month-old KO and littermate floxed Grn mice were studied. Ovariectomy To test the effects of estrogen deficiency on bone mass in female PGRN KO mice, 4.5-month-old PGRN KO and their littermate WT mice were bilaterally sham operated (Sham) or Ovx after being weight matched (body weight per group). After surgery, all animals were maintained without any treatment of 5 weeks before euthanasia. All mice were group-housed at five mice per plastic cage, maintained in a humidity- and temperature-controlled facility with a 12:12-hour light/dark cycle, and fed with a regular Teklad mouse diet (Envigo, Somerset, NJ) and water ad libitum. At the termination of the study, mice were euthanized, and the lumbar vertebral bones, femurs, and tibiae were subjected to skeletal phenotype assessment as indicated later. To study bone formation and mineralization, mice were injected with 20 mg/kg of calcein (Sigma-Aldrich, St. Louis, MO) or 15 mg/kg of demeclocycline (Sigma-Aldrich) at 7 days or 2 days before euthanasia. Micro-CT scan Bones including femurs, tibiae, and lumbar vertebrae were fixed in 10% neutral-buffered formalin for 48 hours at 4°C and then stored in 70% ethanol before being assessed using micro-CT (μCT) scan and histomorphometry. The distal femurs and the third or fourth lumbar (L3 or L4) vertebral bodies were scanned using a Scanco VivaCT-50 μCT system (Scanco Medical, Brüttisellen, Switzerland) with an X-ray energy of 55 kV, a voxel size of 10.0 μm, and an integration time of 500 ms. The cancellous region of interest within the distal femur was defined between 0.10 to 1.35 mm, proximal to the growth plate after excluding the surrounding cortical shell. The cancellous bone within the lumbar vertebral body was scanned and assessed within a region 0.1 mm proximal to the cranial growth plate and 0.1 mm distal to the caudal growth plate after excluding the surrounding cortex. The previously mentioned cancellous regions of interest were assessed using a global thresholding protocol with segmentation values of 0.8/1/270 (21). Quantitative assessment of the diaphyseal cortex, starting at the tibiofibular junction (TFJ) and extending proximally by 0.42 mm, was also conducted, and data were generated from 40 slices using a global thresholding protocol with segmentation values of 0.8/1/365 (21). Histomorphometry Following μCT analysis, the undecalcified femurs were embedded in methyl methacrylate (Sigma-Aldrich) and then sectioned with a Jung 2265 microtome (Leica Microsystems, Bannockburn, IL). Five-micrometer sections of the distal femurs were collected and processed for Von Kossa/Tetrachrome staining, as previously described, for static histomorphometry (21). Cancellous bone at the metaphysis was assessed within a region of 100 µm from the lowest point of growth plate and extending 0.85 mm down. To measure the number of osteoclasts per bone surface (BS) in cancellous bone, sections were stained for tartrate-resistant acid phosphatase (TRAP) prior to histomorphometry (14, 22, 23). To count the number of osteoblasts per BS (N.Ob/BS) and measure the erosion depth created by osteoclasts, sections were stained by toluidine blue. Dynamic histomorphometry of cancellous bone was performed on unstained, 10-μm–thick sections from the distal femur. Percent mineralizing surface at the BS, mineral apposition rate (MAR), and surface-based bone formation rate (BFR/BS) were determined (24). Before histomorphometry, mosaic-tiled images of the distal femur or the lumbar vertebral body were acquired at ×200 magnification with a Zeiss Axioplan Imager M1 microscope (Carl Zeiss MicroImaging, Thornwood, NY) fitted with a motorized stage. The tiled images were stitched and converted to a single image using Axiovision software (Carl Zeiss MicroImaging) prior to blind analysis using BIOQUANT image analysis software (BIOQUANT Image Analysis Corp., Nashville, TN). RNA extraction and real-time PCR The right tibia and both humeri were isolated and cleaned of surrounding tissues immediately after animals were euthanized. The epiphyses were then removed, and the diaphyseal bones and bone marrow contents were separated by centrifugation and kept frozen in liquid nitrogen until processing for RNA extraction. Frozen tissues were then pulverized using a Biospec Products biopulverizer (Bartlesville, OK), followed by RNA extraction using Tel-Test RNA STAT60 (Friendswood, TX) and subsequent purification using an Invitrogen Micro-to-Midi Total RNA Purification Kit (Carlsbad, CA). cDNA was then synthesized using an Applied Biosystems TaqMan Reverse Transcription Reagents (Foster City, CA) and random hexamer primers according to the manufacturer’s recommendation. Gene amplification at the mRNA level was determined with SYBR Green master mixes using a Life Technologies ViiA™ 7 real-time PCR System (Waltham, MA). Analysis was carried out using Life Technologies ViiA™ 7 software supplied with a Waltham thermocycler. The primers designed for the genes of interest in this study are listed in Table 1. We analyzed melting curves to confirm single product formation. The amplification of Gapdh was used as an internal control. The data from real-time PCR experiments were analyzed by relative change in gene expression: 2−ΔCt [Δ threshold cycle (Ct) = Ct gene of interest − Ct internal control] (25). Table 1. Primers Used for Quantitative RT-PCR Analysis Gene Description Forward Primer Reverse Primer Grn Progranulin TCA CTG TGT CTG GGA CTT CCA GCA GCA GTG GTA GCC ATC A Sp7 Osterix TTT CTC ATT AAC TCG TTG CCA TCT CTT CGG GAA AAC GGC AAA TA Bglap2 Osteocalcin CTG ACC TCA CAG ATG CCA AG GTA GCG CCG GAG TCT GTT C Tnfsf11 RANKL TTG CAC ACC TCA CCA TCA AT TCC GTT GCT TAA CGT CAT GT Dmp1 DMP1 GGA GCC AGA GAG GGT AGA GGA A CAA AGG AAC ACA AGG AGA ATG ACA Sost Sclerostin ACC GGG CGG AGA ATG G GCT GTA CTC GGA CAC ATC TTT GG Wnt1 WNT1 CGC TTC CTC ATG AAC CTT CAC TGG CGC ATC TCA GAG AAC AC Ctsk Cathepsin K CAG CTT CCC CAA GAT GTG AT AGC ACC AAC GAG AGG AGA AA Acp5 TRAP CAG CAG CCA AGG AGG ACT AC ACA TAG CCC ACA CCG TTC TC Csf1r C-fms AGC TCT CAG TAC TTC AGG GC CAA AGG CAC CGG CTC CTA GA C1qa C1qa AAA GGC AAT CCA GGC AAT ATC A TGG TTC TGG TAT GGA CTC TCC C3 C3 CCA GCT CCC CAT TAG CTC TG GCA CTT GCC TCT TTA GGA AGT C Adgre1 F4/80 CTC TGT GGT CCC ACC TTC AT GGT GGC CAA GGA TCT GAA AA Osm Oncostatin M AGG CAC GGG CCA GAG TAC CA GGC GGA TAT AGG GCT CCA AGA GTG Gapdh GAPDH TGC ACC ACC AAC TGC TTA G GGA TGC AGG GAT GAT GTT C Gene Description Forward Primer Reverse Primer Grn Progranulin TCA CTG TGT CTG GGA CTT CCA GCA GCA GTG GTA GCC ATC A Sp7 Osterix TTT CTC ATT AAC TCG TTG CCA TCT CTT CGG GAA AAC GGC AAA TA Bglap2 Osteocalcin CTG ACC TCA CAG ATG CCA AG GTA GCG CCG GAG TCT GTT C Tnfsf11 RANKL TTG CAC ACC TCA CCA TCA AT TCC GTT GCT TAA CGT CAT GT Dmp1 DMP1 GGA GCC AGA GAG GGT AGA GGA A CAA AGG AAC ACA AGG AGA ATG ACA Sost Sclerostin ACC GGG CGG AGA ATG G GCT GTA CTC GGA CAC ATC TTT GG Wnt1 WNT1 CGC TTC CTC ATG AAC CTT CAC TGG CGC ATC TCA GAG AAC AC Ctsk Cathepsin K CAG CTT CCC CAA GAT GTG AT AGC ACC AAC GAG AGG AGA AA Acp5 TRAP CAG CAG CCA AGG AGG ACT AC ACA TAG CCC ACA CCG TTC TC Csf1r C-fms AGC TCT CAG TAC TTC AGG GC CAA AGG CAC CGG CTC CTA GA C1qa C1qa AAA GGC AAT CCA GGC AAT ATC A TGG TTC TGG TAT GGA CTC TCC C3 C3 CCA GCT CCC CAT TAG CTC TG GCA CTT GCC TCT TTA GGA AGT C Adgre1 F4/80 CTC TGT GGT CCC ACC TTC AT GGT GGC CAA GGA TCT GAA AA Osm Oncostatin M AGG CAC GGG CCA GAG TAC CA GGC GGA TAT AGG GCT CCA AGA GTG Gapdh GAPDH TGC ACC ACC AAC TGC TTA G GGA TGC AGG GAT GAT GTT C View Large Table 1. Primers Used for Quantitative RT-PCR Analysis Gene Description Forward Primer Reverse Primer Grn Progranulin TCA CTG TGT CTG GGA CTT CCA GCA GCA GTG GTA GCC ATC A Sp7 Osterix TTT CTC ATT AAC TCG TTG CCA TCT CTT CGG GAA AAC GGC AAA TA Bglap2 Osteocalcin CTG ACC TCA CAG ATG CCA AG GTA GCG CCG GAG TCT GTT C Tnfsf11 RANKL TTG CAC ACC TCA CCA TCA AT TCC GTT GCT TAA CGT CAT GT Dmp1 DMP1 GGA GCC AGA GAG GGT AGA GGA A CAA AGG AAC ACA AGG AGA ATG ACA Sost Sclerostin ACC GGG CGG AGA ATG G GCT GTA CTC GGA CAC ATC TTT GG Wnt1 WNT1 CGC TTC CTC ATG AAC CTT CAC TGG CGC ATC TCA GAG AAC AC Ctsk Cathepsin K CAG CTT CCC CAA GAT GTG AT AGC ACC AAC GAG AGG AGA AA Acp5 TRAP CAG CAG CCA AGG AGG ACT AC ACA TAG CCC ACA CCG TTC TC Csf1r C-fms AGC TCT CAG TAC TTC AGG GC CAA AGG CAC CGG CTC CTA GA C1qa C1qa AAA GGC AAT CCA GGC AAT ATC A TGG TTC TGG TAT GGA CTC TCC C3 C3 CCA GCT CCC CAT TAG CTC TG GCA CTT GCC TCT TTA GGA AGT C Adgre1 F4/80 CTC TGT GGT CCC ACC TTC AT GGT GGC CAA GGA TCT GAA AA Osm Oncostatin M AGG CAC GGG CCA GAG TAC CA GGC GGA TAT AGG GCT CCA AGA GTG Gapdh GAPDH TGC ACC ACC AAC TGC TTA G GGA TGC AGG GAT GAT GTT C Gene Description Forward Primer Reverse Primer Grn Progranulin TCA CTG TGT CTG GGA CTT CCA GCA GCA GTG GTA GCC ATC A Sp7 Osterix TTT CTC ATT AAC TCG TTG CCA TCT CTT CGG GAA AAC GGC AAA TA Bglap2 Osteocalcin CTG ACC TCA CAG ATG CCA AG GTA GCG CCG GAG TCT GTT C Tnfsf11 RANKL TTG CAC ACC TCA CCA TCA AT TCC GTT GCT TAA CGT CAT GT Dmp1 DMP1 GGA GCC AGA GAG GGT AGA GGA A CAA AGG AAC ACA AGG AGA ATG ACA Sost Sclerostin ACC GGG CGG AGA ATG G GCT GTA CTC GGA CAC ATC TTT GG Wnt1 WNT1 CGC TTC CTC ATG AAC CTT CAC TGG CGC ATC TCA GAG AAC AC Ctsk Cathepsin K CAG CTT CCC CAA GAT GTG AT AGC ACC AAC GAG AGG AGA AA Acp5 TRAP CAG CAG CCA AGG AGG ACT AC ACA TAG CCC ACA CCG TTC TC Csf1r C-fms AGC TCT CAG TAC TTC AGG GC CAA AGG CAC CGG CTC CTA GA C1qa C1qa AAA GGC AAT CCA GGC AAT ATC A TGG TTC TGG TAT GGA CTC TCC C3 C3 CCA GCT CCC CAT TAG CTC TG GCA CTT GCC TCT TTA GGA AGT C Adgre1 F4/80 CTC TGT GGT CCC ACC TTC AT GGT GGC CAA GGA TCT GAA AA Osm Oncostatin M AGG CAC GGG CCA GAG TAC CA GGC GGA TAT AGG GCT CCA AGA GTG Gapdh GAPDH TGC ACC ACC AAC TGC TTA G GGA TGC AGG GAT GAT GTT C View Large Serum chemistry At the termination of experiments, mice under deep isoflurane anesthesia were bled from the abdominal inferior vein. The blood samples were then processed in MicroTainer serum separator tubes (BD Biosciences, San Jose, CA), and plasma was collected and frozen prior to biochemistry analysis. Serum procollagen type I amino-terminal propeptide (PINP) and pyridinoline (PYD) measurements were carried out using the rat/mouse PINP EIA Kit AC-33F1 from Immunodiagnostic Systems (Scottsdale, AZ) and the Metra Biosystems Serum PYD Kit 8019 (Santa Clara, CA), respectively. Serum PGRN was measured using an Adipogen progranulin (mouse) ELISA kit (RRID: AB_2783011; AG-45A-0019Y) according to the manufacturers’ directions (26). Immunohistochemistry After deparaffinization and rehydration, 5-µm–thick sections were washed with 0.3% Triton X-100 (Sigma-Aldrich) in PBS. Endogenous peroxidase activity and nonspecific binding sites were blocked by incubating the sections in 0.3% H2O2 in PBS and 5% donkey serum in PBS, respectively. Sections were then incubated with a diluted goat anti-human PGRN antibody (RRID: AB_2114489; R&D Systems, Minneapolis, MN) (5 μg/mL) at 4°C overnight (27). After incubation with biotinylated rabbit anti-goat secondary antibody (RRID: AB_2336814; Vector Laboratories, Burlingame, CA) (1:200) (28), sections were then incubated and stained with horseradish peroxidase substrate and diaminobenzidine using a Vector Laboratories Avidin-Biotin Complex kit. The immunostained tissue sections were counterstained with hematoxylin. Bone marrow stromal cell culture and mineralization in vitro Briefly, the bone marrow plugs from 10-week-old PGRN KO and WT mice were flushed out from both femurs and tibiae. Bone marrow stromal cells (BMSCs) were cultured in a primary culture medium (PCM) containing alpha modification of Eagle medium (Thermo Scientific, Waltham, CA) supplemented with 10% fetal bovine serum (HyClone, Pittsburgh, PA), 100 U/mL penicillin (Invitrogen), 100 µg/mL streptomycin (Invitrogen), and 0.25 µg/mL Fungizone (Invitrogen) (29). BMSCs from both male and female mice were plated into six-well plates at a density of 1.8–2.3 × 106 cells per well, separately. Cells were then incubated in a humidified atmosphere of 5% CO2 at 37°C. At day 7 (d7), osteogenic differentiation of the BMSCs were induced by switching to an osteogenic medium, the PCM supplemented with 50 µg/mL ascorbic acid (Sigma-Aldrich) and 3 mM β-glycerol phosphate (Sigma-Aldrich) (29). Starting at d5, media was changed every 2 to 3 days. At d21, alkaline phosphatase activity was measured using a Leukocyte Alkaline Phosphatase kit (#85L2-1KT; Sigma-Aldrich). To assess mineralization, 2% silver nitrate solution (Sigma-Aldrich) was added to cell culture dishes for Von Kossa staining and ultraviolet cross-linked for 10 minutes. The stained culture dishes were scanned and quantified using a BIOQUANT image analysis system. Osteoclastogenesis in vitro and TRAP staining Briefly, the bone marrow from 10-week-old PGRN KO and littermate control mice was collected, and bone marrow cells were cultured in PCM supplemented with 20 ng/mL macrophage-colony stimulating factor (M-CSF; R&D Systems). After M-CSF stimulation for 24 hours, the nonadherent osteoclast precursors were collected and recultured in 48-well plates at a density of 5 × 105 cells per well in PCM supplemented with 10 ng/mL M-CSF and 50 ng/mL RANKL (R&D Systems). The media was changed every 3 days. At the end of the culture period (d9), TRAP staining was performed using a Leukocyte Acid Phosphatase kit (Sigma-Aldrich) to identify osteoclasts as TRAP+ cells containing three or more nuclei. Statistical analysis Because our data were normally distributed and had equivalent variance, statistical significance was ascertained by a Student t test or, where indicated, by two-way analysis of variance. To observe age-related effects, statistical comparisons were made vs measurement at baseline. To observe the effect of deletion of the PGRN gene, statistical comparisons were made vs sex- and age-matched control mice. P ≤ 0.05 was considered significant. Results Loss of PGRN prevents age-related bone loss in female mice The general phenotype of PGRN KO mice has been reported previously (18, 30). In brief, PGRN KO pups were born at the expected Mendelian frequency from the hemizygous parents. Young adult PGRN KO mice were healthy and fertile. Whole-body pathological evaluation was normal at the gross level, and there were no abnormalities in hematogram, serum chemistries, or body weight (18, 30). As expected, there was a sex difference in body weight of the WT mice at 6 months of age (31). However, there were no any significant differences between the littermate WT and KO mice in both sexes (male, WT vs KO: 30.2 ± 3.36 g vs 29.4 ± 1.96 g; female, WT vs KO: 23.4 ± 1.07 g vs 22.5 ± 1.58 g). To study the effects of PGRN on bone homeostasis, we first examined long bones of 2-, 6-, and 12-month-old PGRN KO mice. Peak bone mass of C57BL/6 mouse is established at ∼2 months of age (31). We found that aging was associated with significant bone loss in both male WT and KO mice, and by 12 months, there had been a 45.7% (P < 0.05) and 46% (P < 0.01) decrease in femoral cancellous bone fractional volume (BV)/tissue volume (TV) in WT and KO mice, respectively. No difference in BV/TV was detected between male KO and WT mice at these ages (Fig. 1A). From 2 to 12 months of age, BV/TV decreased by 33% (P < 0.01) in female WT mice. In contrast, no decrease in BV/TV was detected in female KO mice with aging. Compared with age-matched controls, PGRN deletion increased cancellous BV/TV by 101% (P < 0.01) at 6 months and 371% (P < 0.001) at 12 months, respectively, in female KO mice. The increase in BV/TV was associated with a substantial increase in Tb.N and a considerable decrease in Tb.Sp (Fig. 1B). Our results support the idea that loss of PGRN prevents age-related bone loss in female mice. Figure 1. View largeDownload slide μCT assessment of cancellous bone at distal femur in PGRN KO and age-matched WT mice (male: WT, n = 4; KO, n = 4 to 6; female: WT, n = 4 to 6; KO, n = 5 to 7 per age group). (A) Representative three-dimensional reconstruction of cancellous bone at distal femur (scale bar, 100 µm) and μCT assessment of cancellous BV/TV, trabecular thickness (Tb.Th), trabecular number (Tb.N), and trabecular separation (Tb.Sp) at distal femur in 2-, 6-, and 12-mo-old male mice. (B) μCT assessment of cancellous bone at distal femur in female mice. Expression of PGRN gene (Grn) was determined in (C) bone tissue and (D) osteoblasts cultured from BMSCs. Data are presented as mean ± SD. Two-way analysis of variance was used. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs littermate controls; #P < 0.05, ##P < 0.01, ###P < 0.001, ####P < 0.0001 vs baseline measurement at 2 mo. Figure 1. View largeDownload slide μCT assessment of cancellous bone at distal femur in PGRN KO and age-matched WT mice (male: WT, n = 4; KO, n = 4 to 6; female: WT, n = 4 to 6; KO, n = 5 to 7 per age group). (A) Representative three-dimensional reconstruction of cancellous bone at distal femur (scale bar, 100 µm) and μCT assessment of cancellous BV/TV, trabecular thickness (Tb.Th), trabecular number (Tb.N), and trabecular separation (Tb.Sp) at distal femur in 2-, 6-, and 12-mo-old male mice. (B) μCT assessment of cancellous bone at distal femur in female mice. Expression of PGRN gene (Grn) was determined in (C) bone tissue and (D) osteoblasts cultured from BMSCs. Data are presented as mean ± SD. Two-way analysis of variance was used. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs littermate controls; #P < 0.05, ##P < 0.01, ###P < 0.001, ####P < 0.0001 vs baseline measurement at 2 mo. We found that the PGRN gene, Grn, is expressed in mouse bone tissue and in osteoblasts cultured from BMSCs. PGRN gene KO resulted in a marked decrease in Grn mRNA levels in bone and cultured osteoblasts from PGRN KO mice (Fig. 1C and 1D). PGRN selectively inhibits bone formation in female skeleton To study the cellular mechanism by which the loss of PGRN increases BFR in female mice, we performed histomorphometry in a cohort of 6-month-old KO and littermate WT mice (Fig. 2). Histomorphometry of the distal femur demonstrated that PGRN gene deletion had no effect on BV/TV in males (Fig. 2A), but dramatically increased cancellous BV/TV in females (Fig. 2A), confirming the μCT results (Fig. 1). Loss of PGRN significantly increased N.Ob/BS, MAR, and BFR in female mice (Fig. 2B–2D). In contrast, PGRN deficiency did not alter N.Ob/BS or BFR in male mice (Fig. 2B–2D). These data demonstrate that PGRN suppresses osteoblast number and activity in a female-specific fashion. Figure 2. View largeDownload slide Histomorphometry of cancellous bone at distal femur in 6-mo-old PGRN KO and littermate WT mice (male: WT, n = 4, KO, n = 12; female: WT, n = 4; KO, n = 5). (A) Static histomorphometry assessment of cancellous BV/TV, trabecular width (Tb.Wi), trabecular number (Tb.N), and trabecular separation (Tb.Sp). (B) Determination of the N.Ob/BS in PGRN KO mice. (C) Representative images of fluorescence-labeled trabecular BS at distal femur in KO and WT mice. (D) Dynamic histomorphometry assessment of trabecular mineralizing surface (MS/BS), MAR, and BFR/BS at distal femur. Data are presented as mean ± SD. Two-way analysis of variance was used. (E) Mineralization of BMSCs from female PGRN KO and littermate WT mice. At day 21 postculture, cell cultures were stained for determining alkaline phosphatase activity (alkaline phosphatase staining, purple color) and mineralization (Von Kossa staining, black color). The stained colonies were normalized by total culture area (n = 6 for each). Two-tailed Student t test was used. White bar, littermate WT mice; black bar, PGRN KO mice. *P < 0.05, **P < 0.01, ***P < 0.001 vs littermate control mice; #P < 0.05, ###P < 0.001 vs male mice. Figure 2. View largeDownload slide Histomorphometry of cancellous bone at distal femur in 6-mo-old PGRN KO and littermate WT mice (male: WT, n = 4, KO, n = 12; female: WT, n = 4; KO, n = 5). (A) Static histomorphometry assessment of cancellous BV/TV, trabecular width (Tb.Wi), trabecular number (Tb.N), and trabecular separation (Tb.Sp). (B) Determination of the N.Ob/BS in PGRN KO mice. (C) Representative images of fluorescence-labeled trabecular BS at distal femur in KO and WT mice. (D) Dynamic histomorphometry assessment of trabecular mineralizing surface (MS/BS), MAR, and BFR/BS at distal femur. Data are presented as mean ± SD. Two-way analysis of variance was used. (E) Mineralization of BMSCs from female PGRN KO and littermate WT mice. At day 21 postculture, cell cultures were stained for determining alkaline phosphatase activity (alkaline phosphatase staining, purple color) and mineralization (Von Kossa staining, black color). The stained colonies were normalized by total culture area (n = 6 for each). Two-tailed Student t test was used. White bar, littermate WT mice; black bar, PGRN KO mice. *P < 0.05, **P < 0.01, ***P < 0.001 vs littermate control mice; #P < 0.05, ###P < 0.001 vs male mice. We assessed the differentiation of BMSCs into osteoblasts, but did not detect a difference in mineralization between PGRN KO and WT mice (Fig. 2E). We also found that treatment of WT BMSCs with PGRN had no effect on differentiation (data not shown). Mutation of PGRN gene (R493X) replicates the bone phenotype of PGRN KO mice Loss-of-function mutations in the PGRN gene in humans are associated with FTD (32). A mouse line with the homolog of one of these mutations (R493X) was created, and mice homozygous for this mutation were shown to display the behavioral defects seen in PGRN KO mice (20). We studied the effects of the mutation on skeletal homeostasis in 6-month-old homozygous R493X mice and their WT littermates. Mutant female mice displayed markedly increased cancellous BV/TV, associated with a considerable increase in Tb.N and a substantial decrease in Tb.Sp at distal femur in female mice; mutant male mice did not differ from WT males (Fig. 3). Strikingly, the results at the distal femur are the same as that seen in PGRN KO mice at the same age (Figs. 1 and 2). Cortical bone thickness, assessed by μCT at the TFJ, was significantly increased in both male and female R493X mice (Fig. 3B). Figure 3. View largeDownload slide μCT assessment of cancellous bone at distal femur and cortical bone at the TFJ in 6-mo-old PGRN mutant mice (R493X) and littermate WT mice (male: WT, n = 5; R493X, n = 5; female: WT, n = 4; R493X, n = 5). (A) Representative three-dimensional reconstruction of cancellous bone at distal femur (scale bar, 100 µm). (B) μCT assessment of cancellous BV/TV, trabecular thickness (Tb.Th), trabecular number (Tb.N), and trabecular separation (Tb.Sp) at distal femur, and cortical thickness (Ct.Th) at TFJ. Data are presented as mean ± SD. Two-way analysis of variance was used. *P < 0.05, **P < 0.01 vs littermate controls. Figure 3. View largeDownload slide μCT assessment of cancellous bone at distal femur and cortical bone at the TFJ in 6-mo-old PGRN mutant mice (R493X) and littermate WT mice (male: WT, n = 5; R493X, n = 5; female: WT, n = 4; R493X, n = 5). (A) Representative three-dimensional reconstruction of cancellous bone at distal femur (scale bar, 100 µm). (B) μCT assessment of cancellous BV/TV, trabecular thickness (Tb.Th), trabecular number (Tb.N), and trabecular separation (Tb.Sp) at distal femur, and cortical thickness (Ct.Th) at TFJ. Data are presented as mean ± SD. Two-way analysis of variance was used. *P < 0.05, **P < 0.01 vs littermate controls. PGRN promotes bone resorption without a sex selection We studied the effects of PGRN on bone resorption and osteoclastogenesis in both sexes. We counted TRAP+ osteoclasts at trabecular BSs (TRAP+ Oc/BS) in the distal femur. Interestingly, we observed an increase in TRAP+ cells on the BS in both male and female KO mice compared with age- and sex-matched WT mice (Fig. 4A). We further assessed osteoclastic bone resorption by measuring the depth of erosion pits and found that the loss of PGRN led to a notable decrease in erosion depth in both male and female mice (Fig. 4A). In addition, PGRN deletion significantly decreased bone marrow TNF-α mRNA levels in KO mice (Fig. 4B), in line with the previous finding that loss of PGRN promotes insulin sensitivity by lowering bone marrow proinflammatory cytokines (6). Figure 4. View largeDownload slide Assessment of bone resorption in 6-mo-old PGRN KO mice. (A) Numbers of TRAP+ Oc/BS and erosion depth at trabecular surfaces (WT, n = 4; KO, n = 5 for each sex). (B) Gene expression of TNF-α in bone marrow of 6-mo-old PGRN KO and littermate WT mice (male: WT, n = 4, KO, n = 12; female: WT, n = 4; KO, n = 5). Data are presented as mean ± SD. (C and D) Two-way analysis of variance was used. *P < 0.05, **P < 0.01 vs littermate controls. Effect of PGRN on osteoclastogenesis in vitro was studied in osteoclasts cultured from PGRN KO and WT mouse BMMs. (C) Representative images of the multinucleated TRAP+ cells. (D) Determination of TRAP+ cell numbers in the osteoclast cultures. (E) Gene expression of osteoclasts at days 7 to 9 postculture. Data are presented as mean ± SD (WT, n = 3; KO, n = 3, for each sex). Two-way analysis of variance was used. *P < 0.05, **P < 0.01, ***P < 0.001 vs sex and littermate WT mice. Figure 4. View largeDownload slide Assessment of bone resorption in 6-mo-old PGRN KO mice. (A) Numbers of TRAP+ Oc/BS and erosion depth at trabecular surfaces (WT, n = 4; KO, n = 5 for each sex). (B) Gene expression of TNF-α in bone marrow of 6-mo-old PGRN KO and littermate WT mice (male: WT, n = 4, KO, n = 12; female: WT, n = 4; KO, n = 5). Data are presented as mean ± SD. (C and D) Two-way analysis of variance was used. *P < 0.05, **P < 0.01 vs littermate controls. Effect of PGRN on osteoclastogenesis in vitro was studied in osteoclasts cultured from PGRN KO and WT mouse BMMs. (C) Representative images of the multinucleated TRAP+ cells. (D) Determination of TRAP+ cell numbers in the osteoclast cultures. (E) Gene expression of osteoclasts at days 7 to 9 postculture. Data are presented as mean ± SD (WT, n = 3; KO, n = 3, for each sex). Two-way analysis of variance was used. *P < 0.05, **P < 0.01, ***P < 0.001 vs sex and littermate WT mice. PGRN stimulates osteoclastic differentiation of both mouse and human bone marrow–derived macrophages (BMMs) by inducing cell fusion (7). To further study osteoclastogenesis in vitro, BMMs from KO mice were cultured in the presence of RANKL (7). In line with findings by Oh et al. (7), loss of endogenous PGRN dramatically decreased osteoclast formation as determined by counting multinucleated TRAP+ cells. However, no sex differences were found in this antiosteoclastogenic effect (Fig. 4C and 4D). Loss of PGRN significantly decreased the expression of Grn and of markers of mature osteoclasts (Ctsk and Acp5), while increasing the expression of markers of early osteoclasts (Csf1r and Adgre1) (Fig. 4E). Oncostatin M (Osm), a product of F4/80+ cells, has osteogenic effects (33). A very recent study has demonstrated that loss of PGRN increases the production of complement components C1a and C3 in brain microglial cells, a resident macrophage in the brain (19). Because osteoclast-derived C3a is able to stimulate osteogenesis, and C1q activates Wnt signaling pathways (34), we examined the impact of the loss of PGRN on these potential osteogenic factors in osteoclast cultures. Interestingly, loss of PGRN significantly increased the expression of C1qa, C3, and Osm in mouse osteoclast cultures (Fig. 4). Role of PGRN in estrogen deficiency–induced bone loss The sex dimorphic bone phenotype in PGRN KO mice suggested a possible role of PGRN mediating estrogen deficiency–induced bone loss. We therefore ovariectomized PGRN KO and the littermate WT mice. μCT demonstrated that the Ovx WT mice displayed a significant decrease in cancellous BV/TV (percentage) in the third lumbar vertebra (L3), whereas in the Ovx KO mice, estrogen depletion did not significantly alter the L3 BV/TV (Fig. 5B). These results demonstrate that PGRN is required for estrogen deficiency–induced vertebral bone loss. Figure 5. View largeDownload slide Role of PGRN in estrogen deficiency induced bone loss. (A) Representative images of three-dimensional μCT reconstruction of the third lumbar (L3) vertebral body in Ovx and Sham-operated PGRN KO and littermate WT mice. (B) μCT assessment of cancellous bone at the L3 vertebral body 5 wk after ovariectomy or Sham operation. (C) Serum PINP and PYD levels were determined by ELISA. (D) Skeletal gene expression was determined by real-time PCR. Data are presented as mean ± SD. WT mice: Sham, n = 4; Ovx, n = 8; PGRN KO mice: Sham, n = 4; Ovx, n = 5. Two-way analysis of variance was used. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs Sham or littermate controls; ##P < 0.01, ###P < 0.001, ####P < 0.0001 vs littermate WT mice with the same treatment. Tb.N, trabecular number; Tb.Sp, trabecular separation; Tb.Th, trabecular thickness. Figure 5. View largeDownload slide Role of PGRN in estrogen deficiency induced bone loss. (A) Representative images of three-dimensional μCT reconstruction of the third lumbar (L3) vertebral body in Ovx and Sham-operated PGRN KO and littermate WT mice. (B) μCT assessment of cancellous bone at the L3 vertebral body 5 wk after ovariectomy or Sham operation. (C) Serum PINP and PYD levels were determined by ELISA. (D) Skeletal gene expression was determined by real-time PCR. Data are presented as mean ± SD. WT mice: Sham, n = 4; Ovx, n = 8; PGRN KO mice: Sham, n = 4; Ovx, n = 5. Two-way analysis of variance was used. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs Sham or littermate controls; ##P < 0.01, ###P < 0.001, ####P < 0.0001 vs littermate WT mice with the same treatment. Tb.N, trabecular number; Tb.Sp, trabecular separation; Tb.Th, trabecular thickness. Ovx did not affect the levels of PINP in either WT or KO mice. PGRN deficiency resulted in a decrease in serum PYD by 39.4% in Sham KO mice and by 48.2% in Ovx KO mice (Fig. 5C). RNA was extracted from the tibial diaphysis, and real-time PCR demonstrated that Sham KO mice had a higher expression of Sp7 (2.2-fold), Bglap2 (5.2-fold), Tnfsf11 (10.5-fold), Dmp1 (7.0-fold), Sost (8.0-fold), and Wnt1 (10.3-fold) than Sham WT mice. However, these effects of PGRN deficiency on skeletal gene expression were abolished following Ovx (Fig. 5D). Adipocyte-derived PGRN does not affect bone mass PGRN has been identified as an adipokine implicated in the pathophysiology of T2D. To study the role of adipocyte-derived PGRN in bone homeostasis, we specifically deleted PGRN in adipocytes by crossing the floxed Grn mice with male adiponectin Cre mice. To confirm successful deletion of PGRN in adipocytes, we performed real-time PCR and immunohistochemistry on brown fat tissue (interscapular fat pads) and white fat tissue (gonadal fat pads) and found substantial reduction in PGRN at both the mRNA and protein levels (Fig. 6). However, the specific deletion of PGRN did not significantly alter the serum PGRN level as determined by ELISA (Fig. 6C). Surprisingly, deletion of PGRN in adipocytes did not affect bone mass in either male or female mice at 6 months of age (Fig. 6D). Thus, adipocyte-derived PGRN does not contribute to circulating PGRN nor to the skeletal effects of PGRN deficiency. Figure 6. View largeDownload slide Characterization of phenotype of 6-mo-old transgenic mice with a specific deletion of PGRN in adipocytes. (A) Determination of PGRN mRNA levels in interscapular and gonadal fat pads by real time PCR (WT, n = 5 to 8; KO, n = 5 to 8). (B) Determination of PGRN protein expression in interscapular fat pads using immunohistochemistry. (C) Serum PGRN was determined by using ELISA (WT, n = 5 to 8; KO, n = 5 to 8). (D) μCT assessment of cancellous bone at distal femurs in 6-mo-old male and female mice with conditional KO of PGRN in adipocytes (WT, n = 5 to 8; KO, n = 5 to 8). Two-way analysis of variance was used. *P < 0.05, **P < 0.01, ***P < 0.001 vs littermate controls. Tb.N, trabecular number; Tb.Sp, trabecular separation; Tb.Th, trabecular thickness. Figure 6. View largeDownload slide Characterization of phenotype of 6-mo-old transgenic mice with a specific deletion of PGRN in adipocytes. (A) Determination of PGRN mRNA levels in interscapular and gonadal fat pads by real time PCR (WT, n = 5 to 8; KO, n = 5 to 8). (B) Determination of PGRN protein expression in interscapular fat pads using immunohistochemistry. (C) Serum PGRN was determined by using ELISA (WT, n = 5 to 8; KO, n = 5 to 8). (D) μCT assessment of cancellous bone at distal femurs in 6-mo-old male and female mice with conditional KO of PGRN in adipocytes (WT, n = 5 to 8; KO, n = 5 to 8). Two-way analysis of variance was used. *P < 0.05, **P < 0.01, ***P < 0.001 vs littermate controls. Tb.N, trabecular number; Tb.Sp, trabecular separation; Tb.Th, trabecular thickness. Specific deletion of PGRN in microglial cells/macrophages leads to a high bone mass phenotype in female mice Deletion of PGRN from microglial cells/macrophages produces a considerable behavioral phenotype (18, 19). To determine whether microglial cell/macrophage-derived PGRN influences skeletal homeostasis, we studied the bones of female mice in which expression of the PGRN gene was deleted using Cx3cr1-Cre. μCT assessment of the L4 vertebra showed that deletion of PGRN in Cx3cr1+ cells led to an 86% (P < 0.01) increase in cancellous BV/TV at age of 16 months (Fig. 7A and 7B). The increased BV/TV was associated with a marked increase in trabecular number and a substantial decrease in trabecular separation (Fig. 7B). Moreover, specifically deleting PGRN in microglials/macrophages increased both osteoblasts (Fig. 7C) and TRAP+ cells (Fig. 7D) at trabecular BSs, which is similar to that observed in the PGRN null mice (Figs. 2B and 4A). To confirm a valid deletion of Grn in osteoclasts by using a Cx3cr1 promoter, we cultured osteoclasts using the bone marrow cells of the conditional KO mice. Indeed, a conditional deletion of Grn led to a marked decrease (∼90%) in Grn mRNA level in osteoclasts (Fig. 7E). Mature osteoblasts were also cultured using the BMSCs of the same mice (Fig. 7E). It was noted that Grn gene expression is lower in the cells cultured from the conditional KO mice (Fig. 7E). However, the difference is small, which was most likely due to the expansion of coisolated tissue macrophages or monocytes in the osteoblast cultures (35). Figure 7. View largeDownload slide Assessment of bone phenotype of transgenic mice with targeted deletion of PGRN in microglials/microphages. (A) Representative three-dimensional reconstruction of cancellous bone at the L4 vertebral bone (scale bar, 100 µm) in 16-mo-old female mice with PGRN deletion in microglial cells/macrophages. (B) μCT assessment of cancellous bone at the L4 vertebrae. (C and D) Determination of the N.Ob/BS and TRAP+ Oc/BS at the L4 vertebral body. Control, n = 5; KO, n = 5. (E) Osteoclasts (OCs) and osteoblasts (OBs) were cultured using the bone marrow cells derived from the mice with the conditional deletion of PGRN in microglials/macrophages. Grn expression was determined by real-time PCR. Control = 6; KO = 4. Data are presented as mean ± SD. Student t test was used. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs littermate controls. Tb.N, trabecular number; Tb.Sp, trabecular separation; Tb.Th, trabecular thickness. Figure 7. View largeDownload slide Assessment of bone phenotype of transgenic mice with targeted deletion of PGRN in microglials/microphages. (A) Representative three-dimensional reconstruction of cancellous bone at the L4 vertebral bone (scale bar, 100 µm) in 16-mo-old female mice with PGRN deletion in microglial cells/macrophages. (B) μCT assessment of cancellous bone at the L4 vertebrae. (C and D) Determination of the N.Ob/BS and TRAP+ Oc/BS at the L4 vertebral body. Control, n = 5; KO, n = 5. (E) Osteoclasts (OCs) and osteoblasts (OBs) were cultured using the bone marrow cells derived from the mice with the conditional deletion of PGRN in microglials/macrophages. Grn expression was determined by real-time PCR. Control = 6; KO = 4. Data are presented as mean ± SD. Student t test was used. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs littermate controls. Tb.N, trabecular number; Tb.Sp, trabecular separation; Tb.Th, trabecular thickness. Discussion The age-related deterioration of bone architecture differs between male and female mice, with bone loss being more pronounced in females (31, 36). Glatt et al. (31) have characterized The Jackson Laboratory in-bred C57BL/6 mice during aging. They found that cancellous BV/TV at the distal femur peaked between 1 and 2 months of age and declined thereafter (31). At 6 months of age, cancellous BV/TV in the distal femur in female mice had diminished to 5%, as determined by μCT (31). Recently, Noguchi et al. (36) reported that 1-year-old PGRN KO mice had lower distal femoral BV/TV (5%) than age-matched WT C57BL/6 mice. In this study, they used nonlittermate WT mice as controls, and these mice exhibited an extremely high femoral cancellous BV/TV (17%) (36). In contrast, we demonstrated that female PGRN KO mice, including the R493X mutant mice, had higher cancellous BV/TV than age-matched littermate control mice. Our results indicate that PGRN has a direct inhibitory effect on bone formation and that PGRN deficiency prevents age-related bone loss in female mice. In this study, we found that the bone phenotype of adult PGRN KO mice is sexually dimorphic. Loss of PGRN selectively prevented age-related bone loss in female but not in male mice. However, we did not detect a difference in the mineralization of BMSCs between PGRN KO and WT mice. These results suggest that PGRN production by osteoblast lineage cells is not responsible for a suppression of bone formation. PGRN is a secreted peptide that can be detected in cell culture, serum, and body fluid, and the effects of PGRN on bone formation appear to be non–cell autonomous. PGRN is abundantly expressed by adipose tissue and recently identified as a proinflammatory adipokine (6, 11). Adipocytes have been demonstrated to inhibit osteoblast differentiation by producing TNF-α (37). Oh et al. (7) have shown that ovariectomy led to an increase in PGRN levels in mouse bone marrow and serum. In the current study, we found that TNF-α mRNA levels were significantly lower in the bone marrow of PGRN KO mice. These results suggested that adipocytic PGRN might have inhibited bone formation in female mice. However, the specific deletion of PGRN in adipocytes using adiponectin Cre did not produce a bone phenotype. Progranulin is best known as a glial factor that is deficient in a subset of patients with FTD. Recent studies have demonstrated that microglial PGRN contributes to the cognitive effects of PGRN (11, 18, 19). Microglial cells are macrophages residing in the nervous system. We hypothesized that microglial/macrophage-derived PGRN might account for the age-related bone loss in female mice. Indeed, a specific deletion of PGRN in microglial/macrophages using Cx3cr1-Cre resulted in a high bone mass phenotype in aged female mice. This result recapitulates that seen in global PGRN KO mice. Although PGRN is highly expressed in microglial cells, the mechanism by which microglial PGRN inhibits bone formation is not clear. It has recently been demonstrated that disrupting estrogen signaling in kisspeptin neurons in the hypothalamus led to high bone mass in female (but not male) mice (38). Therefore, it is possible that microglial PGRN regulates bone formation through an effect on hypothalamic neurons. Cx3cr1 is expressed in peripheral macrophages as well (39, 40). We cannot exclude the possible contribution of peripheral macrophages to the high bone mass phenotype with the specific deletion of Grn in Cx3cr1+ cells. Loss of PGRN impaired osteoclastic bone resorption independent of sex. However, there was an increase in TRAP+ cells at trabecular BSs in female global or conditional PGRN KO mouse bones. It is not clear if these increased TRAP+ cells are true osteoclasts because PGRN is required for osteoclast fusion and maturation (7). The increased TRAP+ cells were possibly macrophages recruited to BSs. It is well known that macrophages within the bone marrow cavity are involved in maintaining skeletal homeostasis (35, 41). Indeed, we demonstrated that loss of PGRN also stimulated the expression of genes associated with immature osteoclasts (Csf1r, Adgre1, and C1q) (42, 43) and enhanced the mRNA levels of Osm and C3 in the mouse osteoclast cultures. The molecules Osm, C3, and C1q are involved in bone remodeling in favor of bone formation (6, 33, 34). Further work is needed to determine if peripheral macrophage-derived PGRN contributes to the high bone mass phenotype in female mice. In conclusion, our study demonstrated that PGRN is an important factor involved in maintaining bone homeostasis and metabolism in female mice. Loss of PGRN inhibited the maturation and function of osteoclasts regardless of sex and resulted in increased bone formation in females only. The critical source of PGRN in suppressing bone formation is microglial/macrophage lineage cells. Our results implicate PGRN derived from microglial cells/macrophages as a potential therapeutic target for the treatment of osteoporosis in females. Acknowledgments We thank Dr. Robert V. Farese (Gladstone Institute of Cardiovascular Disease, UCSF) for generously providing the PGRN KO mice and the bone samples from R493X mutant mice, and Dr. Eric Huang and Dr. Jiasheng Zhang (Pathology, UCSF) for providing the bone samples from the mice with conditional deletion of PGRN in Cx3cr1+ cells. We also thank the San Francisco VA Medical Center Bone Core for its technical support. Financial Support: This work was supported by a US Department of Veterans Affairs Merit Review grant (1I01BX003212 to R.A.N.), a National Institutes of Health P30 grant (AR066262), and the UCSF Resource Allocation Program grant award (to L.W.). Additional Information Disclosure Summary: The authors have nothing to disclose. Data Availability: All data generated or analyzed during this study are included in this published article or in the data repositories listed in References. Abbreviations: Abbreviations: BFR bone formation rate BMM bone marrow–derived macrophage BMSC bone marrow stromal cell BS bone surface BV bone fractional volume Ct threshold cycle d day FTD frontotemporal dementia Grn progranulin gene KO knockout MAR mineral apposition rate N.Ob number of osteoblasts Oc osteoclast Osm oncostatin M Ovx ovariectomized PCM primary culture medium PGRN progranulin PINP procollagen type I amino-terminal propeptide PYD pyridinoline T2D type 2 diabetes TFJ tibiofibular junction TRAP tartrate-resistant acid phosphatase TV tissue volume UCSF University of California, San Francisco WT wild-type μCT micro-CT References and Notes 1. Mackie EJ . Osteoblasts: novel roles in orchestration of skeletal architecture . Int J Biochem Cell Biol . 2003 ; 35 ( 9 ): 1301 – 1305 . Google Scholar Crossref Search ADS PubMed WorldCat 2. Khosla S , Westendorf JJ , Oursler MJ . Building bone to reverse osteoporosis and repair fractures . J Clin Invest . 2008 ; 118 ( 2 ): 421 – 428 . Google Scholar Crossref Search ADS PubMed WorldCat 3. He Z , Bateman A . Progranulin (granulin-epithelin precursor, PC-cell-derived growth factor, acrogranin) mediates tissue repair and tumorigenesis . J Mol Med (Berl) . 2003 ; 81 ( 10 ): 600 – 612 . Google Scholar Crossref Search ADS PubMed WorldCat 4. Cruts M , Van Broeckhoven C . Loss of progranulin function in frontotemporal lobar degeneration . Trends Genet . 2008 ; 24 ( 4 ): 186 – 194 . Google Scholar Crossref Search ADS PubMed WorldCat 5. Youn BS , Bang SI , Klöting N , Park JW , Lee N , Oh JE , Pi KB , Lee TH , Ruschke K , Fasshauer M , Stumvoll M , Blüher M . Serum progranulin concentrations may be associated with macrophage infiltration into omental adipose tissue . Diabetes . 2009 ; 58 ( 3 ): 627 – 636 . Google Scholar Crossref Search ADS PubMed WorldCat 6. Matsubara T , Mita A , Minami K , Hosooka T , Kitazawa S , Takahashi K , Tamori Y , Yokoi N , Watanabe M , Matsuo E , Nishimura O , Seino S . PGRN is a key adipokine mediating high fat diet-induced insulin resistance and obesity through IL-6 in adipose tissue . Cell Metab . 2012 ; 15 ( 1 ): 38 – 50 . Google Scholar Crossref Search ADS PubMed WorldCat 7. Oh J , Kim JY , Kim HS , Oh JC , Cheon YH , Park J , Yoon KH , Lee MS , Youn BS . Progranulin and a five transmembrane domain-containing receptor-like gene are the key components in receptor activator of nuclear factor κB (RANK)-dependent formation of multinucleated osteoclasts . J Biol Chem . 2015 ; 290 ( 4 ): 2042 – 2052 . Google Scholar Crossref Search ADS PubMed WorldCat 8. Limonard EJ , Veldhuis-Vlug AG , van Dussen L , Runge JH , Tanck MW , Endert E , Heijboer AC , Fliers E , Hollak CE , Akkerman EM , Bisschop PH . Short-term effect of estrogen on human bone marrow fat . J Bone Miner Res . 2015 ; 30 ( 11 ): 2058 – 2066 . Google Scholar Crossref Search ADS PubMed WorldCat 9. Feng JQ , Guo FJ , Jiang BC , Zhang Y , Frenkel S , Wang DW , Tang W , Xie Y , Liu CJ . Granulin epithelin precursor: a bone morphogenic protein 2-inducible growth factor that activates Erk1/2 signaling and JunB transcription factor in chondrogenesis . FASEB J . 2010 ; 24 ( 6 ): 1879 – 1892 . Google Scholar Crossref Search ADS PubMed WorldCat 10. Xu K , Zhang Y , Ilalov K , Carlson CS , Feng JQ , Di Cesare PE , Liu CJ . Cartilage oligomeric matrix protein associates with granulin-epithelin precursor (GEP) and potentiates GEP-stimulated chondrocyte proliferation . J Biol Chem . 2007 ; 282 ( 15 ): 11347 – 11355 . Google Scholar Crossref Search ADS PubMed WorldCat 11. Nguyen AD , Nguyen TA , Martens LH , Mitic LL , Farese RV Jr . Progranulin: at the interface of neurodegenerative and metabolic diseases . Trends Endocrinol Metab . 2013 ; 24 ( 12 ): 597 – 606 . Google Scholar Crossref Search ADS PubMed WorldCat 12. Takeda S , Elefteriou F , Levasseur R , Liu X , Zhao L , Parker KL , Armstrong D , Ducy P , Karsenty G . Leptin regulates bone formation via the sympathetic nervous system . Cell . 2002 ; 111 ( 3 ): 305 – 317 . Google Scholar Crossref Search ADS PubMed WorldCat 13. Kajimura D , Lee HW , Riley KJ , Arteaga-Solis E , Ferron M , Zhou B , Clarke CJ , Hannun YA , DePinho RA , Guo XE , Mann JJ , Karsenty G . Adiponectin regulates bone mass via opposite central and peripheral mechanisms through FoxO1 [published correction appears in Cell Metab. 2014;19(5):891] . Cell Metab . 2013 ; 17 ( 6 ): 901 – 915 . Google Scholar Crossref Search ADS PubMed WorldCat 14. Wattanachanya L , Lu WD , Kundu RK , Wang L , Abbott MJ , O’Carroll D , Quertermous T , Nissenson RA . Increased bone mass in mice lacking the adipokine apelin . Endocrinology . 2013 ; 154 ( 6 ): 2069 – 2080 . Google Scholar Crossref Search ADS PubMed WorldCat 15. Walsh JS , Vilaca T . Obesity, type 2 diabetes and bone in adults . Calcif Tissue Int . 2017 ; 100 ( 5 ): 528 – 535 . Google Scholar Crossref Search ADS PubMed WorldCat 16. Siris ES . Clinical issues with bisphosphonate therapy for osteoporosis. Introduction . Am J Med . 2009 ; 122 ( 2 Suppl ): S1 – S2 . Google Scholar Crossref Search ADS PubMed WorldCat 17. Lewiecki EM . New targets for intervention in the treatment of postmenopausal osteoporosis . Nat Rev Rheumatol . 2011 ; 7 ( 11 ): 631 – 638 . Google Scholar Crossref Search ADS PubMed WorldCat 18. Martens LH , Zhang J , Barmada SJ , Zhou P , Kamiya S , Sun B , Min SW , Gan L , Finkbeiner S , Huang EJ , Farese RV Jr . Progranulin deficiency promotes neuroinflammation and neuron loss following toxin-induced injury . J Clin Invest . 2012 ; 122 ( 11 ): 3955 – 3959 . Google Scholar Crossref Search ADS PubMed WorldCat 19. Lui H , Zhang J , Makinson SR , Cahill MK , Kelley KW , Huang HY , Shang Y , Oldham MC , Martens LH , Gao F , Coppola G , Sloan SA , Hsieh CL , Kim CC , Bigio EH , Weintraub S , Mesulam MM , Rademakers R , Mackenzie IR , Seeley WW , Karydas A , Miller BL , Borroni B , Ghidoni R , Farese RV Jr , Paz JT , Barres BA , Huang EJ . Progranulin deficiency promotes circuit-specific synaptic pruning by microglia via complement activation . Cell . 2016 ; 165 ( 4 ): 921 – 935 . Google Scholar Crossref Search ADS PubMed WorldCat 20. Nguyen AD , Nguyen TA , Zhang J , Devireddy S , Zhou P , Karydas AM , Xu X , Miller BL , Rigo F , Ferguson SM , Huang EJ , Walther TC , Farese RV Jr . Murine knockin model for progranulin-deficient frontotemporal dementia with nonsense-mediated mRNA decay . Proc Natl Acad Sci USA . 2018 ; 115 ( 12 ): E2849 – E2858 . Google Scholar Crossref Search ADS PubMed WorldCat 21. Millard SM , Louie AM , Wattanachanya L , Wronski TJ , Conklin BR , Nissenson RA . Blockade of receptor-activated G(i) signaling in osteoblasts in vivo leads to site-specific increases in cortical and cancellous bone formation . J Bone Miner Res . 2011 ; 26 ( 4 ): 822 – 832 . Google Scholar Crossref Search ADS PubMed WorldCat 22. Kao RS , Abbott MJ , Louie A , O’Carroll D , Lu W , Nissenson R . Constitutive protein kinase A activity in osteocytes and late osteoblasts produces an anabolic effect on bone . Bone . 2013 ; 55 ( 2 ): 277 – 287 . Google Scholar Crossref Search ADS PubMed WorldCat 23. Abbott MJ , Roth TM , Ho L , Wang L , O’Carroll D , Nissenson RA . Negative skeletal effects of locally produced adiponectin . PLoS One . 2015 ; 10 ( 7 ): e0134290 . Google Scholar Crossref Search ADS PubMed WorldCat 24. Dempster DW , Compston JE , Drezner MK , Glorieux FH , Kanis JA , Malluche H , Meunier PJ , Ott SM , Recker RR , Parfitt AM . Standardized nomenclature, symbols, and units for bone histomorphometry: a 2012 update of the report of the ASBMR Histomorphometry Nomenclature Committee . J Bone Miner Res . 2013 ; 28 ( 1 ): 2 – 17 . Google Scholar Crossref Search ADS PubMed WorldCat 25. Schmittgen TD , Lee EJ , Jiang J , Sarkar A , Yang L , Elton TS , Chen C . Real-time PCR quantification of precursor and mature microRNA . Methods . 2008 ; 44 ( 1 ): 31 – 38 . Google Scholar Crossref Search ADS PubMed WorldCat 26. RRID:AB_2783011, https://scicrunch.org/resolver/AB_2783011 . 27. RRID:AB_2114489, https://scicrunch.org/resolver/AB_2114489 . 28. RRID:AB_2336814, https://scicrunch.org/resolver/AB_2336814 . 29. Kao R , Lu W , Louie A , Nissenson R . Cyclic AMP signaling in bone marrow stromal cells has reciprocal effects on the ability of mesenchymal stem cells to differentiate into mature osteoblasts versus mature adipocytes . Endocrine . 2012 ; 42 ( 3 ): 622 – 636 . Google Scholar Crossref Search ADS PubMed WorldCat 30. Yin F , Banerjee R , Thomas B , Zhou P , Qian L , Jia T , Ma X , Ma Y , Iadecola C , Beal MF , Nathan C , Ding A . Exaggerated inflammation, impaired host defense, and neuropathology in progranulin-deficient mice . J Exp Med . 2010 ; 207 ( 1 ): 117 – 128 . Google Scholar Crossref Search ADS PubMed WorldCat 31. Glatt V , Canalis E , Stadmeyer L , Bouxsein ML . Age-related changes in trabecular architecture differ in female and male C57BL/6J mice . J Bone Miner Res . 2007 ; 22 ( 8 ): 1197 – 1207 . Google Scholar Crossref Search ADS PubMed WorldCat 32. Huey ED , Grafman J , Wassermann EM , Pietrini P , Tierney MC , Ghetti B , Spina S , Baker M , Hutton M , Elder JW , Berger SL , Heflin KA , Hardy J , Momeni P . Characteristics of frontotemporal dementia patients with a Progranulin mutation . Ann Neurol . 2006 ; 60 ( 3 ): 374 – 380 . Google Scholar Crossref Search ADS PubMed WorldCat 33. Guihard P , Danger Y , Brounais B , David E , Brion R , Delecrin J , Richards CD , Chevalier S , Rédini F , Heymann D , Gascan H , Blanchard F . Induction of osteogenesis in mesenchymal stem cells by activated monocytes/macrophages depends on oncostatin M signaling . Stem Cells . 2012 ; 30 ( 4 ): 762 – 772 . Google Scholar Crossref Search ADS PubMed WorldCat 34. Naito AT , Sumida T , Nomura S , Liu ML , Higo T , Nakagawa A , Okada K , Sakai T , Hashimoto A , Hara Y , Shimizu I , Zhu W , Toko H , Katada A , Akazawa H , Oka T , Lee JK , Minamino T , Nagai T , Walsh K , Kikuchi A , Matsumoto M , Botto M , Shiojima I , Komuro I . Complement C1q activates canonical Wnt signaling and promotes aging-related phenotypes . Cell . 2012 ; 149 ( 6 ): 1298 – 1313 . Google Scholar Crossref Search ADS PubMed WorldCat 35. Chang MK , Raggatt LJ , Alexander KA , Kuliwaba JS , Fazzalari NL , Schroder K , Maylin ER , Ripoll VM , Hume DA , Pettit AR . Osteal tissue macrophages are intercalated throughout human and mouse bone lining tissues and regulate osteoblast function in vitro and in vivo . J Immunol . 2008 ; 181 ( 2 ): 1232 – 1244 . Google Scholar Crossref Search ADS PubMed WorldCat 36. Noguchi T , Ebina K , Hirao M , Kawase R , Ohama T , Yamashita S , Morimoto T , Koizumi K , Kitaguchi K , Matsuoka H , Kaneshiro S , Yoshikawa H . Progranulin plays crucial roles in preserving bone mass by inhibiting TNF-α-induced osteoclastogenesis and promoting osteoblastic differentiation in mice . Biochem Biophys Res Commun . 2015 ; 465 ( 3 ): 638 – 643 . Google Scholar Crossref Search ADS PubMed WorldCat 37. Abuna RP , De Oliveira FS , Santos TS , Guerra TR , Rosa AL , Beloti MM . Participation of TNF-α in inhibitory effects of adipocytes on osteoblast differentiation . J Cell Physiol . 2016 ; 231 ( 1 ): 204 – 214 . Google Scholar Crossref Search ADS PubMed WorldCat 38. Herber CB , Krause WC , Wang L , Bayrer JR , Li A , Schmitz M , Fields A , Ford B , Zhang Z , Reid MS , Nomura DK , Nissenson RA , Correa SM , Ingraham HA . Estrogen signaling in arcuate Kiss1 neurons suppresses a sex-dependent female circuit promoting dense strong bones . Nat Commun . 2019 ; 10 ( 1 ): 163 . Google Scholar Crossref Search ADS PubMed WorldCat 39. Yona S , Kim KW , Wolf Y , Mildner A , Varol D , Breker M , Strauss-Ayali D , Viukov S , Guilliams M , Misharin A , Hume DA , Perlman H , Malissen B , Zelzer E , Jung S . Fate mapping reveals origins and dynamics of monocytes and tissue macrophages under homeostasis [published correction appears in Immunity. 2013;38(5):1073–1079] . Immunity . 2013 ; 38 ( 1 ): 79 – 91 . Google Scholar Crossref Search ADS PubMed WorldCat 40. Fogg DK , Sibon C , Miled C , Jung S , Aucouturier P , Littman DR , Cumano A , Geissmann F . A clonogenic bone marrow progenitor specific for macrophages and dendritic cells . Science . 2006 ; 311 ( 5757 ): 83 – 87 . Google Scholar Crossref Search ADS PubMed WorldCat 41. Pettit AR , Chang MK , Hume DA , Raggatt LJ . Osteal macrophages: a new twist on coupling during bone dynamics . Bone . 2008 ; 43 ( 6 ): 976 – 982 . Google Scholar Crossref Search ADS PubMed WorldCat 42. Teo BH , Bobryshev YV , Teh BK , Wong SH , Lu J . Complement C1q production by osteoclasts and its regulation of osteoclast development . Biochem J . 2012 ; 447 ( 2 ): 229 – 237 . Google Scholar Crossref Search ADS PubMed WorldCat 43. Cappellen D , Luong-Nguyen NH , Bongiovanni S , Grenet O , Wanke C , Susa M . Transcriptional program of mouse osteoclast differentiation governed by the macrophage colony-stimulating factor and the ligand for the receptor activator of NFkappa B . J Biol Chem . 2002 ; 277 ( 24 ): 21971 – 21982 . Google Scholar Crossref Search ADS PubMed WorldCat Copyright © 2019 Endocrine Society
TI - Female-Specific Role of Progranulin to Suppress Bone Formation
JF - Endocrinology
DO - 10.1210/en.2018-00842
DA - 2019-09-01
UR - https://www.deepdyve.com/lp/oxford-university-press/female-specific-role-of-progranulin-to-suppress-bone-formation-TBSvmCOx7C
SP - 2024
VL - 160
IS - 9
DP - DeepDyve
ER -