TY - JOUR
AU - Shin, Chan Soo
AB - Abstract Postmenopausal osteoporosis is characterized by increased bone resorption due to estrogen deficiency. Receptor activator of nuclear factor‐κB‐Fc (RANK‐Fc), a fusion protein that specifically blocks receptor activator of nuclear factor ligand binding to RANK, has been known to be efficient and well tolerated in animal models of osteoporosis. Here we show that cell‐based gene therapy with RANK‐Fc effectively prevented bone loss in ovariectomized (OVX) mice. Thirty‐one young adult female C57Bl/6 mice were used, and repeated intraperitoneal injection of mesenchymal stem cells (MSCs) transduced with retrovirus was performed as follows: 1) Sham‐operated mice (n = 8); 2) OVX mice treated with phosphate‐buffered saline (OVX‐PBS; n = 8); 3) OVX mice injected with MSCs transduced with control retrovirus (OVX‐green fluorescent protein [GFP]; n = 7); and 4) OVX mice injected with MSCs transduced with RANK‐Fc (OVX‐RANK‐Fc; n = 8). Cellular expression of RANK‐Fc was confirmed by Western blot analysis of cell lysates and conditioned medium and also by enzyme‐linked immunosorbent assay for the mice serum. Measurement of bone mineral density (BMD) by dual‐energy x‐ray absorptiometry (PIXImus) revealed that the OVX‐RANK‐Fc group gained significantly higher BMD than either the OVX‐PBS group or OVX‐GFP group after 8 weeks. The expression of GFP, which is coexpressed with RANK‐Fc, was detected by polymerase chain reaction analysis of DNA isolated from femur and intra‐abdominal fat, whereas no GFP signal was identified in liver, brain, heart, lung, or bone marrow aspirates. These suggest that expression of RANK‐Fc by genetically modified MSCs may be a feasible option for the prevention of bone loss induced by ovariectomy. Receptor activator of nuclear factor‐κB, Receptor activator of nuclear factor‐κB‐Fc, Mesenchymal stem cell, Gene therapy, Osteoporosis, Ovariectomy Introduction Osteoporosis is defined as a skeletal disorder characterized by compromised bone strength, which predisposes increased risk of fracture, most commonly in the vertebrae, wrist, and hip [1]. Most of the current therapies for osteoporosis belong to antiresorptive agents that target the bone resorbing activity of the osteoclasts. Given that osteoporosis is a chronic condition, with slow progression as well as slow response to therapeutic agents, development of an alternative safe and effective measure that could provide sustained antiresorptive effects is necessary. In this regard, ex vivo gene therapy using genetically modified mesenchymal stem cells (MSCs) expressing soluble anti‐resorptive factors might be an option. MSCs represent an attractive cellular vehicle for the in vivo delivery of therapeutic gene products, since they are easily obtained by bone marrow aspiration, rapidly expanded in vitro, readily transduced with a variety of viral vectors, and survive long‐term in vivo following retroviral transduction [2, 3]. Indeed, in baboon model of aplastic anemia, human erythropoietin (hEPO)‐expressing MSCs could be successfully implanted into allogenic recipient and capable of secreting hEPO for up to 137 days, resulting in a significant increase in hematocrit [4]. Osteoclasts, the bone‐resorbing cells in the bone marrow environment, are derived from the hematopoietic macrophage system through the interactions between the receptor activator of the nuclear factor‐κB ligand (RANKL)/RANK system (reviewed in ref. [5]). RANKL, a transmembrane molecule located on bone marrow stromal cells and osteoblasts, binds to RANK present on the surface of osteoclast precursors. This ligand‐receptor interaction activates NF‐κB, which in turn stimulates the differentiation of osteoclast precursors to osteoclasts. Osteoprotegerin (OPG) produced by osteoblasts and stromal cells binds to RANKL, sequestering and preventing it from binding to RANK, which results in inhibition of osteoclast formation and survival [6–8]. Animals with null mutations of the OPG gene exhibit severe osteoporosis [9], whereas overexpression of OPG in transgenic mice and OPG treatment of normal rodents leads to profound osteopetrosis [6]. Mice with ablated RANKL or RANK genes exhibit osteopetrosis, impaired tooth eruption, and lymph node agenesis [10, 11], whereas RANK activation by administration of RANKL or stimulatory RANK antibodies promotes osteoclastogenesis [7, 8]. The critical role of RANKL/RANK/OPG in osteoclastogenesis has led to the development of a new therapeutic strategy, targeting this system. For example, administration of a soluble OPG or inhibitory RANK antibodies blocks osteoclastogenesis and prevents bone loss in animal models of ovariectomy‐associated estrogen deficiency [6], arthritis [12], and osteolytic metastasis [13]. Moreover, results from the first human trial with OPG support its potential as a therapeutic agent for osteoporosis and bone metastasis [14, 15]. RANK‐Fc is a recombinant protein of the extracellular domain of RANK fused to the Fc region of human IgG and functions as a soluble antagonist against RANKL [16]. Fusion of RANK to IgG Fc dictates homodimerization, which probably increases its avidity for RANKL. Addition of RANK‐Fc in vitro virtually eliminates the formation of osteoclasts in cocultures of myeloma bone marrow and osteoblast/stromal cells [17], and transgenic mice overexpressing RANK‐Fc have severe osteopetrosis because of a reduction in osteoclasts, similar to OPG transgenic mice [7]. In addition, subcutaneous injection of RANK‐Fc protein into rapidly growing young mice (aged 3–4 weeks) resulted in a dose‐dependent increase in total bone density as compared with control group [7]. The RANK‐Fc has the potential advantage over OPG of greater specificity for RANKL. The introduction of RANK‐Fc via MSCs could be an option that could in theory overcome a number of conventional problems of recombinant protein technology, including complexity of purification, unstable biological activity, antibody formation, toxicity, and the need for frequent administration due to short half‐life. These problems may be solved using gene delivery to express the targeted protein [18]. This study was undertaken to test whether introduction of the MSCs genetically modified with a retrovirus engineered to express mouse RANK‐Fc could prevent the bone loss induced by ovariectomy in mice. Materials and Methods Animals Female C57Bl/6 mice (5–6 weeks of age) were purchased from Orient Corporation (Kapyoung, Gyeonggi, Korea). All mice used in our experiments were maintained in the animal facility at Seoul National University Hospital Clinical Research Institute, in accordance with the animal care guidelines stipulated by the hospital. Diets and tap water were provided ad libitum throughout the study. Retroviral Vector Construction The pMT‐RANK‐Fc vector containing a DNA sequence encoding the extracellular domain of mouse RANK (Met1‐Pro213) fused via a linker to the Fc region of human immunoglobulin G1 (IgG1) was provided by Dr. Jaerang Rho (Chungnam National University, Daejon, Korea) and has been described previously [17]. To investigate the functional effects of sustained RANK‐Fc on bone resorption, a retroviral vector encoding RANK‐Fc was generated by introducing the RANK‐Fc cDNA into the EcoRI/NotI site of p‐murine stem cell virus‐internal ribosome entry site‐enhanced green fluorescent protein (plasmid [p]MSCV‐IRES‐eGFP) (pMSCV‐eGFP, a kind gift from Dr. Ronald L. Johnson, University of Alabama at Birmingham, Birmingham, AL) [19] downstream of the MSCV long terminal repeat (LTR) and upstream of the IRES‐eGFP cassette (Fig. 1), thereby allowing expression of both RANK‐Fc and eGFP from a single bicistronic mRNA. The resulting construct was designated pMSCV‐RANK‐Fc‐IRES‐eGFP (pMSCV‐RANK‐Fc‐eGFP; Fig. 1). In this system, gene expression is under the transcriptional control of MSCV LTR. The nucleotide sequence of the RANK‐Fc coding region was confirmed by sequencing. Figure 1. Open in new tabDownload slide Schematic representation of retroviral vectors used. RANK‐Fc cDNA is located downstream of MSCV LTR and upstream of IRES‐eGFP cassette, allowing the bicistronic expression of both RANK‐Fc and eGFP from a single bicistronic mRNA. Abbreviations: eGFP, enhanced green fluorescent protein gene; IRES, internal ribosomal entry site of the encephalomyocarditis virus; MSCV‐LTR, Moloney murine leukemia virus long terminal repeat; RANK‐Fc, fusion protein of receptor activator of nuclear factor‐κB (Met1–Pro213) with Fc region of human immunoglobulin G1. Figure 1. Open in new tabDownload slide Schematic representation of retroviral vectors used. RANK‐Fc cDNA is located downstream of MSCV LTR and upstream of IRES‐eGFP cassette, allowing the bicistronic expression of both RANK‐Fc and eGFP from a single bicistronic mRNA. Abbreviations: eGFP, enhanced green fluorescent protein gene; IRES, internal ribosomal entry site of the encephalomyocarditis virus; MSCV‐LTR, Moloney murine leukemia virus long terminal repeat; RANK‐Fc, fusion protein of receptor activator of nuclear factor‐κB (Met1–Pro213) with Fc region of human immunoglobulin G1. Isolation and Culture Expansion of MSCs Female C57Bl/6 mice (6–8 weeks old) were sacrificed by cervical dislocation. MSCs were isolated from bone marrow of the femur and tibia by inserting a 27‐gauge needle into the shaft of the bone and flushing with 1 ml of α‐minimum essential medium (αMEM; Sigma‐Aldrich, St. Louis, http://www.sigmaaldrich.com) supplemented with 10% fetal bovine serum (FBS; Cambrex, Walkersville, MD, http://www.cambrex.com), 2 mM l‐glutamine, and 100 U/ml penicillin/100 μg/ml streptomycin (Invitrogen, Grand Island, NY, http://www.invitrogen.com). Cells were filtered through a 70‐μm nylon filter (Falcon, Franklin Lakes, NJ), centrifuged for 5 minutes at 260g, suspended, and plated out in complete Dulbecco's modified Eagle's medium‐high glucose (DMEM; Invitrogen) containing 15% horse serum (Invitrogen), 15% fetal calf serum (Invitrogen), 2 mM l‐glutamine, 10–4 M 2‐mercaptoethanol (Sigma‐Aldrich), 10–6 M hydrocortisone (Sigma‐Aldrich), and 100 U/ml penicillin‐100 μg/ml streptomycin, at a density of 1 × 106 nucleated cells per ml in 25‐cm2 culture flasks (Techno Plastic Products AG, Trasadingen, Switzerland). Cells were grown in complete medium at 37°C at 5% CO2 for 7 days, and the medium was replaced every 2–3 days. Adherent cells were grown to 90% confluence, washed with phosphate‐buffered saline (PBS), and incubated with 0.25% trypsin/2 mM EDTA (Invitrogen). Nondetached cells were discarded, and the remaining cells were designated passage 1 (P1). Confluent MSCs were passaged further and plated at a dilution of 1:2 to 1:3. P2 to P5 MSCs were transduced with retrovirus. Establishment of MSCs Overexpressing RANK‐Fc To obtain the viral supernatant, 293T packaging cells were transfected with the retroviral construct using Lipofectamine PLUS (Invitrogen). Briefly, 24 hours prior to transfection, 5 × 106 293T cells were plated in 100‐mm dishes containing DMEM supplemented with 10% FBS and 100 U/ml penicillin‐100 μg/ml streptomycin. The transfection solution contained DNA (4 μg of pMD‐gag‐pol, 4 μg of pMD‐VSVG, and 4 μg of retroviral vector pMSCV‐eGFP or pMSCV‐RANK‐Fc‐eGFP alone), Lipofectamine PLUS, and serum‐free DMEM. The medium was replaced after 3 hours. Viral supernatant fractions were collected at 48 hours, filtered through a 0.45‐μm syringe filter (Nalgene, Rochester, NY), and stored at –80°C. MSCs were plated on six‐well plates at a density of 1.5 × 105 cells per well, 24 hours prior to retroviral transduction. The culture medium was removed, and retroviral supernatant (MSCV‐eGFP or MSCV‐RANK‐Fc‐eGFP) with polybrene (Sigma‐Aldrich) at a final concentration of 4 μg/ml was added to the culture. Culture plates were centrifuged at 2,800 rpm at 32°C for 90 minutes and incubated at 37°C for 2 hours. A second round of infection was performed as described above, and the retrovirus supernatant was replaced with culture medium for overnight incubation. Following the third and fourth rounds of infection, cells were cultured until near confluency. Transduced cells were directly evaluated for GFP expression using a standard fluorescent microscope (Olympus IX70) and flow cytometry with a fluorescence‐activated cell sorter (FACS) (FACS Vantage SE; Becton, Dickinson and Company, Franklin Lakes, NJ, http://www.bd.com) at 530 ± 30 nm. Expression of RANK‐Fc In Vitro To assess the cellular expression of RANK‐Fc, Western blot analysis was performed using cell lysates from MSCs transduced with MSCV‐RANK‐Fc‐eGFP or empty MSCV‐eGFP retrovirus as described previously [20]. Briefly, cell lysates were prepared as follows: cells were treated with lysis buffer (150 mM NaCl, 50 mM Tris‐Cl, pH 7.4, 20 mM EDTA, 1% Triton X‐100, 1% sodium deoxycholate, 0.1% SDS, and protease inhibitors), fractionated by SDS‐polyacrylamide gel electrophoresis (10% gels), and transferred to nitrocellulose membranes. Blots were probed with goat polyclonal anti‐mouse RANK antibody (0.1 μg/ml) or goat polyclonal anti‐human IgG Fcγ fragment‐specific antibody (1:1,000; Jackson Laboratory, Bar Harbor, ME, http://www.jax.org), followed by horseradish peroxidase (HRP)‐conjugated polyclonal anti‐goat IgG (1:3,000; Santa Cruz Biotechnology Inc., Santa Cruz, CA, http://www.scbt.com) and detected by enhanced chemiluminescence using an enhanced chemiluminescence kit (Amersham Biosciences, Piscataway, NJ, http://www.amersham.com), according to the manufacturer's instructions. For assaying RANK‐Fc secretion in vitro, serum‐free DMEM was added to confluent plates of MSCs transduced with MSCV‐RANK‐Fc‐eGFP or MSCV‐eGFP. After 24 hours, culture supernatants were incubated at 4°C overnight with goat polyclonal anti‐mouse RANK antibody (1 μg/ml; R&D Systems), followed by a 4°C overnight incubation with 20 μl of the agarose conjugate suspension (protein A/G‐agarose; Santa Cruz Biotechnology Inc.). Samples were centrifuged at 2,500 rpm for 30 seconds at 4°C, and the supernatant was aspirated. Next, pellets were washed three times and suspended with 40 μl of 2× electrophoresis sample buffer. Samples were boiled for 5 minutes and then subjected to Western blot analysis using anti‐mouse RANK antibody or anti‐human IgG Fcγ fragment‐specific antibody as above. Experimental Design Mice were randomly divided into four groups as follows: 1) Sham‐operated mice (SHAM; n = 8); 2) OVX mice treated with PBS (OVX‐PBS; n = 8); 3) OVX mice injected with MSCs transduced with control retrovirus (OVX‐GFP; n = 7); and 4) OVX mice injected with MSCs transduced with RANK‐Fc (OVX‐RANK‐Fc; n = 8). Bilateral ovariectomy in mice is a standard method to evaluate agents that ameliorate bone loss associated with estrogen deficiency [21]. On study day –1, mice were anesthetized with ketamine (50 mg/kg) and xylazine (15 mg/kg) to allow exposure of the ovaries. Specifically, the gonads were removed in OVX groups but only manipulated in the SHAM cohorts. To determine whether prophylactic treatment with RANK‐Fc prevents ovariectomy‐induced bone loss in an osteoporosis model, animals underwent repeated intraperitoneal injection of the indicated doses of transduced cells (OVX‐GFP, OVX‐RANK‐Fc) or PBS (SHAM, OVX‐PBS) on days 0, 2, 4, and 6. Transduced MSCs were trypsinized, centrifuged, and suspended at a concentration of 1 × 107 nucleated cells per milliliter of PBS. Suspensions of transduced MSCs were intraperitoneally injected into C57Bl/6 mice in a total volume of 0.2 ml. Intraperitoneal injections were repeated every other day for a total of four doses, based on a previous study [22]. Sera were collected on the indicated days. Blood samples from retro‐orbital capillaries were taken under anesthesia with diethyl ether (Duksan Chemical, Duksan, Korea), and serum was recovered by centrifugation at 8,000g for 10 minutes. Samples were stored at –80°C until analysis. Bone Mineral Density Measurement Bone mineral density (BMD) and body composition were measured using a Lunar PIXImus densitometer (software version 2.0; GE Lunar, Madison, WI) on indicated days. PIXImus scanner is a proven device for bone mineral and body composition measurements of mice and other small animals weighing 10–50 g. PIXImus acquires images in less than 5 minutes with dual‐energy x‐ray absorptiometry (DEXA). Mice were anesthetized with a mixture of ketamine (50 mg/kg) and xylazine (15 mg/kg) in PBS and placed prone on the platform of the PIXImus, and whole body bone mineral content (BMC; mg) and areal BMD (g/cm2) with the exclusion of head was acquired with mouse‐specific software (version 1.47). Basically, BMD is calculated by dividing the BMC by projected body area. Body fat content can be also obtained by the same software. All DEXA scans were conducted by the same researcher (D.K.). Percent coefficient of variation of BMC and BMD for the repeated scans was 1%–2%. Expression of RANK‐Fc In Vivo For enzyme‐linked immunosorbent assay (ELISA), 96‐well microplates were coated with goat polyclonal anti‐mouse RANK antibody (1 μg/well) overnight at 4°C, washed three times with PBS‐Tween (0.5% Tween 20 per liter, pH 7.4), and incubated at room temperature with blocking solution (1.59% Na2CO3, 2.93% NaHCO3, 0.2% NaN3, 1% BSA, 5% sucrose) for 1–2 hours. Wells were washed, incubated overnight at 4°C with RANK‐Fc protein standard prepared in Sf9 cells (a kind gift from Dr. Jaerang Rho, Chungnam National University, Daejeon, Korea) [17] and diluted culture supernatant or mouse serum samples (100 μl); rewashed; and treated with HRP‐conjugated goat anti‐human IgG Fcγ fragment‐specific antibody (1:5,000; Jackson Laboratory) for 2–4 hours. Plates were washed and incubated at room temperature with substrate solution (tetramethylbenzidine base; Sigma‐Aldrich) for 20 minutes. Following the addition of stop solution (1 M H2SO2), the optical density of each well was determined using a microplate reader set (ThermoMax; Scientific Surplus, Belle Mead, NJ) at 450 nm. DNA PCR for GFP Femurs from sacrificed mice were dissected free of surrounding tissue and fixed in 4% paraformaldehyde‐PBS (pH 7.4 with 10 N NaOH) at 4°C for 3 days. Soft tissues (liver, brain, heart, lung, and intra‐abdominal fat) were dissected and placed in paraformaldehyde overnight. Bone marrow cells were isolated from the femur and tibia by inserting a 27‐gauge needle into the shaft of the bone and flushing with 1 ml of α‐minimum essential medium as described above. To detect GFP by polymerase chain reaction (PCR), tissues and bone marrow cells were placed in a lysis buffer containing 100 mM NaCl, 100 mM Tris‐Cl (pH 8.0), 25 mM EDTA (pH 8), 0.5% sodium dodecyl sulfate, and 0.1 mg/ml proteinase K. The bones were homogenized, and then lysis buffer was added. DNA was extracted from the resultant fluid by incubating in a 50°C incubator overnight, followed by addition of a phenol:chloroform mixture (1:1) to layer out the DNA from the organic phase. The supernatant of this solution was combined with 10 mM ammonium acetate, and the DNA was precipitated out with 7.3 ml of 100% ethanol per ml of sample, centrifuged, washed in 70% ethanol, and suspended in Tris:EDTA buffer. PCR was performed using 2 μl of template DNA, 20 pmol of each primer (synthesized by Bioneer Corp., Chungwon, Korea), 200 μM dNTPs, 1 mM MgCl2, and 1 U of Taq polymerase in a 50‐μl reaction volume containing 1× Taq polymerase buffer using a Perkin‐Elmer Gene Amp PCR System 2400. The sense and antisense primers, 5′‐CACATGAAGCAGCACGACTT‐3′ and 5′‐AGTTCACCTTGATGCCGTTC‐3′, were used to amplify GFP‐producing bands of 265 base pairs (bp). Statistical Methods Data were assessed using Statistical Package for the Social Sciences (SPSS) software (version 11.0) and summarized with descriptive statistics (mean and standard error). The differences between the four treatment groups at each time point were analyzed with the nonparametric Kruskal‐Wallis test. All tests were two‐sided, and a significance level of 5% was assigned. Results Establishment of Cells Stably Expressing RANK‐Fc To establish cells stably overexpressing RANK‐Fc, P2 to P5 MSCs were transduced with MSCV‐RANK‐Fc‐eGFP and directly evaluated for GFP expression by visualizing under a fluorescent microscope. Since transcription of both RANK‐Fc and eGFP genes is driven by the same promoter, MSCV LTR, GFP‐positive cells should signify RANK‐Fc expression. Using FACS analysis, we have confirmed that the transduction efficiencies were 94.1% and 92.5% in MSCs transduced with MSCV‐RANK‐Fc‐eGFP and MSCV‐eGFP, respectively. To confirm the cellular expression of RANK‐Fc, protein isolated from lysates of transduced MSCs were subjected to Western blot analysis using antibodies against mouse RANK or human IgG Fcγ. As shown in Figure 2A and 2B, cells transduced with the pMSCV‐RANK‐Fc‐eGFP vector overexpressed RANK‐Fc, compared with those infected with pMSCV‐eGFP. Moreover, to confirm the extracellular secretion of RANK‐Fc, we analyzed the culture medium of MSCs overexpressing RANK‐Fc by immunoprecipitation‐Western blot analysis. Immunoblots disclosed RANK‐Fc protein secretion in medium from cells transduced with MSCV‐RANK‐Fc‐eGFP, but no immunoreactivity was observed in conditioned medium from cells transduced with MSCV‐eGFP (Fig. 2C, 2D). Figure 2. Open in new tabDownload slide Cellular expression and secretion of RANK‐Fc in vitro. (A, B): Western blot analysis of cellular lysates from transduced mesenchymal stem cells (MSCs). Cell lysates from confluent cultures of MSCs transduced with either plasmid murine stem cell virus (pMSCV)‐eGFP or pMSCV‐RANK‐Fc‐eGFP were analyzed by SDS‐polyacrylamide gel electrophoresis (SDS‐PAGE) and Western blotting with goat polyclonal anti‐mouse RANK antibody (A) or goat polyclonal anti‐human IgG Fc fragment‐specific antibody (B). (C, D): Protein samples from conditioned medium were immunoprecipitated with goat polyclonal anti‐mouse RANK antibody and resolved by SDS‐PAGE. Blots were probed with goat polyclonal anti‐mouse RANK antibody (C) or goat polyclonal anti‐human IgG Fc fragment‐specific antibody (D). Abbreviations: GFP, green fluorescent protein; IP, immuno‐precipitation; mRANK, mouse receptor activator of nuclear factor‐κB; RANK, receptor activator of nuclear factor‐κB; WB, Western blot. Figure 2. Open in new tabDownload slide Cellular expression and secretion of RANK‐Fc in vitro. (A, B): Western blot analysis of cellular lysates from transduced mesenchymal stem cells (MSCs). Cell lysates from confluent cultures of MSCs transduced with either plasmid murine stem cell virus (pMSCV)‐eGFP or pMSCV‐RANK‐Fc‐eGFP were analyzed by SDS‐polyacrylamide gel electrophoresis (SDS‐PAGE) and Western blotting with goat polyclonal anti‐mouse RANK antibody (A) or goat polyclonal anti‐human IgG Fc fragment‐specific antibody (B). (C, D): Protein samples from conditioned medium were immunoprecipitated with goat polyclonal anti‐mouse RANK antibody and resolved by SDS‐PAGE. Blots were probed with goat polyclonal anti‐mouse RANK antibody (C) or goat polyclonal anti‐human IgG Fc fragment‐specific antibody (D). Abbreviations: GFP, green fluorescent protein; IP, immuno‐precipitation; mRANK, mouse receptor activator of nuclear factor‐κB; RANK, receptor activator of nuclear factor‐κB; WB, Western blot. RANK‐Fc Expression In Vivo Next, we investigated whether MSCs transduced with RANK‐Fc retrovirus continuously secrete biologically active RANK‐Fc in vivo after intraperitoneal injection. Blood was collected from retro‐orbital capillaries, and serum samples were prepared and analyzed for RANK‐Fc activity (Fig. 3). Serum RANK‐Fc levels measured by ELISA peaked at 2 weeks after the first injection (9.65 ± 2.66 ng/ml, mean ± SE), and gradually decreased to 0.60 ± 0.21 ng/ml at 8 weeks after the first injection. Serum RANK‐Fc levels were much lower (approximately 50‐fold) than those from conditioned medium from the cultured MSCs (data not shown). No RANK‐Fc was detected in control animals. Figure 3. Open in new tabDownload slide Production of RANK‐Fc from transplanted mesenchymal stem cells (MSCs) in vivo. Sera from the C57Bl/6 mice that underwent intraperitoneal injection of MSCs transduced with p‐murine stem cell virus‐RANK‐Fc‐enhanced green fluorescent protein (pMSCV‐RANK‐Fc‐eGFP) retrovirus were collected on days 5, 14, 28, and 56 after the first injection. The samples were subjected to enzyme‐linked immunosorbent assay analysis using anti‐Fc antibody. Data are expressed as means and standard errors. n = 7–8 per group. Abbreviation: RANK, receptor activator of nuclear factor‐κB. Figure 3. Open in new tabDownload slide Production of RANK‐Fc from transplanted mesenchymal stem cells (MSCs) in vivo. Sera from the C57Bl/6 mice that underwent intraperitoneal injection of MSCs transduced with p‐murine stem cell virus‐RANK‐Fc‐enhanced green fluorescent protein (pMSCV‐RANK‐Fc‐eGFP) retrovirus were collected on days 5, 14, 28, and 56 after the first injection. The samples were subjected to enzyme‐linked immunosorbent assay analysis using anti‐Fc antibody. Data are expressed as means and standard errors. n = 7–8 per group. Abbreviation: RANK, receptor activator of nuclear factor‐κB. RANK‐Fc Expression Attenuates Bone Loss in the OVX Model After ovariectomy, all OVX mice exhibited higher body weight and body fat mass (%) than sham‐operated animals, indicating that estrogen deficiency had successfully induced (Fig. 4A, 4B). As a surrogate marker for the functional capacity of the genetically modified cells in vivo, we determined the changes in BMD in the mice that were given MSCs. As shown in Figure 4C and 4D, all OVX mice (OVX‐PBS or OVX‐GFP group) gained significantly less BMD and BMC, from baseline compared with the sham‐operated animals (p < .01). Individual values (mean ± S.D.) of the body weight, body fat, BMD, and BMC at each time point are shown in Table 1. Figure 4. Open in new tabDownload slide Changes in the body weight, fat mass, BMD, and BMC in the animals. Seven‐week‐old female C57Bl/6 mice were sham‐operated or ovariectomized. OVX‐RANK‐Fc mice, n = 8; OVX‐GFP mice, n = 7; OVX‐PBS mice, n = 8). SHAM mice (n = 8) were used as a control. Body weight (g), body fat mass percent (%), BMD (g/cm2), and BMC (g) were measured at baseline and at 4 and 8 weeks after the operation using dual‐energy x‐ray absorptiometer (PIXimus; GE Healthcare). Shown are percent changes in body weight (A), body fat mass (B), BMD (C), and BMC (D) compared with the baseline. Data are expressed as means and standard errors. *, p < .01 versus SHAM; **, p < .05 versus SHAM; #, p < .01 versus OVX‐PBS. Abbreviations: BMC, bone mineral content; BMD, bone mineral density; OVX‐GFP, ovariecetomized and intraperitoneally injected with control green fluorescent protein virus; OVX‐PBS, ovariecetomized and intraperitoneally injected with phosphate‐buffered saline; OVX‐RANK‐Fc, ovariecetomized and intraperitoneally injected with mesenchymal stem cells transduced with receptor activator of nuclear factor‐κB‐Fc retrovirus; RANK‐Fc, receptor activator of nuclear factor‐κB‐Fc; SHAM, sham‐operated. Figure 4. Open in new tabDownload slide Changes in the body weight, fat mass, BMD, and BMC in the animals. Seven‐week‐old female C57Bl/6 mice were sham‐operated or ovariectomized. OVX‐RANK‐Fc mice, n = 8; OVX‐GFP mice, n = 7; OVX‐PBS mice, n = 8). SHAM mice (n = 8) were used as a control. Body weight (g), body fat mass percent (%), BMD (g/cm2), and BMC (g) were measured at baseline and at 4 and 8 weeks after the operation using dual‐energy x‐ray absorptiometer (PIXimus; GE Healthcare). Shown are percent changes in body weight (A), body fat mass (B), BMD (C), and BMC (D) compared with the baseline. Data are expressed as means and standard errors. *, p < .01 versus SHAM; **, p < .05 versus SHAM; #, p < .01 versus OVX‐PBS. Abbreviations: BMC, bone mineral content; BMD, bone mineral density; OVX‐GFP, ovariecetomized and intraperitoneally injected with control green fluorescent protein virus; OVX‐PBS, ovariecetomized and intraperitoneally injected with phosphate‐buffered saline; OVX‐RANK‐Fc, ovariecetomized and intraperitoneally injected with mesenchymal stem cells transduced with receptor activator of nuclear factor‐κB‐Fc retrovirus; RANK‐Fc, receptor activator of nuclear factor‐κB‐Fc; SHAM, sham‐operated. Table 1. Body weight, BMC, BMD, and body fat changes of the animals Open in new tabDownload slide Open in new tabDownload slide Open in new tabDownload slide Open in new tabDownload slide Open in new tab Table 1. Body weight, BMC, BMD, and body fat changes of the animals Open in new tabDownload slide Open in new tabDownload slide Open in new tabDownload slide Open in new tabDownload slide Open in new tab It is noteworthy that all the mice in this experiment actually gained BMD because they were still in the growing phase of development [18]. However, the percentage increase in BMD in mice transplanted with MSCs overexpressing RANK‐Fc was significantly greater than that of the OVX‐PBS group (p < .05). The change in BMC in the animals showed a similar pattern. These results suggest that the genetically modified cell therapy could prevent the ovariectomy‐induced reduction in BMD in C57Bl/6 mice. Detection of GFP Expression by PCR We next tested whether the bone‐protective effect of RANK‐Fc was a result of engraftment of RANK‐Fc producing cells into bone. To address this, PCR analysis for GFP was performed using genomic DNA isolated from various tissues or bone marrow aspirates. As shown in Figure 5A, expected GFP DNA fragment (∼265 bp) was weakly but clearly amplified from femur and intra‐abdominal fat of OVX‐RANK‐Fc mice, whereas tissues from OVX+PBS group did not reveal any positive signal (Fig. 5B). DNA isolated from liver, brain, heart, lung, or bone marrow aspirates did not give a positive band in either of the two groups. Essentially the same pattern of GFP expression (i.e., only from femur and fat tissue) was also observed from OVX+GFP mice (data not shown). These results suggest that some degree of engraftment to bone has occurred after systemic injection of retrovirus‐transduced MSCs, although a comparable amount of expression was also observed in fat tissue. Figure 5. Open in new tabDownload slide Localization of RANK‐Fc‐secreting cells in tissue sections of transduced mesenchymal stem cell transplants. Tissues and bone marrow aspirates from the C57Bl/6 mice that underwent intraperitoneal injection of the transduced mesenchymal stem cells or PBS were harvested 56 days after the first injection and prepared for the analysis of green fluorescent protein (GFP) by polymerase chain reaction (PCR). DNA PCR using oligonucleotide primers for the GFP gene was performed using in genomic DNA isolated from various tissues as template. Lane 1, liver; lane 2, brain; lane 3, heart; lane 4, lung; lane 5, intra‐abdominal fat; lane 6, femur; lane 7; bone marrow aspirates; lane 8, positive control. Abbreviations: M, 100‐base pair DNA ladder; OVX, ovariectomized; PBS, phosphate‐buffered saline; RANK‐Fc, receptor activator of nuclear factor‐κB‐Fc. Figure 5. Open in new tabDownload slide Localization of RANK‐Fc‐secreting cells in tissue sections of transduced mesenchymal stem cell transplants. Tissues and bone marrow aspirates from the C57Bl/6 mice that underwent intraperitoneal injection of the transduced mesenchymal stem cells or PBS were harvested 56 days after the first injection and prepared for the analysis of green fluorescent protein (GFP) by polymerase chain reaction (PCR). DNA PCR using oligonucleotide primers for the GFP gene was performed using in genomic DNA isolated from various tissues as template. Lane 1, liver; lane 2, brain; lane 3, heart; lane 4, lung; lane 5, intra‐abdominal fat; lane 6, femur; lane 7; bone marrow aspirates; lane 8, positive control. Abbreviations: M, 100‐base pair DNA ladder; OVX, ovariectomized; PBS, phosphate‐buffered saline; RANK‐Fc, receptor activator of nuclear factor‐κB‐Fc. Discussion The objective of this study was to examine the feasibility of using genetically modified MSCs as a platform for sustained systemic delivery of therapeutic proteins into the circulation in an osteoporosis model. Our data show that MSCs are effectively transduced with MSCV‐based retrovirus and subsequently are capable of secreting biologically active RANK‐Fc in vitro and in vivo. Moreover, RANK blockade by RANK‐Fc is an effective method to prevent ovariectomy‐induced bone loss. MSCs attract considerable attention in efforts to develop cell and gene therapies [2, 22, 23] since they are readily obtained from the patient, thus avoiding any immune responses. Additionally, extensive experiments over several decades disclose no evidence of tumorgenicity that is prominently observed with embryonic stem cells [24]. Promising results have been reported on the use of MSCs or closely related cells from bone marrow in animal models for a number of diseases, including osteogenesis imperfecta [22], Parkinsonism [25], spinal cord injury [26], stroke [27], myelin deficiency [28], cardiac disorders [29], and lung diseases [30]. Our data demonstrate that mesenchymal cell‐based gene therapy with RANK‐Fc has clearly protected bone loss associated with ovariectomy in C57Bl/6 mice. Effects on bone density are small, however, possibly due to the rapid skeletal growth of the mice that renders the evaluation of a “bone‐protective” effect difficult in this rodent OVX model. In our study, the circulating level of RANK‐Fc peaked at 1 week after the last injection, consistent with previous data [31], and decreased gradually over 2 months. The durability of expression of the desired gene is a major limitation of retroviral transduction of MSCs. No report has extended beyond 4 months post‐transplantation of transduced cells [3, 31]. In most instances, expression decreases with time [32] due to promoter inactivation [33] and/or loss of transduced cells. Although these results are promising, they highlight the need for careful consideration of ex vivo methods, choice of promoter to direct the desired biological activity, and assessment of the self‐maintenance of transduced MSCs upon in vivo transplantation. It remains a matter of controversy whether donor‐derived MSCs are capable of homing and engraftment into the marrow cavity following transplantation. A number of reports indicate that marrow‐derived cells engraft as differentiated cells into multiple tissues following infusion into experimental animals [2, 22, 23], whereas others suggest that MSCs do not engraft in marrow after systemic infusion [34, 35]. In our experiments, GFP signals were observed in bone and intra‐abdominal fat, but not from any other tissues, including bone marrow aspirates by PCR analysis. The presence of GFP signal from intra‐abdominal fat seems to be a consequence of intraperitoneal injection of cells resulting in initial accumulation of a large number of cells locally. Our finding of positive PCR bands from femur but not from freshly isolated bone marrow cells is intriguing. One can speculate that the injected cells engrafted, differentiated, or moved, and then resided within the bone tissue, such as bony lacunae or lining cells. However, we have not traced the migration of cells directly, nor were we able to identify the cell types that express GFP using sensitive assays, such as fluorescent in situ hybridization. Therefore, the positive PCR results from femur do not definitely prove the engraftment of donor cells. In addition, the cells we used as a vehicle were not isolated by specific selection method using cell surface markers, and introducing retrovirus by multiple rounds of transduction may have resulted in alteration of stem cell characteristics, including the capacity for homing. Indeed, in our study, the therapeutic effects of RANK‐Fc could have been obtained regardless of the location of the transplanted cells once they are connected to the systemic circulation. To confirm whether the systemically administered cells have engrafted, in situ immunological analysis will be required in our model. MSCs are easily isolated from iliac crest bone marrow aspirates. The systemic infusion of autologous MSCs appears to be well tolerated in a Phase I clinical trial [36]. Additionally, several reports suggest that engraftment of whole marrow or MSCs is possible in mice or dogs without the need for marrow ablation, provided that cells are infused in large numbers or at regularly spaced intervals [37, 38]. Accordingly, it may be possible to use gene‐engineered MSCs from patients for therapy of common diseases, such as osteoporosis, in which marrow ablation cannot be justified. However, notwithstanding the potential benefits of MSC as a cellular vehicle to convey therapeutic gene product, strategies to overcome the risks imposed by retroviral transduction, that is, uncontrolled proliferation and even tumor formation, which can occur depending on the integration of the virus, should be developed before any clinical application could be tried. The main limitation of our study is that by using mice in a growing phase, it is difficult to directly extrapolate our results to postmenopausal or involutional model of osteoporosis. Indeed, the role of the RANK/OPG system in the growing mice has not been well established. Moreover, since BMD measurement using DEXA is significantly affected by the size of the bones, interpreting the changes in BMD in animals of different size is very complicated [39, 40]. Nonetheless, our results are consistent with the study by Hsu et al. who demonstrated in vivo bone protective effects of RANK‐Fc in immature growing mice (aged 3–4 weeks) [7]. In summary, genetically modified MSCs are capable of sustaining to secrete RANK‐Fc up to 8 weeks after intraperitoneal injection, and RANK blockade by RANK‐Fc may be an effective method to prevent ovariectomy‐induced bone loss. These data support the use of bone marrow‐derived MSCs as cellular vehicles for gene therapy and pave the way for a future therapeutic approach for osteoporosis using genetically modified MSCs. Acknowledgements This work was supported by grant 06‐2005‐053‐0 from Seoul National University Hospital. D.K. is currently affiliated with the Department of Internal Medicine, Dankook University College of Medicine, Cheonan, Korea. Disclosures The authors indicate no potential conflicts of interest. References 1 NIH Consensus Development Panel on Osteoporosis Prevention, Diagnosis, and Therapy . Osteoporosis prevention, diagnosis, and therapy . JAMA 2001 ; 285 : 785 – 795 . Crossref Search ADS PubMed WorldCat 2 Prockop DJ. Marrow stromal cells as stem cells for nonhematopoietic tissues . Science 1997 ; 276 : 71 – 74 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Bianco P , Riminucci M, Gronthos S et al. Bone marrow stromal stem cells: Nature, biology, and potential applications . Stem Cells 2001 ; 19 : 180 – 192 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Bartholomew A , Patil S, Mackay A et al. Baboon mesenchymal stem cells can be genetically modified to secrete human erythropoietin in vivo . Hum Gene Ther 2001 ; 12 : 1527 – 1541 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Hofbauer LC , Heufelder AE. Role of receptor activator of nuclear factor‐kappaB ligand and osteoprotegerin in bone cell biology . J Mol Med 2001 ; 79 : 243 – 253 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Simonet WS , Lacey DL, Dunstan CR et al. Osteoprotegerin: A novel secreted protein involved in the regulation of bone density . Cell 1997 ; 89 : 309 – 319 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Hsu H , Lacey DL, Dunstan CR et al. Tumor necrosis factor receptor family member RANK mediates osteoclast differentiation and activation induced by osteoprotegerin ligand . Proc Natl Acad Sci U S A 1999 ; 96 : 3540 – 3545 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Lacey DL , Timms E, Tan HL et al. Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation . Cell 1998 ; 93 : 165 – 176 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Bucay N , Sarosi I, Dunstan CR et al. Osteoprotegerin‐deficient mice develop early onset osteoporosis and arterial calcification . Genes Dev 1998 ; 12 : 1260 – 1268 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Kong YY , Yoshida H, Sarosi I et al. OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph‐node organogenesis . Nature 1999 ; 397 : 315 – 323 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Dougall WC , Glaccum M, Charrier K et al. RANK is essential for osteoclast and lymph node development . Genes Dev 1999 ; 13 : 2412 – 2424 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Nakashima T , Wada T, Penninger JM. RANKL and RANK as novel therapeutic targets for arthritis . Curr Opin Rheumatol 2003 ; 15 : 280 – 287 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Honore P , Luger NM, Sabino MA et al. Osteoprotegerin blocks bone cancer‐induced skeletal destruction, skeletal pain and pain‐related neurochemical reorganization of the spinal cord . Nat Med 2000 ; 6 : 521 – 528 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Bekker PJ , Holloway D, Nakanishi A et al. The effect of a single dose of osteoprotegerin in postmenopausal women . J Bone Miner Res 2001 ; 16 : 348 – 360 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Body JJ , Greipp P, Coleman RE et al. A phase I study of AMGN‐0007, a recombinant osteoprotegerin construct, in patients with multiple myeloma or breast carcinoma related bone metastases . Cancer 2003 ; 97 : 887 – 892 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Sordillo EM. Pearse RN RANK‐Fc: A therapeutic antagonist for RANK‐L in myeloma . Cancer 2003 ; 97 : 802 – 812 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Pearse RN , Sordillo EM, Yaccoby S et al. Multiple myeloma disrupts the TRANCE/osteoprotegerin cytokine axis to trigger bone destruction and promote tumor progression . Proc Natl Acad Sci U S A 2001 ; 98 : 11581 – 11586 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Bolon B , Carter C, Daris M et al. Adenoviral delivery of osteoprotegerin ameliorates bone resorption in a mouse ovariectomy model of osteoporosis . Mol Ther 2001 ; 3 : 197 – 205 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Bailey EC , Milenkovic L, Scott MP et al. Several PATCHED1 missense mutations display activity in patched1‐deficient fibroblasts . J Biol Chem 2002 ; 277 : 33632 – 33640 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Jeon MJ , Kim JA, Kwon SH et al. Activation of peroxisome proliferator‐activated receptor‐gamma inhibits the Runx2‐mediated transcription of osteocalcin in osteoblasts . J Biol Chem 2003 ; 278 : 23270 – 23277 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Bellino FL. Nonprimate animal models of menopause: Workshop report . Menopause 2000 ; 7 : 14 – 24 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Pereira RF , O'Hara MD, Laptev AV et al. Marrow stromal cells as a source of progenitor cells for nonhematopoietic tissues in transgenic mice with a phenotype of osteogenesis imperfecta . Proc Natl Acad Sci U S A 1998 ; 95 : 1142 – 1147 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Azizi SA , Stokes D, Augelli BJ et al. Engraftment and migration of human bone marrow stromal cells implanted in the brains of albino rats‐similarities to astrocyte grafts . Proc Natl Acad Sci U S A 1998 ; 95 : 3908 – 3913 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Prockop DJ , Gregory CA, Spees JL. One strategy for cell and gene therapy: Harnessing the power of adult stem cells to repair tissues . Proc Natl Acad Sci U S A 2003 ; 100 (suppl) : 11917 – 11923 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 25 Schwarz EJ , Alexander GM, Prockop DJ et al. Multipotential marrow stromal cells transduced to produce L‐DOPA: Engraftment in a rat model of Parkinson disease . Hum Gene Ther 1999 ; 10 : 2539 – 2549 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Hofstetter CP , Schwarz EJ, Hess D et al. Marrow stromal cells form guiding strands in the injured spinal cord and promote recovery . Proc Natl Acad Sci U S A 2002 ; 99 : 2199 – 2204 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Li Y , Chen J, Chen XG et al. Human marrow stromal cell therapy for stroke in rat: Neurotrophins and functional recovery . Neurology 2002 ; 59 : 514 – 523 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Jin HK , Carter JE, Huntley GW et al. Intracerebral transplantation of mesenchymal stem cells into acid sphingomyelinase‐deficient mice delays the onset of neurological abnormalities and extends their life span . J Clin Invest 2002 ; 109 : 1183 – 1191 . Google Scholar Crossref Search ADS PubMed WorldCat 29 Toma C , Pittenger MF, Cahill KS et al. Human mesenchymal stem cells differentiate to a cardiomyocyte phenotype in the adult murine heart . Circulation 2002 ; 105 : 93 – 98 . Google Scholar Crossref Search ADS PubMed WorldCat 30 Ortiz LA , Gambelli F, McBride C et al. Mesenchymal stem cell engraftment in lung is enhanced in response to bleomycin exposure and ameliorates its fibrotic effects . Proc Natl Acad Sci U S A 2003 ; 100 : 8407 – 8411 . Google Scholar Crossref Search ADS PubMed WorldCat 31 Krebsbach PH , Zhang K, Malik AK et al. Bone marrow stromal cells as a genetic platform for systemic delivery of therapeutic proteins in vivo: Human factor IX model . J Gene Med 2003 ; 5 : 11 – 17 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Ding LM , Saylors R, Munshi NC. The stromal cell as a vehicle for ex vivo gene transfer . Blood 1997 ; 89 : 446 – 456 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 33 Chuah MK , Van Damme A, Zwinnen H et al. Long‐term persistence of human bone marrow stromal cells transduced with factor VIII‐retroviral vectors and transient production of therapeutic levels of human factor VIII in nonmyeloablated immunodeficient mice . Hum Gene Ther 2000 ; 11 : 729 – 738 . Google Scholar Crossref Search ADS PubMed WorldCat 34 Murry CE , Soonpaa MH, Reinecke H et al. Haematopoietic stem cells do not transdifferentiate into cardiac myocytes in myocardial infarcts . Nature 2004 ; 428 : 664 – 668 . Google Scholar Crossref Search ADS PubMed WorldCat 35 Balsam LB , Wagers AJ, Christensen JL et al. Haematopoietic stem cells adopt mature haematopoietic fates in ischaemic myocardium . Nature 2004 ; 428 : 668 – 673 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Lazarus HM , Haynesworth SE, Gerson SL et al. Ex vivo expansion and subsequent infusion of human bone marrow‐derived stromal progenitor cells (mesenchymal progenitor cells): Implications for therapeutic use . Bone Marrow Transplant 1995 ; 16 : 557 – 564 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 37 Stewart FM , Crittenden RB, Lowry PA et al. Long‐term engraftment of normal and post‐5‐fluorouracil murine marrow into normal nonmyeloablated mice . Blood 1993 ; 81 : 2566 – 2571 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Bienzle D , Abrams‐Ogg AC, Kruth SA et al. Gene transfer into hematopoietic stem cells: Long‐term maintenance of in vitro activated progenitors without marrow ablation . Proc Natl Acad Sci U S A 1994 ; 91 : 350 – 354 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Dickson GR , Luczak M, Wlodarski KH. The limitation of DEXA analysis for bone mass determination in mice . Folia Biol (Krakow) 2004 ; 52 : 125 – 129 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 40 Holdsworth DW , Thornton MM, Drost D et al. Rapid small‐animal dual‐energy X‐ray absorptiometry using digital radiography . J Bone Miner Res 2000 ; 15 : 2451 – 2457 . Google Scholar Crossref Search ADS PubMed WorldCat Copyright © 2006 AlphaMed Press This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Retrovirus‐Mediated Gene Transfer of Receptor Activator of Nuclear Factor‐κB‐Fc Prevents Bone Loss in Ovariectomized Mice
JF - Stem Cells
DO - 10.1634/stemcells.2005-0480
DA - 2006-07-01
UR - https://www.deepdyve.com/lp/oxford-university-press/retrovirus-mediated-gene-transfer-of-receptor-activator-of-nuclear-DbwfGU3tfh
SP - 1798
EP - 1805
VL - 24
IS - 7
DP - DeepDyve
ER -