SEARCH

SEARCH BY CITATION

Keywords:

  • IMMUNE SYSTEM;
  • ESTROGEN DEFICIENCY;
  • PROINFLAMMATORY CYTOKINES;
  • T CELLS;
  • B LYMPHOPOESIS

ABSTRACT

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

Activated T cell has a key role in the interaction between bone and immune system. T cells produce proinflammatory cytokines, including receptor activator of NF-κB ligand (RANKL), tumor necrosis factor α (TNF-α), and interleukin 17 (IL-17), all of which augment osteoclastogenesis. RANKL and TNF-α are targeted by inhibitors such as denosumab, a human monoclonal RANKL antibody, and infliximab, which neutralizes TNF-α. IL-17 is also an important mediator of bone loss, and an antibody against IL-17 is undergoing phase II clinical trial for rheumatoid arthritis. Although there are a few studies showing suppression of Th17 cell differentiation and induction of regulatory T cells (Tregs) by infliximab, the effect of denosumab remains poorly understood. In this study, we investigated the effects of anti-TNF-α, anti-RANKL, or anti-IL-17 antibody administration to estrogen-deficient mice on CD4+ T-cell proliferation, CD28 loss, Th17/Treg balance and B lymphopoesis, and finally, the translation of these immunomodulatory effects on skeletal parameters. Adult Balb/c mice were treated with anti-RANKL/-TNF-α/-IL-17 subcutaneously, twice a week, postovariectomy (Ovx) for 4 weeks. Animals were then autopsied; bone marrow cells were collected for FACS and RNA analysis and serum collected for ELISA. Bones were dissected for static and dynamic histomorphometry studies. We observed that although anti-RANKL and anti-TNF-α therapies had no effect on Ovx-induced CD4+ T-cell proliferation and B lymphopoesis, anti-IL-17 effectively suppressed both events with concomitant reversal of CD28 loss. Anti-IL-17 antibody reduced proinflammatory cytokine production and induced Tregs. All three antibodies restored trabecular microarchitecture with comparable efficacy; however, cortical bone parameters, bone biomechanical properties, and histomorphometry were best preserved by anti-IL-17 antibody, likely attributable to its inhibitory effect on osteoblast apoptosis and increased number of bone lining cells and Wnt10b expression. Based on the superior immunoprotective effects of anti-IL-17, which appears to translate to a better skeletal preservation, we propose beginning clinical trials using a humanized antibody against IL-17 for treatment of postmenopausal osteoporosis. © 2014 American Society for Bone and Mineral Research.


Introduction

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

Osteoporosis is characterized by a decrease in bone strength, leading to an increased risk of fractures.[1, 2] It is caused by an uncoupling of bone formation from bone resorption in the remodeling process so that osteoclast activity exceeds that of the osteoblast activity.[3-6] With increasing age, this imbalance in bone remodeling could happen in both men and women; however, in postmenopausal women, this process is accelerated.[7] Although there are several antiresorptive options available for the treatment of osteoporosis, parathyroid hormone (PTH) is the sole anabolic therapy. Antiresorptive drugs include bisphosphonates, raloxifene (a selective estrogen receptor modulator), and denosumab (a humanized monoclonal antibody against receptor activator of NF-κB ligand [RANKL]).[8]

Recent studies have established that bone and immune cells are functionally connected because they share the same progenitors and are influenced by the same cytokines.[1, 9] Immune cells, especially the activated T cells, have a major role in the interplay between bone and immune system.[10, 11] These cells produce proinflammatory cytokines, including RANKL, tumor necrosis factor α (TNF-α), and interkeukin 17 (IL-17). All have strong osteoclast-promoting and bone-resorbing effects, but RANKL is considered to have the most potent resorptive effect of all the osteoclastogenic cytokines.[10, 12-14]

Systemic levels of RANKL, TNF-α, and IL-17 are enhanced after ovariectomy (Ovx), and their inhibition/functional block is likely to afford effective skeletal protection post-Ovx.[15, 16] Indeed, neutralizing RANKL action by denosumab has been approved by the FDA for postmenopausal women who are at high risk of osteoporotic fractures and are nonresponsive to other antiresorptive therapies.[17] Studies have established its efficacy in reducing fracture rates and increasing bone mineral density (BMD) in postmenopausal women, but side effects include risk of osteonecrosis of jaw and are contraindicated in patients with hypocalcaemia.[18, 19] There are no studies investigating the effect of blocking RANKL function, eg, by denosumab on effector T and B cells. We and others have shown that T and B lymphocytes are key mediators of Ovx-induced bone loss.[11, 14] The other important proinflammatory cytokine is TNF-α, a key mediator of rheumatoid arthritis (RA)-induced bone loss, and its neutralization by an antibody, infliximab, is recommended for the treatment of this disease.[20-22] Recent studies suggest that infliximab may be effective in the treatment of osteoporosis, particularly in patients with poor tolerance to bisphosphonates, the mainstay of antiresorptive therapy.[21] The major side effect associated with infliximab is increased risk of contracting mycobacterial infection.[23-25] There are reports showing that infliximab suppresses Th17 cell differentiation and induces T regulatory cells (Tregs) in patients with uveitis,[26] ulcerative colitis,[27] and RA.[28] However, there are no reports on the effect of infliximab on immune-modulation in osteopenic condition, which is characterized by inappropriate activation of T and B cells that are key mediators of the disease.

RANKL and IL-17 are key mediators of other diseases affecting the bone.[29] IL-17-deficient mice are resistant to collagen-induced arthritis (CIA), and blocking IL-17 in a mouse CIA model reduces disease symptoms.[30, 31] Phase II trial with an antibody against IL-17 (AIN457) in RA patients is presently under way.[32] We have shown that blocking IL-17 prevented bone loss in estrogen deficiency-induced osteopenia and increased differentiation of Th17 cells.[13]

Although denosumab has been approved for treatment of osteoporosis in antiresorptive mode, there are few studies assessing the effects of anti-RANKL on the skeleton of Ovx rodents. Because denosumab is not cross-reactive with rodent RANKL, its evaluation in preclinical studies was performed mainly in cynomolgus monkeys or human RANKL knock in mice.[33, 34] Previously, studies have been carried out that show that RANKL inhibition by osteoprotegerin leads to improved bone strength in Ovx rats.[35] A study by Furuya and colleagues has shown that single subcutaneous injection of OYC1, an anti-mouse RANKL neutralizing monoclonal antibody, to normal mice significantly augments bone mineral density.[36] RANKL is considered an acute and specific activator of osteoclastogenesis, resulting in rapid bone loss.[37] There is a report describing that a robust osteoclastogenic effect of soluble RANKL was suppressed by anti-RANKL in ovary-intact mice, suggesting that it has potent osteoclast-specific action.[38] However, the effect of anti-RANKL on other immune cell types, such as Th17 and Tregs, has not been investigated despite the fact that one of the major contributors of RANKL is T cells. In case of anti-TNF-α therapy, there is a report that shows that functional block of TNF binding protein prevents Ovx-induced bone loss.[15] A recent report by Kim and colleagues has shown that TNF-α blocker prevents the increase of sclerostin, a wnt antagonist implicated in the pathogenesis of postmenopausal osteoporosis, in bone of ovariectomized mice.[39] However, the effect of blocking/neutralizing RANKL and TNF-α on T cell-mediated bone wasting induced by estrogen deficiency has not been investigated. Here, we made a comparative study on the effects of neutralizing TNF-α, RANKL, and IL-17 by monoclonal antibodies on CD4+ T-cell proliferation, Th17/Treg balance, and B lymphopoesis in Ovx mice. We have also determined the effect of these therapies on Ovx-induced loss of CD28, a membrane glycoprotein that provides the requisite costimulatory signal to T cells. CD28 is also an established immunosenescence biomarker.[12] Finally, we have investigated how these immunomodulatory effects are translated to skeletal preservation in Ovx mice.

Materials and Methods

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

Flow cytometry

Cells from the bone marrow (BM) were labeled with anti-CD3, -CD4, -CD28, and -B220 antibodies (APC-conjugated anti-mouse CD3, PE-conjugated anti-mouse CD4, FITC-conjugated anti-mouse CD28, and APC-conjugated anti-B220 antibodies) to assess the percentage of CD4+, CD4+CD28+ in CD3+ cells, and B220+ (CD45RO) cells as per previously published protocol.[12, 14] Specificity of immunostaining was ascertained by the background fluorescence of cells incubated with Ig isotype controls. Fluorescence data from at least 10,000 cells were collected from each sample. Immunostaining was done as per manufacturer's instructions. In brief, single-cell suspension of the BM was prepared in phospate-buffered saline (PBS). Cells were counted using a hemocytometer and were resuspended in FACS buffer as 106 cells/500 µL PBS and antibody was added as 10 µL/106 cells and further incubated for 45 minutes at room temperature. After incubation, cells were washed twice with PBS and transferred to FACS tubes for analysis. FACS Caliber and FACS Arya (BD Biosciences, Mississauga, Canada) were used to quantify the percentage of CD4+, CD4+CD28+ T cells in CD3+ cells, and B220+ B cells in all the groups.

Cytometric bead array flex

Levels of IL-6, IFN-γ, TNF-α, IL-17A, IL-12, and IL-10 were assessed in serum samples by fluorescent bead-based technology using cytometric bead array (CBA) Flex sets according to manufacturer's instructions (BD Biosciences). Fluorescent signals were read and analyzed on a FACS caliber flow cytometer (BD Biosciences) with the help of BD FCAP Array v1.0.1 software (BD Biosciences).

Total RNA isolation and quantitative real-time-PCR

Total RNA was extracted from isolated CD4+ T cells and B220+ cells of all the in vivo groups using Trizol (Invitrogen, Carlsbad, CA, USA). cDNA was synthesized from 1 µg total RNA with the RevertAid H Minus First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Pittsburgh, PA, USA). SYBR green chemistry was used for quantitative determination of the mRNAs for Stat 3, Stat 5, Wnt 10b, SOST, ROR-α, RORγt, Foxp3, RANKL, OPG, and a housekeeping gene, GAPDH, following an optimized protocol.

The design of sense and antisense oligonucleotide primers was based on published cDNA sequences using the Universal probe library (Roche Diagnostics, Indianapolis, IN, USA). Primer sequences are given in Supplemental Table S1. For real-time PCR, the cDNA was amplified with Light Cycler 480 (Roche Diagnostics). The double-stranded DNA-specific dye SYBR Green I was incorporated into the PCR buffer provided in the Light Cycler 480 SYBER green I master (Roche Diagnostics) to allow for quantitative detection of the PCR product in a 20-µL reaction volume. The temperature profile of the reaction was 95°C for 5 minutes, 40 cycles of denaturation at 94°C for 2 minutes, annealing and extension at 62°C for 30 seconds, and extension at 72°C for 30 seconds. GAPDH was used to normalize differences in RNA isolation, RNA degradation, and the efficiencies of the reverse transcription.

In vivo studies

The study was conducted in accordance with current legislation on animal experiments and was approved by the Institutional Animal Ethics Committee, Central Drug Research Institute (CPSCEA registration no. 34/1999, dated November 3, 1999, extended to 2012; approval reference no. IAEC/2010/127/renew 01, dated September 2, 2012). Adult Balb/c mice (9 to 10 weeks old) were used for the studies.[40-43] All mice were housed at 25°C, in 12-hour light/dark cycles. Normal chow diet and water were provided ad libitum. For the experiments, 10 mice per group were taken. The groups were sham-operated (ovary intact) mice; ovariectomized (Ovx); Ovx + 1.0 mg/kg anti-TNF-α antibody;[15] Ovx + 300 µg/kg anti-RANKL antibody;[44] and Ovx + 100 ng/kg anti-IL-17 antibody twice/week sc.[13] The sham-operated group served as the positive control and was given vehicle (IgG in normal saline). All treatments were continued for 4 weeks. At the completion of study, animals were autopsied. After autopsy, bones were dissected and the BM was flushed out. Total lymphocytes from the BM were isolated by using HiSep LSM 1084 (HiMedia, Mumbai, India) by means of density (1.08460 ± 0010 g/mL) gradient centrifugation technique.[12, 14, 45, 46] Long bones were kept in 70% isopropanol for micro-computed tomography (µCT) study. Pure CD4+ cells were retrieved from the BM by positive selection using microbead-based isolation by MACS separator according to the manufacturer's protocol (EasySep Biotin Selection Kit, Stem Cell Technologies Inc., Vancouver, Canada). These purified cells were then collected in Trizol for real-time PCR (qPCR). Serum was collected for ELISA.

Micro-computed tomographic (μCT) determination of excised bones was carried out using the Sky Scan 1076 CT scanner (Aartselaar, Belgium) using previously published protocol.[13] To analyze trabecular region, region of interest (ROI) was drawn at a total of 100 slices in the region of secondary spongiosa situated 1.5 mm from the distal border of growth plate, excluding all primary spongiosa and cortical bone. For cortical bone analysis, 350 serial image slides were discarded from growth plate to exclude the trabecular region, and 100 consecutive image slides were selected and quantification was done using CTAn software. Various trabecular parameters (3D) and cortical parameters (2D) were analyzed by following previously published protocols.[47] Histomorphometric analyses were conducted using Bioquant Image Analysis software (Bioquant, Nashville, TN, USA). Tartrate-resistant acid phosphatase (TRAP) staining of osteoclasts was performed using a leukocyte acid phosphatase staining kit (Sigma, St. Louis, MO, USA). Bone mechanical strength was examined by 3-point bending strength of femur mid-diaphysis using Bone Strength Tester Model TK 252C as reported earlier.[48] The load-displacement curves generated were used to calculate the ultimate load (N) and energy to failure (mJ).

Measurement of bone-relevant serum parameters

Serum levels of CTx and type I collagen N-terminal propeptide (PINP) were determined by ELISA kits purchased from Immunodiagnostic Systems Inc. (Scottsdale, AZ, USA) by following the manufacturer's protocols.

Osteoblast apoptosis

Isolated femur bones (n = 3) from each group were fixed in 4% paraformaldehyde and decalcified in 20% ethylene diamine tetraacetic acid (EDTA) for 2 weeks, then embedded in paraffin. Transverse sections (5 µm) were cut. After deparaffinization and hydration, sections were incubated in a buffer containing PBS with 1% bovine serum albumin (BSA), 0.1% rat serum, and 0.1% Triton X-100 (all v/v) for 30 minutes at room temperature to avoid the nonspecific binding. Sections were washed with PBS, and apoptotic cells were detected with deoxynucleotidyl transferase-mediated nick end labeling assay (Tunel assay according to Roche manual kit). Sections were then washed with PBS and incubated with mouse anti-Runx-2 antibody (1:1600) dilution in PBS containing 0.5% BSA (Abcam, Cambridge, UK) overnight at 4°C in a humidified chamber. Sections were washed with PBS and incubated with fluorescent Alexa Fluor-488 goat anti-mouse IgG (H + L) (1:1600 dilution in PBS) (Molecular Probes, Eugene, OR, USA) for 1 hour at room temperature. Sections were also stained with DAPI for 15 minutes in dark. After 15 minutes' incubation, slides were rinsed with PBS and mounted with Vectashield (Vector Laboratories, Burlingame, CA, USA). Sections were viewed under fluorescence microscope. Tunel as well as RUNX-2-positive cells were counted in five randomly selected fields from three bone sections of each group and quantified with Image-Pro Plus 6.1 software (Media Cybernetics, Rockville, MD, USA).

Determination of the bone lining cells

Deparaffinized and hydrated femoral epiphysis sections of different groups were immersed in Mayer's hematoxylin for 1 minute, washed with running tap water for 15 minutes, and incubated in 1% eosin aqueous solution for 1 to 2 minutes at room temperature. After this, the sections were dehydrated with ascending concentrations of ethanol solutions, cleared with xylene, and mounted with a cover slip using DPX (Sigma) mounting medium. Lining cells on the periosteal bone surface were calculated with Image-Pro Plus 6.1 software.

Statistical analysis

Data are expressed as mean ± SD. The data obtained in experiments were subjected to one-way ANOVA followed by Newman–Keuls test of significance using Prism version 3.0 software. Qualitative observations have been represented following assessments made by three individuals blinded to the experimental designs.

Results

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

Effects on proliferation of effector T cells, CD4+CD28+ T cells, and B lymphopoesis

Ovx is known to increase the proliferation of proinflammatory T helper cells including CD4+ and CD8+ T cells.[11] Mice Ovx for 4 weeks have higher frequency of CD4+ T cells in BM compared with the sham group (Fig. 1A). Treatment of Ovx mice with either anti-RANKL or anti-TNF-α antibody failed to change the Ovx-induced increase in CD4+ T cells. However, the percentage of CD4+ T cells in the BM of Ovx mice treated with anti-IL-17 antibody was comparable to the sham group (Fig. 1A).

image

Figure 1. Effect of anti-TNF-α/-RANKL/-IL-17 on Ovx-induced increases in T cells, B cells, and OPG/RANKL ratio in B220+ cells of BM. (A) The CD4+ T-cell population was increased in the BM of Ovx mice compared with the sham and anti-IL-17, but the other two neutralizing antibodies did not reverse the Ovx-induced effect. (B) Compared with the sham, CD4+CD28+ T cells in the BM were diminished in the Ovx mice, anti-IL-17 and anti-TNF-α, but anti-RANKL did not reverse the Ovx-induced effect. The reversal effect of anti-IL-17 was greater than anti-TNF-α. (C) The population of B220+ cells was increased in BM of Ovx mice compared with the sham and anti-IL-17, but the other two neutralizing antibodies did not reverse the Ovx-induced increase. (D) The OPG/RANKL ratio in the BM B220+ cells was measured in various groups by qPCR. Compared with the sham, the ratio was decreased in the Ovx group. All three antibody-treated groups reversed the Ovx-induced changes, but only the anti-IL-17 group was comparable to the sham. n = 10 mice/group; data are presented as mean ± SD; ***p < 0.001 compared with Ovx + vehicle group; *p < 0.05 compared with Ovx + vehicle group; ap < 0.001 compared with sham + vehicle group; ep < 0.001 compared with Ovx + anti-RANKL group; gp < 0.05 compared with Ovx + anti-RANKL group; pp < 0.001 compared with Ovx + anti-IL-17 group; rp < 0.05 compared with Ovx + anti-IL-17 group.

Download figure to PowerPoint

We next studied the effect of various treatments on Ovx-induced alteration of CD28 expression in T cells of BM. CD28 is a surface glycoprotein that is constitutively expressed on all CD4+ T cells, and decrease of its expression serves as a surrogate of T-cell senescence.[12, 49] CD4+CD28+ T-cell population in BM of the Ovx group was dramatically reduced compared with the sham group (p < 0.001) (Fig. 1B). CD4+CD28+ T-cell population in BM of the sham and anti-IL-17 groups were comparable (Fig. 1B). Anti-RANKL antibody treatment of Ovx mice failed to alter Ovx-induced loss of CD28 expression in CD4+ cells (Fig. 1B). Anti-TNF-α treatment to Ovx mice led to increase in CD4+CD28+ T cells, but it was significantly less than the sham or anti-IL-17 groups (p < 0.001) (Fig. 1B).

Ovx is known to increase the proliferation of B220+ cells in BM, whereas estrogen treatment completely reverses this change.[14] Mice Ovx for 4 weeks had a higher number of B220+ cells in BM compared with the sham group (p < 0.001) (Fig. 1C). Treatment of Ovx mice with either anti-RANKL or anti-TNF-α antibody failed to diminish the Ovx-induced increase in the B220+ population. However, the B220+ population was comparable between the sham and anti-IL-17 groups (Fig. 1C). Estrogen deficiency upregulates the production of RANKL from B cells,[50] and this cytokine is the most potent osteoclastogenic factor. B cells are also a rich source of osteoprotegrin (OPG), the decoy receptor for RANKL.[51] In B220+ cells isolated from Ovx mice, the OPG/RANKL ratio was significantly lower than the sham group (Fig. 1D). Treatment of Ovx mice with anti-TNF-α or anti-IL-17 antibody led to a significant increase in the OPG/RANKL ratio. The Ovx mice treated with anti-IL-17 antibody exhibited a robust increase in the ratio that was comparable to the sham group. Increase of OPG/RANKL ratio was least in the Ovx + anti-RANKL group when compared with other antibody-treated groups (Fig. 1D).

Effects on production of proinflammatory and anti-inflammatory cytokines

Aberrant immune response contributes to the pathogenesis of postmenopausal osteoporosis via an altered proinflammatory to anti-inflammatory balance. Various human and animal experiments have confirmed that there is an alteration in the levels of various proinflammatory and anti-inflammatory cytokines during bone loss.[9] Therefore, we examined the effect of various antibody treatments on the levels of major proinflammatory and anti-inflammatory cytokines. In the Ovx group, there was a significant increase in the levels of IL-6, IL-17, IFN-γ, and TNF-α and decrease in IL-10 over the sham group (Table 1). Interestingly, among the antibody treatment groups, the anti-IL-17 group had the most favorable cytokine profile, exemplified by substantial lowering of proinflammatory cytokines such as IL-6, IFN-γ, and TNF-α and increase in anti-inflammatory cytokine, IL-10, when compared with the Ovx group (Table 1). The levels of IL-6 and TNF-α, the two potent osteoclastogenic cytokines, were comparable between the anti-IL-17 and the sham groups. By contrast, the levels of these two osteoclastogenic cytokines as well as IL-10 were comparable between the Ovx, anti-TNF-α, and anti-RANKL groups. IL-17 and IL-12 (another proinflammatory cytokine) were also decreased significantly in the anti-IL-17 group compared with the Ovx group, but the magnitude was less than that observed with IL-6, IFN-γ and TNF-α. IL-17 and IL-12 levels were comparable between the groups that received antibody therapies. Taken together, these results suggested that amongst the three, anti-IL-17 treatment had the best potential to inhibit the production of proinflammatory cytokines and induce anti-inflammatory cytokines.

Table 1. Effect of Anti-TNF-α, Anti-RANKL, and Anti-IL-17 on Expression of Various Cytokines
CytokineShamOvxOvx + anti-TNF-αOvx + anti-RANKLOvx + anti-IL-17
  • Data are mean ± SE;

  • ***

    p < 0.001,

  • **

    p < 0.01,

  • *

    p < 0.05 compared with Ovx vehicle group;

  • d

    p < 0.001,

  • e

    p < 0.01 compared with sham group; and

  • f

    p < 0.001 compared with anti IL-17 group.

IL-65.35 ± 0.16***11.55 ± 0.2610.55 ± 0.45f,d10.30 ± 0.56f,d5.60 ± 0.24***
IL-107.46 ± 3.22***3.30 ± 0.183.34 ± 0.18f,d4.03 ± 0.23f,d9.12 ± 0.21***,d
IL-1237.49 ± 1.0538.30 ± 0.8634.78 ± 0.55*35.17 ± 0.41*35.38 ± 0.74*
IL-1777.55 ± 0.81***119.26 ± 3.22102.93 ± 2.63***,d ***,d106.22 ± 1.69***,d108.47 ± 1.49**,d
IFN-γ7.64 ± 0.17***20.34 ± 0.6718.28 ± 0.30**,f,d17.67 ± 0.33***,f,d9.75 ± 0.56***,e
TNF-α177.73 ± 4.46***265.80 ± 4.51276.94 ± 4.69f,d271.55 ± 4.34f,d182.34 ± 5.91***

Effect on Treg/Th17 balance

The two T-cell subsets (Th17 and Tregs) have important regulatory roles in immune function. Th17 cell is a key player in the pathogenesis of autoimmune diseases and protection against bacterial infections, whereas Tregs restrain excessive effector T-cell responses.[52] Proinflammatory cytokines are known to induce the production of Th17 and suppress Tregs.[52] Because proinflammatory cytokines were increased in the Ovx group, the effect of anti-RANKL/-TNF-α/-IL-17 was determined on Treg/Th17 balance in purified CD4+ cells from the BM. Data showed that in comparison to the controls, mRNA levels of retinoic-acid-receptor-related orphan receptor-α (ROR-α) (p < 0.001) and ROR-γt (p < 0.001) were robustly increased in the Ovx group (Fig. 2A, B). All three antibody-treated groups showed marked reduction of ROR-α and ROR-γt levels compared with the Ovx group; however, the anti-IL-17 group had the highest suppressive effect (Fig. 2A, B). As differentiation of Th17 cells is triggered by signal transducer and activator of transcription 3 (Stat 3),[53] expression of Stat 3 was determined. All three antibody-treated groups showed marked reduction in Stat 3 levels compared with the Ovx group (Fig. 2C). Expression of Foxp3 (p < 0.01) gene was significantly decreased in Ovx mice compared with the sham group, whereas treatment with anti-RANKL (p < 0.01), anti-TNF-α (p < 0.05), and anti-IL-17 (p < 0.001) antibodies to Ovx mice significantly upregulated Foxp3 expression with the anti-IL-17 effect being equivalent to the sham group (Fig. 2D). As Stat5 is a negative regulator of Th17 and promotes the development of Tregs,[53] mRNA levels of Stat 5 were also determined in all three antibody-treated groups. It was observed that Stat 5 mRNA levels were increased in anti-RANKL and anti-IL-17 groups. Interestingly, we found no difference in Stat 5 levels between Ovx and anti-TNF-α–treated groups (Fig. 2E). Thus, overall, all three regimens induced Tregs and suppressed polarization toward Th17 cells, but anti-IL-17 had the most potent effect owing to its robust effect on expression of key transcription factors ROR-α, ROR-γt, and Foxp3.

image

Figure 2. Effect of anti-TNF-α/-RANKL/-IL-17 on Th17/Treg balance studied using BM CD4+ T cells. (A) ROR-α mRNA level was robustly increased in the Ovx group compared with the sham, and all antibody-treated groups reversed this change. The ROR-α level in the anti-IL-17 group was lower than the sham. (B) Compared with the sham, the ROR-γt mRNA level was robustly increased in the Ovx group, and all antibody-treated groups reversed this change. The ROR-γt level in the anti-IL-17 group was lower than the sham. (C) Compared with the sham, the Stat 3 mRNA level was robustly increased in the Ovx group, and all antibody-treated groups reversed this change. (D) Foxp3 was decreased in the Ovx group compared with the sham. All three antibody-treated groups reversed the Ovx-induced changes, but only the anti-IL-17 group was comparable to sham. (E) Compared with the sham, the Stat 5 mRNA level was decreased in the Ovx group, and all antibody-treated groups except anti-TNF-α reversed this change. Data represent three independent experiments and expressed as mean ± SD with 95% confidence interval. Statistical analysis was performed by ANOVA method followed by the Newman–Keuls test of significance using Prism version 3.0 software. ***p < 0.001 compared with Ovx + vehicle group; **p < 0.01 compared with Ovx + vehicle group; *p < 0.05 compared with Ovx + vehicle group; ap < 0.001 compared with sham + vehicle group; bp < 0.01 compared with sham + vehicle group; fp < 0.01 compared with Ovx + anti-RANKL group, gp < 0.05 compared with Ovx + anti-RANKL group; qp < 0.01 compared with Ovx + anti-IL-17 group; rp < 0.05 compared with Ovx + anti-IL-17 group;. xp < 0.001 compared with Ovx + anti-TNF-α group; yp < 0.01 compared with Ovx + anti-TNF-α group; zp < 0.05 compared with Ovx + anti-TNF-α group.

Download figure to PowerPoint

Effect on trabecular bones

In gross observation by 3D µCT, deterioration of the trabecular architecture attributable to destruction of trabecular bone of femur was readily observed in the Ovx group compared with the sham group (Fig. 3A). Femoral response to various treatments was quantified. The Ovx group had reduced bone volume/trabecular volume (BV/TV; p < 0.001; Fig. 3B), trabecular thickness (Tb.Th; p < 0.05; Fig. 3C), and trabecular number (Tb.N; p < 0.001; Fig. 3D) and increased structure model index (SMI; p < 0.05) compared with the sham group (Fig. 3E). There were no differences in the trabecular parameters between the various antibody-treated groups, except SMI, where the anti-TNF-α group had no effect on the Ovx-induced increase of this parameter. Thus, all three antibody treatments equally prevented the decrease in trabecular bone volume, thickness, and number induced by Ovx.

image

Figure 3. Effect of anti-TNF-α/-RANKL/-IL-17 on Ovx-induced deterioration of trabecular microarchitecture in femur. After the end of treatment, femurs were collected in 70% isopropanol. (A) Representative images of femur epiphysis. (B) Bone volume/trabecular volume (BV/TV), (C) trabecular thickness (Tb.Th), and (D) trabecular number (Tb.N) were decreased in the Ovx compared with the sham, and all treatments significantly increased these parameters over the Ovx group. (E) Compared with the sham, the structure model index (SMI) was increased in the Ovx group, and anti-RANKL and anti-IL-17 treatments reversed the effect. n = 10 mice/group; data expressed as mean ± SD with 95% confidence interval. Statistical analysis was performed by one-way ANOVA nonparametric method followed by the Newman–Keuls test of significance using Prism version 3.0 software. ***p < 0.001, **p < 0.01, and *p < 0.05 compared with Ovx animals.

Download figure to PowerPoint

Effect on cortical bones

Two-dimensional µCT measurements at the site of femur mid-diaphysis showed that relative to the sham group, the Ovx group had decreased periosteal area (Tt.Ar) and cortical thickness (Ct.Th) (p < 0.001) (Fig. 4A, B). Compared with the Ovx group, the anti-IL-17 and anti-RANKL groups showed greater Tt.Ar, but the anti-TNF-α group had no effect (Fig. 4A). Ct.Th was not different between the Ovx, Ovx + anti-TNF-α, and Ovx + anti-RANKL groups, and this parameter was comparable between the sham and Ovx + anti-IL-17 groups (Fig. 4B).

image

Figure 4. Effect of anti-TNF-α/-RANKL/-IL-17 on Ovx-induced decrease of cortical bone and bone strength at femur mid-diaphysis. Compared with the sham, (A) periosteal area (Tt.Ar) and cortical thickness (Ct.Th) were decreased in Ovx and only anti-IL-17 reversed both effects. Tt.Ar was also reversed by anti-RANKL. In a 3-point bending experiment in femurs, both power and energy to break were reduced in the Ovx, and only anti-IL-17 reversed both effects. n = 10 mice/group; data are presented as mean ± SD; ***p < 0.001 compared with Ovx + vehicle group;**p < 0.01 compared with Ovx + vehicle group; *p < 0.05 compared with Ovx + vehicle group; bp < 0.01 compared with sham + vehicle group; cp < 0.05 compared with sham + vehicle group; qp < 0.01 compared with Ovx + anti-IL-17 group; rp < 0.05 compared with Ovx + anti-IL-17 group.

Download figure to PowerPoint

In the 3-point bending test of femur, all three antibody treatment groups exhibited power comparable to the sham (Fig. 4C). Bending energy was comparable between the sham and anti-IL-17 groups, and both were higher than the other three groups (Fig. 4D).

Effects on osteoclast functions

Ovx mice treated with anti-IL-17 (p < 0.001), anti-TNF-α (p < 0.001), and anti-RANKL (p < 0.001) antibodies had decreased osteoclast number compared with the Ovx group, but maximum reduction was observed in the anti-IL-17 group (Fig. 5A). Osteoclast surface/bone surface was also decreased in all antibody-treated groups compared with the Ovx group. However, the decrease in osteoclast surface produced by anti-IL-17 treatment was significantly greater (p < 0.05) than the anti-RANKL and anti-TNF-α treated Ovx animals (Fig. 5B). Osteoclast parameters were not different between the sham and anti-IL-17 groups. As circulating levels of CTX correlate with total osteoclast number and bone-resorbing activity,[54] serum CTX was determined. CTX level was increased in the Ovx group (p < 0.01) compared with sham. A decrease in CTX levels was observed in all three treatment groups with the anti-IL-17 group exhibiting levels even less than the sham group (Fig. 5C).

image

Figure 5. Effect of anti-TNF-α/-RANKL/-IL-17 on osteoclasts. Histomorphometric analysis of femoral epiphysis showing increased (A) number of osteoclasts (OC)/bone perimeter (N.Oc/B.Pm) and (B) osteoclast surface/bone surface (%) in the Ovx group over sham. All three treatments significantly reversed the Ovx-induced changes on osteoclasts. Only anti-IL-17 completely reversed Ovx-induced increase in N.Oc/B.Pm. (C) Ovx-induced increase in the serum CTX level was decreased in all antibody treatment groups. n = 5 mice/group; data are presented as mean ± SD; ***p < 0.001 compared with Ovx + vehicle group; **p < 0.01 compared with Ovx + vehicle group; rp < 0.01 compared with Ovx + anti-IL-17 group; gp < 0.05 compared with Ovx + anti-RANKL group; yp < 0.01 compared with Ovx + anti-TNF-α group.

Download figure to PowerPoint

Effect on osteoblast functions

Because treatments that promote osteoblast survival in Ovx skeleton are considered bone anabolic,[55] evidence of osteoblast apoptosis in decalcified bone sections of various treatment groups was investigated by immunostaining with Runx-2 (osteoblast marker) followed by TUNEL staining (indicative of DNA fragmentation). A representative photomicrograph showed staining with individual marker to localize Runx-2-positive nuclei on sections (Supplemental Fig. S1). Determination of ratio of Runx-2/TUNEL-positive cells showed that Ovx had a dramatic fall compared with the sham group, suggesting robust osteoblast apoptosis (Fig. 6A). The ratio was not different between the sham and anti-IL-17 groups. The ratio was higher in the anti-TNF-α and anti-RANKL over the Ovx group but significantly lower than the anti-IL-17 or sham group (Fig. 6A). To confirm the bone anabolic effect of anti-IL-17, mRNA levels of T-cell–produced Wnt 10b was determined. Wnt 10b induces the expression of osteoblastogenic transcription factors like Runx2, Dlx5, and osterix and strongly stimulates osteoblastogenesis.[56] It was observed that Wnt 10b expression was significantly low in Ovx animals compared with the sham group (p < 0.001). Treatment with anti-IL-17 antibody significantly increased the mRNA levels of Wnt 10b (p < 0.05), whereas anti-RANKL and anti-TNF-α had no effect (Fig. 6B). Because sclerostin (SOST) is a Wnt pathway inhibitor and one of the mechanisms by which PTH is believed to exert its anabolic effect by downregulating SOST,[57] the effect of antibody treatments on the SOST level was determined. All three antibodies equally inhibited the Ovx-induced increase in SOST mRNA levels (Fig. 6C).

image

Figure 6. Effect of anti-TNF-α/-RANKL/-IL-17 on osteoblast and bone lining cells. (A) Percentage of Runx-2/TUNEL-positive cells was calculated in various groups. Ovx bone sections had robust increase in apoptotic osteoblasts compared with the sham. Anti-IL-17 completely reversed this effect, whereas anti-RANKL had partial reversal and anti-TNF-α had no effect. (B) Wnt 10b mRNA was significantly reduced in the Ovx group compared with the sham. Only anti-IL-17 treatment to Ovx group led to significant enhancement of Wnt 10b mRNA levels. (C) SOST mRNA was significantly enhanced in the bones of Ovx group compared with the sham. All three antibody treatments to the Ovx group isolated from CD4+ T cells led to a significant reduction of SOST mRNA levels. (D) Anti-IL-17 treatment to Ovx animals maximally stimulated the production of lining cells on femoral surface compared with anti-TNF-α and anti-RANKL groups. (E) All three antibody treatments reversed the Ovx-induced decrease in serum PINP levels, but anti-IL-17 had the most potent effect. n = 5 mice/group; data are presented as mean ± SD; ***p < 0.001 compared with Ovx + vehicle group; **p < 0.01 compared with Ovx + vehicle group; *p < 0.05 compared with Ovx + vehicle group; bp < 0.01 compared with sham + vehicle group; cp < 0.05 compared with sham + vehicle group; ep < 0.001 compared with Ovx + anti-RANKL group; fp < 0.01 compared with Ovx + anti-RANKL group; gp < 0.05 compared with Ovx + anti-RANKL group; qp < 0.01 compared with Ovx + anti-IL-17 group; xp < 0.001 compared with Ovx + anti-TNF-α group; yp < 0.01 compared with Ovx + anti-TNF-α group; zp < 0.05 compared with Ovx + anti-TNF-α group.

Download figure to PowerPoint

Because reactivation of quiescent osteoblasts constitutes a major osteogenic response,[58] we quantified lining cell number in various groups. Lining cells were significantly reduced in Ovx mice compared with mice (p < 0.001). Treatment of Ovx mice with anti-IL-17 antibody led to a significant increase in the number of lining cells (p < 0.001), and this effect was better than that observed with anti-RANKL– (p < 0.05) and anti-TNF-α– (p < 0.05) treated groups (Supplemental Fig. S2 and Fig. 6D). Serum PINP levels, an accepted osteogenic marker,[59] was also determined and were significantly reduced in Ovx mice compared with the sham (Fig. 6E). Although all three antibody treatments led to an increase in serum PINP levels over the Ovx group, anti-IL-17 produced a significantly better effect than anti-RANKL and anti-TNF-α treatments (Fig. 6E).

Discussion

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

In this study, we compared the immunomodulatory potential of anti-IL-17, anti-RANKL, and anti-TNF-α antibodies in an estrogen-deficient bone loss condition and how these effects impact skeletal parameters. We report that anti-IL-17 treatment to Ovx mice presents with the best immunoprotective effects and these then translate to superior skeletal consequence than the anti-RANKL– and anti-TNF-α–treated groups.

Ovx causes an expansion of the T-cell pool in BM by increasing T-cell proliferation and life span.[11, 60] We observed that anti-IL-17 antibody reduced Ovx-induced increases in BM CD4+ T cells, whereas anti-TNF-α and anti-RANKL antibodies had no effect. Additionally, anti-TNF-α partially inhibited the Ovx-induced loss of CD28 expression on CD4+ T cells, whereas anti-IL-17 completely mitigated it. Anti-RANKL treatment, on the other hand, had no effect on the diminished CD28 levels on T cells, thus suggesting that neutralization of RANKL was ineffective for preventing Ovx-induced T-cell senescence. Senescent T cells acquire an enhanced ability to produce osteoclastogenic proinflammatory cytokines.[12] Ovx-induced increase in osteoclastogenic cytokines were best suppressed in the serum of the anti-IL-17 group compared with the other two antibody-treated groups. In fact, the anti-TNF-α and anti-RANKL groups failed completely in suppressing the Ovx-induced systemic increases in IL-6 and TNF-α and decreased IFN-γ only marginally, whereas all three cytokine levels in the anti-IL-17 group were comparable to the sham group. On the other hand, serum IL-10 was significantly elevated in the anti-IL-17 group over the sham, whereas anti-TNF-α and anti-RANKL antibodies failed to change Ovx-induced decrease of this cytokine at all, suggesting a more efficient anti-osteoclastogenic effect of anti-IL-17 at the systemic level over the other two antibodies.

It has been reported that Ovx selectively stimulates B-lymphopoiesis, resulting in an increased accumulation of pre-B cells in the BM, which, in turn, promotes resorption.[61] An Ovx-induced increase in B220+ cells in the BM was completely prevented by anti-IL-17, whereas anti-RANKL and anti-TNF-α had no effect. Mature B cells regulate osteoclastogenesis via the production of RANKL and OPG.[62] A decreased OPG/RANKL ratio in the osteoblasts of Ovx rodents has been reported.[14] We observed that this ratio was dramatically decreased in the B220+ cells of Ovx mice over the sham, and only the anti-IL-17 group exhibited a comparable ratio with the sham group. OPG/RANKL ratio correlated well with the osteoclast parameters as both osteoclast surface and number were comparable between the sham and anti-IL-17 groups. Serum CTX, whose circulating levels correlate with total osteoclast number and bone-resorbing activity, was also determined. Serum CTX was found to be maximally reduced in the anti-IL-17 treatment group, thus corroborating our data.

Recent discovery of two novel subsets of CD4+ T lymphocytes (Th17 and Tregs) has led to a paradigm shift in the understanding of how autoimmune responses are both mediated and regulated.[31] Whereas Th17 lymphocytes are key effector cells in diseases such as RA,[13] experimental autoimmune encephalomyelitis, and osteoporosis, Treg cells are essential for dominant immunologic tolerance.[63] Treg cells suppress immune response through production of anti-inflammatory cytokines, whereas Th17 cells promote the production of proinflammatory cytokines. The differentiation of TH17 cells is initiated by STAT3, whose activation induces the expression of ROR-α and ROR-γt. These two factors establish the TH17-cell-associated gene-expression program. STAT5, on the other hand, is a negative regulator of TH17-cell differentiation and promotes development of Tregs. In addition, the transcription factor forkhead box P3 (FOXP3) antagonizes the TH17-cell developmental program.[53] Thus, the effect of the three antibodies was studied on the Treg/Th17 balance. Data showed that although the Ovx-induced suppression of the Treg response was enhanced, the Th17 response was suppressed by all three antibody treatments out of which anti-IL-17 had the strongest effect, as it completely prevented Ovx-induced alteration of Foxp3, ROR-α, and ROR-γt genes. Thus, our data show that functional blocking of IL-17 is more potent in overcoming the Ovx-induced Treg-suppressive effect than blocking TNF-α or RANKL.

We next studied the skeletal effects of the three treatment groups. In an earlier report, we showed that anti-IL-17 significantly protected trabecular bones against Ovx-induced loss. Based on the trabecular response, it appeared that all three treatments improved trabecular microarchitecture equally when compared with the Ovx. Increase in periosteal apposition is one of the anabolic outcomes of PTH therapy. Inhibition of bone resorption by cathepsin K inhibitor, odanacatib has been shown to stimulate periosteal bone formation in rabbit, which has been purported to be a bone anabolic outcome.[64] We observed that cortical thickness of femur mid-diaphysis was comparable among the anti-IL-17 and sham groups and greater than the other two antibody-treated groups, thus suggesting that anti-IL-17 antibody was more effective in maintaining cortical bone than the other two therapies. In agreement with cortical histomorphometry data, femur biomechanical parameters showed that anti-IL-17 was comparable to the sham group in conferring a greater energy required for breaking force. Increased cortical deposition in the anti-IL-17 group over the other antibody-treated groups may be the result of complete mitigation of Ovx-induced apoptosis of osteoblasts and increase in number of bone lining cells in the femur diaphysis region by anti-IL-17 treatment. Additionally, anti-IL-17 antibody treatment induces the expression of T-cell–produced Wnt 10b, which then activates canonical Wnt signaling, leading to increased osteoblast differentiation and bone mass. A report by Li and colleagues[65] proposed that the sclerostin-independent bone anabolic activity of PTH treatment is mediated by T-cell–produced Wnt10b. Thus, it appears that increased Wnt 10b in the bones of Ovx + anti-IL-17 mice over the Ovx group could afford an osteogenic effect by anti-IL-17 treatment. Serum PINP, which is a specific and sensitive bone turnover marker for monitoring the anabolic mode of any treatment,[59] was also most robustly increased by anti-IL-17 treatment compared with anti-RANKL and anti-TNF-α treatments, which attested to an osteogenic effect of anti-IL-17 treatment.

TNF binding protein (TNFbp), which binds to both TNF-α and TNF-β with equal affinity and blocks total TNF activity, prevents ovariectomy-induced bone loss.[15] In this study, the dose regimen and frequency (1 mg/kg daily for 2 weeks) of dosing was more aggressive compared with our study, where a 4-week treatment of anti-TNF-α (1 mg/kg sc twice a week) significantly ameliorated Ovx-induced deterioration of trabecular, cortical, and strength parameters but failed to afford complete protection.[15] It is also possible that blocking of both TNF-α and TNF-β as achieved with TNFbp might have a stronger bone preservation effect over blocking TNF-α alone by anti-TNF-α.

Transgenic mouse expressing a chimeric human/mouse RANKL (huRANKL) upon Ovx exhibited complete skeletal preservation by human anti-RANKL (denosumab) treated for 8 weeks.[66] In our study, the use of anti-RANKL at 300 μg/kg twice a week that had ameliorated bone loss induced by collagen-induced arthritis in mice resulted in a partial protection from bone loss induced by Ovx at trabecular and cortical sites. This dose was about 30 times less than in the transgenic mouse study. In addition, skeletal metabolism may be different between a transgenic and a wild-type mouse.

In conclusion, anti-IL-17 antibody protects against Ovx-induced osteopenia by suppressing osteoclast function, promoting osteoblast survival and differentiation, and reversing immune-senescent proinflammatory changes triggered by Ovx. All these effects are better with anti-IL-17 than anti-RANKL or anti-TNF-α therapy. This study forms a strong basis for using a humanized antibody against IL-17 cytokine toward the treatment of postmenopausal osteoporosis.

Acknowledgments

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

Generous funding from the Centre for Research in Anabolic Skeletal Targets in Health and Illness (ASTHI) is acknowledged. Other supporting grants are from Council of Scientific and Industrial Research (CSIR), Department of Biotechnology (DBT), Department of Science and Technology (DST), Government of India. CDRI communication number 8643.

We acknowledge AL Vishwakarma for helping with the FACS analysis.

Authors' roles: Study design: AMT, DS, and MNM. Study conduct: AMT, MNM, MD, and JK. Data analysis: KS, PS, RT, NC, and DS. Drafting manuscript: DS, NC, AMT, and KS.

References

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

Supporting Information

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

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
jbmr2228-sm-0001-SupFig-S1.docx1320KSupplementary Figure S1.
jbmr2228-sm-0002-SupFig-S2.doc1695KSupplementary Figure S2.

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