Abstract
Objectives.
The Wnt signaling antagonist Dickkopf-1 (DKK1) inhibits osteoblast differentiation and function and has been described to play a central role in promoting bone loss, while blockade of DKK1 increases bone formation. We investigated the effects of DKK1 on periosteal new bone formation in two murine models of inflammatory arthritis, the antigen-induced arthritis (AIA) and K/BxN serum transfer arthritis (STA) models.
Methods.
The flare variant of AIA was induced in wild type mice and a blocking antibody to DKK1, control rat IgG, or PBS was administered starting on day 14, a time at which inflammation and erosions are known to be established. Knees were assessed for histologic inflammation and periosteal new bone formation was quantitated. Additionally, STA was generated in transgenic mice with osteoblast specific overexpression of Dkk1 and littermate controls. New bone formation around the wrists of these mice was quantified by micro-CT.
Results.
Blockade of DKK1 in arthritic mice resulted in significantly more periosteal new bone formation compared to mice treated with control rat IgG or PBS. Conversely, in the setting of increased Dkk1 expression, arthritic Dkk1 Tg mice developed significantly less periosteal new bone than arthritic controls.
Conclusion.
DKK1 is a regulator of periosteal bone formation in inflammatory arthritis. Thus, regulation of DKK1 may be considered as a therapeutic approach in inflammatory diseases in which patients suffer from excessive periosteal bone formation, such as spondylitis.
Keywords: bone, inflammatory arthritis, Wnt signaling, dickkopf-1 (DKK1), serum transfer, periosteal
Introduction
New bone formation at periosteal sites where tendons and ligaments insert into bone is a hallmark of spondylitis (SpA). However, the pathogenesis of this bone formation remains unclear. Inflammation-induced trabecular bone loss has been associated with new bone formation in ankylosing spondylitis patients (1) and candidate molecules that drive new bone formation include those in the anabolic wingless (Wnt) signaling pathway. Antagonists of this pathway, including DKK1, impair osteoblast differentiation and function. DKK1 is secreted by cells within inflamed synovial tissues and is a key player in bone remodeling (2). Elevated DKK1 levels result in bone loss, while reduced levels increase bone formation and bone mass in mice (3, 4).
These findings translate to human disease, as DKK1 levels are increased in serum from RA patients compared to healthy controls, but are low in patients with SpA (2, 5–7). Blockade of DKK1 in mice promotes bone fracture healing (8), sacroiliac joint ankylosis (9) and peripheral joint osteophyte formation (2). Antigen-induced arthritis (AIA) and K/BxN serum transfer arthritis (STA) are models of inflammatory arthritis characterized by bone erosion. However, periosteal bone formation occurs reproducibly in both models and recapitulates the osteophytes and enthesophytes that form in SpA (10–12). We sought to determine whether modulating DKK1 alters periosteal bone formation through two complementary approaches: DKK1 blockade in AIA and Dkk1 overexpression in osteoblasts (3) in STA.
Methods
AIA flare variant with anti-DKK1 antibody treatment.
Procedures were approved by the IACUC at University of Massachusetts Medical School. Induction of the flare variant of AIA (13) was modified. 9-week-old male C57BL/6J mice were immunized on day −21 with 100μl of 4mg/ml methylated bovine serum albumin (mBSA, Sigma-Aldrich) emulsified in complete Freund’s adjuvant (Sigma-Aldrich), supplemented to a final concentration of 2.5mg/ml M. tuberculosis H37Ra (Difco). 4mg/ml mBSA emulsified in incomplete Freund’s adjuvant (Sigma-Aldrich) was administered on day −7. On day 0, AIA was induced by knee injection of 60μg mBSA in PBS with an intraperitoneal (IP) injection of 200μg of lipopolysaccharide (LPS, Sigma-Aldrich) as adjuvant. On days 10 and 20, 2μg mBSA were injected into the knee to induce arthritis flares. Mice were treated either with PBS (n=10), a rat antibody directed against murine DKK-1 (10mg/kg, n=16) or rat IgG control antibody (10mg/kg, n=16), both provided by Lilly Research Laboratories, 2x/week IP from day 14 to day 38. Untreated mice euthanized on day 14 (peak inflammation and erosion) served as comparators (n=16).
Serum transfer arthritis.
Procedures were approved by the IACUC at University of Massachusetts Medical School and Harvard Center for Comparative Medicine. Arthritogenic serum was harvested from 9-week-old arthritic K/BxN mice (14, 15). Arthritis was induced in 13-week-old male Dkk1 transgenic mice (Dkk1 Tg) (3) and WT littermates (n=7 each) with 150μl of arthritogenic serum on days 0, 2 and 7. Dkk1 overexpression in Dkk1 Tg mice was confirmed by qPCR on RNA isolated from diaphyseal bone marrow (Supplemental Figure). An observer blinded to genotype assessed clinical inflammation according to published protocols (16). Mice were euthanized on day 14 and forepaws were fixed in 70% ethanol for imaging. Age-matched Dkk1 Tg male mice without arthritis were controls.
Histopathologic analysis.
AIA knees and STA ankles were fixed (24 hours, 4% paraformaldehyde), decalcified (15% EDTA in PBS/0.5% PBS), and paraffin embedded. 5μm AIA knee sections (100, two/slide) or 5μm STA ankle sections (50, one/slide) were cut for analysis. Every tenth serial section was stained with hematoxylin and eosin (H&E). Slides of AIA knees were scored using previously defined histopathologic scoring criteria (17).
Quantitation of periosteal bone formation.
Images of H&E-stained sections (#10, 30, and 50) were captured using a Nikon DS-Ri1 camera at four reproducible sites of periosteal bone formation at the knee joint: the medial and lateral patella, and medial and lateral femur. Areas of bone formation were measured using the NIS Elements BR software (Nikon, Melville, NY, USA). The average area for each site was calculated for each knee and areas at all sites were totaled to determine total area of bone formation, which was then multiplied by 0.4mm (the distance through sections 10 to 50) to calculate volume. Total periosteal bone volume for each murine knee was determined by totaling the volumes of bone formation from all four sites.
Micro-CT imaging.
Periosteal bone formation at the forepaws was quantitated on a μCT-35 (Scanco Medical, Wayne, PA, USA). Forearms were scanned with an isotropic voxel size of 7μm. A region of interest (ROI) starting from the radial or ulnar growth plate and extending to the end of visible periostitis (2.1–2.8mm) was analyzed using manufacturer’s software. For no arthritis controls, an ROI extending 2.4mm proximally was analyzed. A semi-automated contouring approach was used to distinguish cortical and periosteal bone for the radius and ulna separately. Periosteal bone in the ROI was thresholded using a global threshold that set the bone/soft tissue cut-off at 389.1mg HA/cm3 and bone volume was calculated. Periosteal bone volumes at the ulna and radius were added to calculate total periosteal bone volume for each forearm. Bone formation in STA is driven by inflammation. As paw thickness changes reflect inflammation severity, we excluded six mice (2 Dkk1 Tg and 4 WT) from analysis based on change in paw thickness <0.5 mm in one or more paws. These mice with minimal change in paw thickness developed very little to no inflammation and thus, would lack periosteal new bone formation.
Statistical analysis.
Prism 9.0 (GraphPad Software, La Jolla, CA, USA) was used for graphing and statistical analysis. The Mann-Whitney test was used to calculate the statistical significance between different treatment groups. All graphs show the median with interquartile range. P values <0.05 were considered significant.
Results
Blockade of DKK1 promotes periosteal bone formation in AIA.
We investigated whether DKK1 blockade affects periosteal bone formation in the flare model of AIA. WT mice were induced with AIA and two additional arthritis flares were administered to sustain the arthritis. Beginning at peak inflammation (day 14), mice were administered PBS, an anti-DKK1 antibody or a control rat IgG twice weekly to day 38. Histologic inflammation was similar among treatment groups at day 38, indicating that inhibition of DKK1 does not affect inflammation (Figure 1A). Similar inflammation among treatment groups at day 38 allows for the comparison of periosteal bone formation. Quantitation of bone formation volume at the patella and femur, reproducible sites of periosteal bone formation in AIA, revealed significantly more new bone in arthritic mice treated with the anti-DKK1 antibody compared to either control rat IgG antibody or PBS treated groups (Figures 1B and C).
Figure 1: Inhibition of DKK1 results in increased periosteal bone formation in the flare model of AIA.
(A) Histologic inflammation scores at peak inflammation (day 14) and the experimental endpoint (day 38). Each symbol represents the histologic score of the left arthritic knee from a single mouse. **** P<0.0001 versus peak erosion. (B) Quantitation of periosteal new bone formation volume at peak erosion (day 14) and the experimental endpoint (day 38). Each symbol represents the volume of periosteal bone formation of the left arthritic knee from a single mouse. ** P=0.0050, **** P<0.0001. (C) Representative H&E stained sections of the patella (top row) and femur (bottom row) of arthritic knees from each group. Arrows indicate areas of periosteal new bone formation and dotted lines delineate the cortical bone. Scale bar represents 100μm.
Overexpression of Dkk1 inhibits periosteal new bone formation in STA.
We evaluated the effects of increased expression of Dkk1 on periosteal bone formation in STA. In Dkk1 Tg mice, Dkk1 overexpression is under the control of the 2.3-kb Col1a1 promoter and DKK1 is overexpressed in osteoblasts (3). Dkk1 Tg and WT mice induced with STA developed a similar degree of inflammation, as assessed by arthritis score (Figure 2A) and change in paw thickness (Figure 2B), suggesting that overexpression of Dkk1 does not affect inflammation severity in this model. The new bone that developed around the wrists (ulna and radius) was analyzed by micro-CT scanning. Arthritic WT and arthritic Dkk1 Tg mice developed significant new bone at the forepaws compared to non-arthritic Dkk1 Tg mice, but arthritic Dkk1 Tg mice developed significantly less new bone compared to arthritic WT mice (Figures 2C and D), suggesting that overexpression of Dkk1 limits the new bone formation in an inflammatory setting.
Figure 2: Periosteal bone formation is suppressed in Dkk1 Tg mice induced with STA.
Average (A) clinical arthritis score and (B) change in paw thickness over time by genotype for WT (black circles) and Dkk1 Tg (gray triangles) mice induced with STA. (C) Periosteal new bone volume at each forepaw was quantitated by micro-CT in non-arthritic Dkk1 Tg and both arthritic WT and Dkk1 Tg mice at day 14 of STA. Each data point represents one forepaw. ** P=0.0017, **** P<0.0001. (D) Representative 3D reconstructions of forepaws from non-arthritic Dkk1 Tg (left panel), arthritic WT (middle panel) and arthritic Dkk1 Tg (right panel) mice induced with STA depicting periosteal new bone formation.
Discussion
The Wnt signaling antagonist DKK1 is a crucial regulator of bone remodeling. We examined the impact of DKK1 on periosteal bone formation in the setting of inflammatory arthritis in STA and in the flare variant of AIA. We demonstrate that inhibition of DKK1 increased, while Dkk1 overexpression in osteoblasts decreased, development of periosteal new bone. In neither model did modulation of DKK1 affect inflammation, indicating that DKK1 exerts its effects primarily on bone. These results are consistent with a seminal study of an inflammatory model of arthritis in which mice that overexpress TNF formed osteophytes after treatment with a DKK1 neutralizing antibody (2) and reflect the expected inhibitory effect of DKK1 on osteoblast differentiation and function.
The periosteal bone observed in the flare variant of AIA occurred specifically at tendon and ligament insertion sites in arthritic knees, resulting in enthesophytes. Notably, these enthesial sites are also areas of mechanical stress, which has been identified as a contributing factor in bone formation in SpA (18). Thus, the formation of new bone evident in the control rat IgG and PBS treatment groups was not unexpected. However, mice treated with the anti-DKK1 antibody developed strikingly more bone, presumably because of the relief of antagonism of the Wnt pathway, resulting in increased Wnt signaling and osteoblast differentiation. Conversely, Dkk1 Tg mice developed less new bone around the wrists compared to controls. This new bone was confined to the periosteal surfaces and did not develop from entheses, suggesting that modulating DKK1 function can affect inflammatory new bone formation at multiple locations.
In conclusion, using two distinct methods of inflammation-driven bone formation, we demonstrate that inhibition or overexpression of DKK1 impacts bone formation. These results indicate that DKK1 plays a significant role in directing new bone formation in the arthritic inflammatory microenvironment. Our findings are particularly relevant to patients with SpA who suffer from excess periosteal and enthesial bone formation.
Supplementary Material
Relative expression of Dkk1 measured by qPCR is increased in Dkk1 Tg (gray triangles) mice compared to WT littermates (black circles). Briefly, RNA was extracted from the diaphyseal region of the femur of Dkk1 Tg and WT mice by homogenizing in TRIzol (Invitrogen) using the Navy RINO lysis kit and Bullet Blender (Next Advance) followed by column purification (Qiagen). RNA was reverse transcribed using Affinity Script cDNA synthesis kit (Agilent) and Dkk1 expression relative to the Tbp housekeeping gene was assessed by qPCR using the delta delta CT method and the following primers: Dkk1-F CTCATCAATTCCAACGCGATCA, Dkk1-R GCCTCATAGAGAACTCCCG, Tbp-F CTACCGTGAATCTTGGCTGTAAAC and Tbp-R AATCAACGCAGTTGTCCGTGGC. Each data point represents one animal. * P=0.014.
Acknowledgments
We would like to thank Lilly Research Laboratories for providing the anti-DKK-1 antibody and rat IgG and Dr. Henry Kronenberg for the gift of Dkk1 Tg mice. KRN T cell transgenic mice were kindly provided by Drs. Benoist and Mathis, Harvard Medical School, Boston, MA, USA and Institut de Genetique et de Biologie Moleculaire et Cellulaire, Illkirch, France.
Funding
This work was supported by grants from the National Institutes of Health [R01-AR055952 to E.G., K08 AR062590 to J.F.C.]; the Rheumatology Research Foundation and the Department of Orthopedics [J.F.C.].
Footnotes
Conflict of interest
JFC and EG receive author royalties from Up To Date, Inc. EG receives royalties from the textbook Rheumatology and salary from The New England Journal of Medicine. EG has consulted for Eli Lilly and received grant funding from AbbVie; JFC has served on advisory boards for Ultragenyx. All other authors have declared no conflicts of interest.
References
- 1.Kang KY, Jung J-Y, Lee SK, Min HK, Hong YS, Park S-H, et al. Trabecular bone score value is associated with new bone formation independently of fat metaplasia on spinal magnetic resonance imaging in patients with ankylosing spondylitis. Scand J Rheumatol 2020;49:292–300. [DOI] [PubMed] [Google Scholar]
- 2.Diarra D, Stolina M, Polzer K, Zwerina J, Ominsky MS, Dwyer D, et al. Dickkopf-1 is a master regulator of joint remodeling. Nat Med 2007;13:156–63. [DOI] [PubMed] [Google Scholar]
- 3.Guo J, Liu M, Yang D, Bouxsein ML, Saito H, Galvin RJ, et al. Suppression of Wnt signaling by Dkk1 attenuates PTH-mediated stromal cell response and new bone formation. Cell Metab 2010;11:161–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Morvan F, Boulukos K, Clement-Lacroix P, Roman-Roman S, Suc-Royer I, Vayssière B, et al. Deletion of a single allele of the Dkk1 gene leads to an increase in bone formation and bone mass. J Bone Miner Res 2006;21:934–45. [DOI] [PubMed] [Google Scholar]
- 5.Kwon SR, Lim MJ, Suh CH, Park SG, Hong YS, Yoon BY, et al. Dickkopf-1 level is lower in patients with ankylosing spondylitis than in healthy people and is not influenced by anti-tumor necrosis factor therapy. Rheumatol Int 2012;32:2523–7. [DOI] [PubMed] [Google Scholar]
- 6.Liao H-T, Lin Y-F, Tsai C-Y, Chou T-C. Bone morphogenetic proteins and Dickkopf-1 in ankylosing spondylitis. Scand J Rheumatol 2018;47:56–61. [DOI] [PubMed] [Google Scholar]
- 7.CiCi D, Corrado A, Rotondo C, Cantatore F. Wnt signaling and biological therapy in rheumatoid arthritis and spondyloarthritis. Int J Mol Sci 2019;20:5552. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Jin H, Wang B, Li J, Xie W, Mao Q, Li S, et al. Anti-DKK1 antibody promotes bone fracture healing through activation of beta-catenin signaling. Bone 2015;71:63–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Uderhardt S, Diarra D, Katzenbeisser J, David JP, Zwerina J, Richards W, et al. Blockade of Dickkopf (DKK)-1 induces fusion of sacroiliac joints. Ann Rheum Dis 2010;69:592–7. [DOI] [PubMed] [Google Scholar]
- 10.Ruiz-Heiland G, Horn A, Zerr P, Hofstetter W, Baum W, Stock M, et al. Blockade of the hedgehog pathway inhibits osteophyte formation in arthritis. Ann Rheum Dis 2012;71:400–7. [DOI] [PubMed] [Google Scholar]
- 11.Schett G, Stolina M, Dwyer D, Zack D, Uderhardt S, Krönke G, et al. Tumor necrosis factor alpha and RANKL blockade cannot halt bony spur formation in experimental inflammatory arthritis. Arthritis Rheum 2009;60:2644–54. [DOI] [PubMed] [Google Scholar]
- 12.Shaw AT, Maeda Y, Gravallese EM. IL-17A deficiency promotes periosteal bone formation in a model of inflammatory arthritis. Arthritis Res Ther 2016;18:104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Koenders MI, Lubberts E, Oppers-Walgreen B, van den Bersselaar L, Helsen MM, Di Padova FE, et al. Blocking of interleukin-17 during reactivation of experimental arthritis prevents joint inflammation and bone erosion by decreasing RANKL and interleukin-1. Am J Pathol 2005;167:141–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Korganow AS, Ji H, Mangialaio S, Duchatelle V, Pelanda R, Martin T, et al. From systemic T cell self-reactivity to organ-specific autoimmune disease via immunoglobulins. Immunity 1999;10:451–61. [DOI] [PubMed] [Google Scholar]
- 15.Kouskoff V, Korganow AS, Duchatelle V, Degott C, Benoist C, Mathis D. Organ-specific disease provoked by systemic autoimmunity. Cell 1996;87:811–22. [DOI] [PubMed] [Google Scholar]
- 16.Matzelle MM, Gallant MA, Condon KW, Walsh NC, Manning CA, Stein GS, et al. Resolution of inflammation induces osteoblast function and regulates the Wnt signaling pathway. Arthritis Rheum 2012;64:1540–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Mould AW, Tonks ID, Cahill MM, Petit AR, Thomas R, Hayward NK, et al. Vegfb gene knockout mice display reduced pathology and synovial angiogenesis in both antigen-induced and collagen-induced models of arthritis. Arthritis Rheum 2003;48:2660–9. [DOI] [PubMed] [Google Scholar]
- 18.Jacques P, Lambrecht S, Verheugen E, et al. Proof of concept: enthesitis and new bone formation in spondyloarthritis are driven by mechanical strain and stromal cells. Ann Rheum Dis 2014;73:437–45. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Relative expression of Dkk1 measured by qPCR is increased in Dkk1 Tg (gray triangles) mice compared to WT littermates (black circles). Briefly, RNA was extracted from the diaphyseal region of the femur of Dkk1 Tg and WT mice by homogenizing in TRIzol (Invitrogen) using the Navy RINO lysis kit and Bullet Blender (Next Advance) followed by column purification (Qiagen). RNA was reverse transcribed using Affinity Script cDNA synthesis kit (Agilent) and Dkk1 expression relative to the Tbp housekeeping gene was assessed by qPCR using the delta delta CT method and the following primers: Dkk1-F CTCATCAATTCCAACGCGATCA, Dkk1-R GCCTCATAGAGAACTCCCG, Tbp-F CTACCGTGAATCTTGGCTGTAAAC and Tbp-R AATCAACGCAGTTGTCCGTGGC. Each data point represents one animal. * P=0.014.