PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Osteoarthritis Cartilage. Author manuscript; available in PMC 2017 April 1.
Published in final edited form as:
PMCID: PMC4799757
NIHMSID: NIHMS734618

Notch Signaling in Postnatal Joint Chondrocytes, but not Subchondral Osteoblasts, is Required for Articular Cartilage and Joint Maintenance

Abstract

Objective

Notch signaling has been identified as a critical regulator in cartilage development and joint maintenance, and loss of Notch signaling in all joint tissues results in an early and progressive osteoarthritis (OA)-like pathology. This study investigated the targeted cell population within the knee joint in which Notch signaling is required for normal cartilage and joint integrity.

Methods

Two loss-of-function mouse models were generated with tissue-specific knockout of the core Notch signaling component, RBPjκ. The AcanCreERT2 transgene specifically removed Rbpjκ floxed alleles in postnatal joint chondrocytes, while the Col1Cre(2.3kb) transgene deleted Rbpjκ in osteoblast populations, including subchondral osteoblasts. Mutant and control mice were analyzed via histology, immunohistochemistry, real-time qPCR, X-ray, and microCT imaging at multiple time-points.

Results

Loss of Notch signaling in postnatal joint chondrocytes results in a progressive OA-like pathology, and triggered the recruitment of non-targeted fibrotic cells into the articular cartilage potentially due to mis-regulated chemokine expression from within the cartilage. Upon recruitment, these fibrotic cells produced degenerative enzymes that may lead to the observed cartilage degradation and contribute to a significant portion of the age-related OA-like pathology. On the contrary, loss of Notch signaling in subchondral osteoblasts did not affect normal cartilage development or joint maintenance.

Conclusions

RBPjκ-dependent Notch signaling in postnatal joint chondrocytes, but not subchondral osteoblasts, is required for articular cartilage and joint maintenance.

Keywords: Notch signaling, osteoarthritis, cartilage, chondrocytes, subchondral bone, synovium

Introduction

Osteoarthritis (OA), the most common degenerative joint disorder observed worldwide, is characterized by fibrosis and loss of joint cartilage, degradation of menisci, osteophyte formation, subchondral bone sclerosis, and variable degrees of synovial tissue dysplasia [1, 2]. Many risk factors have been identified to contribute to the development and progression of OA, including aging, obesity, traumatic injury, and genetics [14]. Though the etiology of OA is still poorly understood, several genetic pathways that regulate cartilage anabolism and catabolism have been associated with the pathogenesis of OA [3, 5, 6].

Notch signaling has recently been identified as a critical regulator in cartilage development and joint maintenance [711]. Many Notch pathway components are expressed in both developing and postnatal cartilages [12, 13], suggesting a functional role for Notch signaling in regulating both cartilage development and homeostasis. In mammals, the Notch signaling pathway is activated when one of the Notch ligands (Jagged1 and 2, Delta-like 1, 3, 4) bind to a Notch receptor (Notch1-4), leading to a series of receptor cleavages mediated by the γ-secretase complex that culminates in the release of the Notch intracellular domain (NICD) into the cytoplasm. The NICD then translocates into the nucleus and binds to the transcriptional regulator, RBPjκ, to form a transcriptional activator complex, which induces the expression of downstream Notch target genes that include several Hes/Hey gene family members [1416].

Previously we reported that loss of RBPjκ-dependent Notch signaling in all joint tissues (Prx1Cre; Rbpjκf/f) including the articular cartilage, meniscus, synovium, ligaments, and subchondral bone, resulted in a severe, early, and progressive OA-like pathology [10]. Additionally, partial removal of Rbpjκ floxed alleles in postnatal joint cartilages using an inducible, although inefficient, Col2CreERT2 transgene led to a much less severe but progressive cartilage degeneration phenotype with age [10]. These data suggested that RBPjκ-dependent Notch signaling is required for postnatal joint maintenance, and it may function partially through signaling within postnatal joint chondrocytes. Due to recombination inefficiencies of our prior cartilage-specific mutant mouse model and the broad tissue targets within the joint of the Prx1Cre transgene, several questions remain concerning the cell-specific and functional role of RBPjκ-dependent Notch signaling in joint cartilage development and maintenance: 1) Would a more complete removal of RBPjκ in postnatal joint cartilages lead to the development of a degenerative joint phenotype that more closely resembles the Prx1Cre; Rbpjκf/f mutant phenotype?, 2) Is RBPjκ-dependent Notch signaling required in other joint tissues/cells, such as the subchondral osteoblasts, for normal joint development and/or maintenance?, and 3) What are the cellular and molecular events responsible for joint degeneration in the absence of RBPjκ-dependent Notch signaling?

To address some of these outstanding issues and questions, we developed two tissue-specific loss-of-function mouse models with effective knockout of RBPjκ in postnatal joint cartilages and subchondral bone, respectively. We also applied immunohistochemistry to our cartilage-specific mutant mouse model to examine the origin of the fibrotic cells that are observed in the Notch deficient and degenerating joint cartilages. Here we report that removal of RBPjκ-dependent Notch signaling from postnatal joint chondrocytes, but not subchondral osteoblasts, triggers the recruitment of non-targeted fibrotic cells into the articular cartilage that produce degenerative enzymes and contribute to the age-related OA-like pathology.

Materials and Methods

Mouse strains

Animal studies were approved by the University of Rochester Committee on Animal Resources. All mouse strains, including AcanCreERT2 [17], Col1a1Cre(2.3kb) [18], Rbpjκf/f [19], and Rosa-LacZ reporter (R26Rf/f) [20], and Rosa26 Td-Tomato reporter (R26Tomatof/f) [21] have been described previously. AcanCreERT2; Rbpjκf/f (RBPjκAcanTM), AcanCreERT2; R26Rf/+ (R26RAcanTM), Col1a1Cre(2.3kb); Rbpjκf/f (RBPjκCol1) and Col1a1Cre(2.3kb); R26Tomatof/+ (R26TomatoCol1) mice were viable and produced in Mendelian ratios. Tamoxifen (TM, 1mg/10g body weight) was administered daily via intraperitoneal (i.p.) injection to RBPjκAcanTM mice and littermate controls or R26RAcanTM mice and littermate controls for 5 continuous days starting at either P14-18 or 2 months of age. Mice were harvested at P14, P20, 2 months, 4 months, and 8 months of age. Overall, RBPjκAcanTM mutant mice showed no obvious difference in skeletal size or body weight compared to littermate controls regardless of the tamoxifen regimen.

Analyses of mice

Postnatal knee joints were harvested and fixed in 10% neutral buffered formalin for 3 days, decalcified in Formic Acid Bone Decalcifier (Immunocal, Decal Chemical Corp.) for 7–10 days, paraffin processed, and embedded for sectioning. Tissues were sectioned at 5 μm and stained with Alcian blue, hematoxylin, orange-g (ABH/OG) and Safranin-O (Saf-O). IHC analyses were performed on sections using traditional antigen retrieval and colorimetric development methodologies with the following primary antibodies: RBPJκ (Cell Signaling), Sry box factor 9 (SOX9) (Santa Cruz), Collagen type II (COL2A1) (Thermo Scientific), Collagen type X (COL10A1) (Quartett), Collagen type III (COL3A1) (Abcam), Matrix metalloproteinase 13 (MMP13) (Thermo Scientific), NITEGE (MDBioProducts), Chemokine C-C motif ligand 20 (CCL-20) (Abcam), and Stromal cell derived factor 1 (SDF-1) (R&D Systems). TUNEL cell death assay was performed on paraffin sections using the in situ Cell Death Detection Kit, Fluorescein (Roche) according to the manufacturer’s instructions. X-gal and immunofluorescent staining was performed on frozen sections as previously described [22]. Trichrome staining was performed on frozen sections using the Trichrome Stain kit (Abcam) according to the manufacturer’s instructions. Briefly, postnatal knee joints were harvested and fixed in 4% paraformaldehyde (PFA) for 2 hours at 4 °C and decalcified with 14% EDTA at 4 °C for 10 days. Tissues were washed in sucrose gradient, embedded with Tissue-TEK OCT medium, snap frozen in liquid nitrogen and sectioned at 10 μm using a Lecia CM 1850 cryotome. MicroCT analyses were performed on 2- and 8-month-old mouse knees prior to decalcification using a VivaCT 40 scanner (Scanco USA) as previously described [23]. OARSI scoring of knee joints was performed on Safranin-O stained sections as previously described [24]. At least three independent mice were analyzed for each group. The precise number of animals used for each experiment is highlighted in the figure legends.

Articular cartilage isolation and gene expression analyses

Articular cartilage was isolated from the knee joints of at least three independent 4-month-old control and mutant mice. Cartilages were then pooled together and digested to chondrocyte cell suspensions as previously described [10]. RNA was isolated from the cell suspensions using an Rneasy Mini kit (Qiagen). Complementary DNA synthesis and real-time quantitative polymerase chain reaction (qPCR) were performed on RNA extracted from the articular chondrocytes as previously described [10]. Primer sequences for Hes1, Sox9, Col2a1, Acan1, Mmp13 and Adamts5 are available upon request.

Western blot

Bone marrow cells were flushed away and bone proteins were extracted from femora and tibiae of RBPjκCol1 mice and littermate controls with RIPA buffer supplemented with protease inhibitors and phosphatase inhibitors (Thermo Scientific). Proteins were fractionated in an SDS-PAGE gel, transferred to a nitrocellulose membrane, and detected with the RBPjκ antibody (Cell Signaling).

Statistical analysis

All experiments except for the real-time qPCR analyses were performed with at least three biological replicates. Data are presented as the mean ± 95% confidence intervals (CI). For the real-time qPCR analyses, cartilages from at least three independent mice were used and pooled for each group. Individual experiments were performed in triplicate as technical replicates. Real-time qPCR data are presented as the mean ± standard deviation (SD). Two-tailed Student’s t-test was used to determine the significance between the two groups. P value < 0.05 was considered statistically significant.

Results

Efficiency and specificity of the AcanCreERT2 transgene within murine joints

To determine whether Notch signaling within postnatal chondrocytes is directly required for joint cartilage maintenance, we previously generated a conditional RBPjκ loss-of-function mouse model Col2a1CreERT2; Rbpjκf/f [10]. Following tamoxifen administration (TM, 1mg/10g body weight from postnatal days 25–29), mutant mice displayed a progressive OA-like pathology by 8 months of age even though we only observed a 50% reduction in Rbpjκ expression [10]. These data suggested that Col2a1CreERT2 activation mediates only low levels of targeted recombination in postnatal joint cartilages, and may not be an ideal model to study cartilage specific aspects of joint maintenance and OA. Therefore, we utilized the AcanCreERT2 knock-in mouse to induce Cre activity specifically within postnatal cartilages to achieve greater recombination efficiency and specificity [17]. We first generated AcanCreERT2; R26Rf/+ (R26RAcanTM) reporter mice to assess recombination efficiency and specificity following TM administration during joint development at P14-18 or at 2 months of age during articular cartilage homeostasis. X-gal staining on P20 (TM P14-18) and 8-month old (TM 2-months) R26RAcanTM reporter mouse knee sections revealed more than 80% recombination efficiency within the articular cartilage (Figure S1A,B). Interestingly, early TM injections (P14-18) demonstrated Cre activation and lineage tracing within the joint cartilages and subchondral bone marrow region (Figure S1A), while later TM injections (2-months) were more restricted to the joint cartilages (Figure S1B). These data confirm that the AcanCreERT2-mediated recombination is highly effective in targeting cartilage tissues, although specificity to the joint cartilages is compromised when tamoxifen is administered during earlier postnatal developmental periods.

We next generated AcanCreERT2; Rbpjκf/f (RBPjκAcanTM) mutant and control mice that were either induced with tamoxifen at P14-18 and harvested at 2-, 4-, and 8-months of age or induced at 2-months with tamoxifen and harvested at 8-months of age. To assess recombination efficiency of the Rbpjκ floxed alleles, we performed immunohistochemistry for RBPJκ protein on knee joint sections. Tissue sections from 2-, 4-, and 8-month old wild-type (WT) control mice showed robust immunoreactivity for RBPJκ within the articular and meniscal cartilage and bone marrow, with weaker signals within subchondral osteoblasts (Figure 1A,B). RBPjκAcanTM mutant immunostained sections showed a near complete depletion of RBPJκ at all time points within the joint cartilages regardless of the timing of the TM administration, while maintaining a strong signal in bone marrow and some expression in subchondral osteoblasts (Figure 1A,B). Interestingly, 8-month old RBPjκAcanTM mutant sections showed an invasion of non-targeted (RBPJκ positive) fibrotic cells often extending from regions of the synovium and lining damaged and fibrous regions of the articular or meniscal cartilages (Figure 1A,B, S2A). These non-targeted fibrotic cells were only observed in 8 month old RBPjκAcanTM mutant sections regardless of whether the TM was administered at P14-18 during development or at 2-months during joint homeostasis. Collectively, these data demonstrate that the AcanCreERT2 mouse line is highly efficient at recombining Rbpjκ floxed alleles within joint cartilages regardless of the timing for TM inductions.

Figure 1
Targeted recombination of the RBPjκ floxed alleles using the AcanCreERT2 mouse line

Loss of Notch signaling within postnatal chondrocytes results in a progressive OA-like pathology

To assess the joint integrity of RBPjκAcanTM mutants and WT control mice, we analyzed Saf-O stained histology at 4 months of age (TM administered P14-18) and 8 months of age (TM administered at either P14-18 or 2-months). Saf-O staining and OARSI scoring of knee joint sections demonstrated that RBPjκAcanTM mutants with TM administered at P14-18 showed no significant changes in the thickness or quality of the articular cartilage as compared to WT sections at 4 months of age (Figure 2A,C). However, by 8-months of age we observed many features consistent with an OA-like pathology in RBPjκAcanTM mutants. These include joint cartilage degeneration and proteoglycan depletion, which is evidenced by the loss of Saf-O staining, articular cartilage fibrosis and altered chondrocyte morphology, as well as, synovial hyperplasia with fibroblast-like synoviocytes invading the joint space and lining many of the articular cartilage and meniscus surfaces (Figure 2A, S2B). Similar histology results were observed in RBPjκAcanTM mutans at 8-months of age when tamoxifen was administered at 2-months of age (Figure 2B, S2B). OARSI scoring of combined 8-month samples shows a dramatic increase in the cartilage degeneration phenotype in RBPjκAcanTM mutants as compared to WT controls (Figure 2C). Saf-O staining also revealed an abundance of fibrotic cells often lining or in close proximity to regions of severe cartilage degeneration (Figure 2B, black asterisk and arrows, S2B red arrows). These data demonstrate that removal of RBPjκ-dependent Notch signaling within cartilage during development or at stages of homeostasis results in a progressive and age dependent OA-like pathology, and therefore highlights the importance of RBPjκ-dependent Notch signaling for maintaining healthy articular cartilage throughout adult life.

Figure 2
Cartilage-specific loss of RBPjκ-dependent Notch signaling in postnatal chondrocytes results in histological and extracellular matrix changes consistent with a progressive OA-like pathology

To characterize specific changes in the expression of ECM-related and Notch signaling molecules, we performed real-time qPCR analyses for Sox9, Col2a1, Acan1, Mmp13, Adamts5, and Hes1 on articular chondrocytes isolated form 4-month-old WT and RBPjκAcanTM mutant mice (Figure 3A). Gene expression analyses revealed a strong reduction RBPjκ-dependent Notch signaling as assessed by Hes1 target gene expression (50% decreased), reduced expression of major chondrogenic genes including Sox9 (78% decreased), Col2a1 (20% decreased) and Acan1 (19% decreased), and induced expression of major matrix degradation enzymes, such as Mmp13 (13.7 fold increased) and Adamts5 (1.5 fold increased) (Figure 3A). Immunohistochemistry (IHC) analyses for SOX9, COL2A1, COL10A1, COL3A1, MMP13 and NITEGE on 8-month-old RBPjκAcanTM mutant knee sections were consistent with the gene expression profiles. Significant depletion of SOX9 and COL2A1 were observed in the mutant articular cartilage, especially in regions of severe fibrosis (Figure 3B). COL10A1 expression, which is normally localized in the deep zone chondrocytes and up-regulated in the OA cartilage, was moderately enhanced in the superficial zone of the mutant articular cartilage (Figure 3B), indicating an inappropriate distribution of hypertrophic chondrocytes. Consistent with the fibrosis phenotype observed via ABH/OG staining, enhanced signals of the fibrosis marker COL3A1 was evident in the degenerative and fibrotic regions of the mutant cartilage, where increased MMP13 and NITEGE signals were also observed (Figure 3B). To better understand the mechanism of cartilage loss, we performed TUNEL staining on both WT and RBPjκAcanTM mutant knee sections at 8-months of age (TM administered at 2-months). A trend for increased apoptosis was observed in the mutant mice, especially in the intermediated and deep zone of the articular cartilage (Figure S3A). At 8-months of age, the number of TUNEL positive cells was significantly increased by 30% in the RBPjκAcanTM mutant knee section (Figure S3B). All of the molecular and ECM changes identified in RBPjκAcanTM mutants are consistent with a progressive OA-like pathology, which is caused by the cartilage-specific loss of RBPjκ-dependent Notch signaling in postnatal joint chondrocytes.

Figure 3
Cartilage-specific loss of RBPjκ-dependent Notch signaling in postnatal chondrocytes results in molecular and extracellular matrix changes consistent with a progressive OA-like pathology

Loss of Notch signaling within postnatal chondrocytes results in the recruitment of non-targeted fibrotic cells into the articular cartilage that may lead to cartilage degradation

As indicated previously in our IHC and histological assessments, at least one to several layers of fibroblast-like cells were often located on the surface of or adjacent to degenerating articular and meniscal cartilages within the RBPjκAcanTM mutants (Figures 1, ,2,2, ,3,3, S2). High magnification microscopic analyses of knee joint sections of 8-month old RBPjκAcanTM mutants (TM administered at 2-months) using Trichrome staining revealed an abundance of fibrotic cells lining damaged cartilage regions that also stained red for non-cartilaginous collagen fibers (Figure 4). As indicated previously (Figure 1, S2A), these fibrotic cells express an abundance of RBPjκ protein, while the surrounding cartilage in RBPjκAcanTM mutants was devoid of RBPjκ protein expression. Therefore, this population of fibrotic cells, which were not targeted by the Cre recombinase, likely represents an invasive cell type originating from non-cartilaginous tissues within the knee joint, and is often found connecting to the synovium at peripheral regions of the joint.

Figure 4
Cartilage-specific loss of RBPjκ-dependent Notch signaling in postnatal chondrocytes results in recruitment of non-targeted fibrotic cells into the articular cartilage that may lead to cartilage degradation

To understand how these invading fibrotic cells affect the mutant cartilage, we performed IHC analyses for MMP13 and NITEGE on the 8-month old RBPjκAcanTM mutant knee sections. We observed a local increase in MMP13 and NITEGE immunoreactivity specifically within the fibrous cells and the fibrotic regions of the RBPjκAcanTM mutant cartilages adjacent to the invading cells (Figure 4, green arrow heads and red arrow heads), which was consistent with our prior observations (Figure 3B). These data suggest that loss of Notch signaling in postnatal cartilages results in the recruitment of non-targeted fibrotic cells onto the articular cartilage and within the joint space that may lead to articular cartilage degradation, and contribute to a significant portion of the age-related OA-like pathology observed in RBPjκAcanTM mutant mice.

To address potential mechanisms by which fibrotic and degenerative cells were recruited into the joints of RBPjκAcanTM mutant mice, we performed IHCs for a panel of chemokines on knee joint sections from RBPjκAcanTM mutant and WT control mice. Here we show that chemokines previously implicated in OA and other joint diseases[2529], such as Chemokine C-C motif Ligand 20 (CCL-20) and Stromal Cell Derived Factor 1 (SDF-1), were differentially expressed in the joint cartilages of RBPjκAcanTM mutant cartilages only at 8-months of age when there is an obvious OA-like phenotype. Specifically, IHC analyses showed very low levels of CCL-20 and SDF-1 expression in 4-month old WT and RBPjκAcanTM mutant knee joint sections (Figure S4A). However, RBPjκAcanTM mutant knee sections showed high immunoreactivity for both CCL-20 and SDF-1 within the joint cartilages at 8-months of age (Figure S4B). Therefore, it is likely that the up-regulation of specific chemokines induces the invasion of fibroblast-like cells from the synovium into the joint space and promotes adherence to and destruction of the articular and meniscal cartilages.

Loss of Notch signaling within subchondral osteoblasts does not affect joint cartilage development and maintenance

We next generated a RBPjκ loss-of-function mouse model using the Col1a1Cre (2.3kb) transgene, Col1a1Cre (2.3kb); Rbpjκf/f (RBPjκCol1), to examine the role of Notch signaling in subchondral osteoblasts and how it may contribute to normal articular cartilage development and maintenance. Western Blot analysis performed on total proteins extracted from femoral and tibial osteoblasts of 2-month old RBPjκCol1 mutants and Cre- control mice (WT) demonstrated that RBPjκ was efficiently knocked down in RBPjκCol1 mutant osteoblasts (Figure S5A). Furthermore, lineage tracing analyses using a Col1a1Cre(2.3kb); R26Tomatof/+ (R26TomatoCol1) reporter mouse demonstrates the specificity of the gene targeting (Tomato fluorescence) to subchondral, trabecular, and cortical osteoblasts and osteocytes, without targeting the cartilage, meniscus, or synovial tissues (Figure S5B). Joint architecture analyses using MicroCT was performed on of 2- and 8-month old RBPjκCol1 mutant and control bones. Consistent with previous studies [30, 31], osteoblast-specific deletions of RBPjκ did not cause a bone phenotype by 2 months-of-age (Figure 5A). Here we further demonstrated that no significant bone phenotype is observed in 8-month-old RBPjκCol1 mutants (Figure 5A). In addition, comparable bone volume was observed in RBPjκCol1 mutant and WT mice at both 2- and 8-months of age (Figure 5B). Joint integrity was analyzed via ABH/OG and Saf-O stained histology at 2-weeks, 2-months, and 8-months of age. Normal secondary ossification center formation and articular cartilage development were observed in 2-week old RBPjκCol mutants (Figure 6A), and comparable proteoglycan staining of cartilage ECM in both RBPjκCol1 mutant and WT mice were evident at 2- and 8-months of age (Figure 6A,B). OARSI scoring performed on 8-month old RBPjκCol1 mutant and WT knee joint sections indicates that RBPjκCol1 mutant mice have no detectable OA-related cartilage changes (Figure 6B). Additionally, a normal expression pattern of major cartilage components, COL2A1 and COL10A1, were observed in both 2- and 8-month old RBPjκCol1 mutant and WT mice (Figure S6A,B). These histological and molecular analyses confirmed that loss of RBPJκ-dependent Notch signaling in subchondral osteoblasts does not affect normal articular cartilage development or joint maintenance.

Figure 5
Osteoblast-specific loss of RBPjκ-dependent Notch signaling does not lead to a bone phenotype
Figure 6
RBPjκ-dependent Notch signaling in osteoblasts is dispensable for joint cartilage development and maintenance

Discussion

In this study we provide the first genetic evidence that RBPjκ-dependent Notch signaling in postnatal joint chondrocytes, but not subchondral osteoblasts, is required for articular cartilage and joint maintenance. This study establishes that loss of RBPjκ-dependent Notch signaling in postnatal joint cartilages resulted in a progressive OA-like pathology, and triggered the recruitment of non-targeted fibrotic cells into the articular cartilage and joint space potentially due to the up-regulation of specific joint disease related chemokines. These invasive fibrotic cells expressed degenerative enzymes and likely contribute to much of the cartilage degradation that is observed. We also demonstrated that loss of Notch signaling in subchondral osteoblasts did not affect normal cartilage development or joint maintenance. Collectively, these data reveal a requisite role for Notch signaling within postnatal joint chondrocytes, but not subchondral osteoblasts, to maintain cartilage and joint integrity throughout adult life.

The role of Notch signaling in cartilage homeostasis and OA development is complex. On one hand, Notch signaling components are widely expressed in murine and human adult articular cartilage [12, 32, 33], and suppression of the Notch pathway in all joint tissues or postnatal chondrocytes results in OA-like pathology [10] (this study), indicating that Notch signaling is a critical regulator of postnatal cartilage and joint maintenance. On the other hand, the expression of Notch signaling molecules is significantly up-regulated in OA cartilage [3236] and the short-term loss or reduction of cartilage-specific Notch signaling can delay cartilage degeneration [35], indicating a role for abnormal or pathogenic Notch signaling in regulating OA on-set and progression. Since the Notch pathway naturally operates in an oscillatory manner as a non-amplifying signal that terminates in a negative feedback loop [3739], it is reasonable to believe that completely removing that signal or abnormally amplifying that signal could result in pathological tissue events. Different levels of Notch signaling have been shown to regulate dynamic biological events, including quiescence, proliferation and differentiation in various tissues [37, 39, 40]. Transient activation of Notch signaling can induce Sox9 expression in vitro [41], and we also found that intermittent Notch activation in articular cartilage promoted cartilage ECM synthesis and induced the expression of anabolic genes including Sox9, Col2a1 and Acan in vivo [42]. On the contrary, when Notch signaling is hyper activated in the context of OA, it simultaneously suppresses the expression of chondrogenic genes and stimulates the expression of catabolic factors, such as MMP13 and ADAMTS5 [10, 35, 42]. Recent work has suggested that the OA process could be modulated by HES1, an important target of Notch signaling that can induce the expression of catabolic factors including Mmp13, Adamts5, Il6 and Il1rl1 in cooperation with calcium/calmodulin-dependent protein kinase 2 (CaMK2) [43]. Collectively, these data suggest that an appropriate balance of Notch signaling in articular cartilage is required to maintain normal cartilage and joint integrity. Only when Notch signaling is significantly impaired or pathologically activated would articular cartilage become prone to damage, due to either an inability to maintain appropriate ECM synthesis, or an excessive degradation result from induced catabolic factors, respectively.

Articular chondrocytes are “maturationally arrested” cells that are metabolically active to maintain the ECM synthesis and turnover but do not proliferate significantly [44]. We have previously reported that suppression of Notch signaling lead to loss of articular chondrocyte morphology in vivo and accelerated chondrocyte de-differentiation in vitro [10]. Here we demonstrated that these affected chondrocytes may ultimately undergo apoptosis and contribute to cartilage loss and degradation (Figure S3). Furthermore, we discovered that loss of Notch signaling in articular chondrocytes resulted in the up-regulation of joint destructive chemokines, such as CCL-20 and SDF-1, that likely promote the recruitment of fibrotic cells from the synovium or periphery of the joint into the joint space and these cells likely contribute to the joint cartilage degeneration through the expression of catabolic enzymes and/or cell death inducing agents (Figure S4)[2529]. We speculate that the most plausible cell sources are the synoviocytes from the expanded and hyperplastic synovial tissue. Synoviocytes are metabolically highly active cells that can nourish chondrocytes via the synovial fluid and remove metabolites and matrix degradation products under physiological condition. During the OA process, cartilage breakdown products are phagocytosed by synovial cells and amplify synovial inflammation, which in return stimulates synoviocytes to produce more catabolic and pro-inflammatory factors responsible for cartilage breakdown, thereby creating a vicious positive feedback loop [45].

Notch signaling also plays a critical regulatory role in osteoblast-lineage cells. Previous studies have shown that removal of both catalytic subunits of the γ-secretase complex, PS1 and PS2, or both Notch1 and Notch2 in embryonic limb mesenchyme lead to excessive bone formation in adolescent mice, but precipitous bone loss in older mice because of the diminution of the progenitor pool [11]. However, deletion of Rbpjκ in more mature osteoblastic cells with either OsxCre or Col1Cre (2.3kb) transgene did not cause any obvious effect in the mutant mice at 8- or 21-weeks of age [31]. These data suggest that Notch signaling maintains mesenchymal progenitors by suppressing osteoblasts differentiation, but RBPjκ-dependent Notch signaling may not play a major role in more committed osteoblast-lineage cells. Consistent with these findings, our data further showed that loss of RBPjκ-dependent Notch signaling in subchondral osteoblasts did not affect normal articular cartilage development or joint maintenance, and caused no significant change in bone mass by the age of 8-months (Figure 5 and and6).6). In conclusion, our findings have identified RBPjκ-dependent Notch signaling in postnatal joint chondrocytes, but not subchondral osteoblasts, as a novel and essential regulator involved in joint maintenance and articular cartilage homeostasis, which triggered chemokine expression and the recruitment of non-targeted fibrotic cells into the articular cartilage that likely contribute to a significant portion of the age-related OA-like pathology.

Supplementary Material

1

Supplementary Figure 1. Targeted recombination of the AcanCreERT2 transgene with postnatal TM administration:

X-gal staining of (A) R26RAcanTM knee sections at P20 (TM administered from P14-18) and (B) R26RAcanTM knee sections at 8-months of age (TM administered at 2-months). LacZ activity can be observed in nearly all articular and meniscal cartilage chondrocytes, with some additional activity observed in subchondral bone of early time-point TM administration. AC, articular cartilage; M, meniscus; SB: subchondral bone. Scale bars, 100μm. N=3 and N=4 for P20 and 8-month old RBPjκAcanTM mutant mice, respectively.

2

Supplementary Figure 2. Cartilage-specific loss of RBPjκ-dependent Notch signaling in 8-month old animals shows invading non-targeted fibrotic cells on the superficial cartilage surfaces:

(A) High magnification for Figure 1A-B of 8 month old mutant mice showing Rbpjκ IHC. (B) High magnification for Figure 2 of 8-month old mutant showing safranin O staining. (C) High magnification for Supplementary Figure 4 of 8-month mutant showing CCL-20 and SDF-1 IHC. Enlarged area is labelled in the yellow box, red arrows indicated non-targeted fibrotic cells on the cartilage surfaces.

3

Supplementary Figure 3. Cartilage-specific loss of RBPjκ-dependent Notch signaling in postnatal chondrocytes results in increased chondrocyte apoptosis:

(A) TUNEL staining of WT and RBPjκAcanTM mutant knee sections at 8-months of age. White dashed lines indicate articular cartilage surface. Scale bars, 200μm. (B) Quantification of the total number of TUNEL positive cells in the articular cartilage. Bars represent means with 95% CI. “*” denotes P=0.031. N=3 and 5 for WT and RBPjκAcanTM mutant mice respectively.

4

Supplementary Figure 4. Cartilage-specific loss of RBPjκ-dependent Notch signaling in early postnatal chondrocytes results in enhanced chemokine expression:

IHC for CCL-20 and SDF-1 on (A) 4-month old WT and RBPjκAcanTM mutant knee sections (TM administered from P14-18) and (B) 8-month old WT and RBPjκAcanTM mutant knee sections (TM administered at 2-months). Red arrows indicate areas of enhance CCL-20 and SDF-1 immunoreactivity in 8-month old RBPjκAcanTM mutant knee sections. N=5 for WT and N=3 for RBPjκAcanTM mutant mice at 4-months of age (TM administered at P14-18). N=4 for WT and N=4 for RBPjκAcanTM mutant mice at 8-months of age (TM administered at 2-months).

5

Supplementary Figure 5. RBPjκ protein is efficiently knocked down in the RBPjκCol1 mutant osteoblasts:

(A) Western blot analyses for RBPjκ in proteins extracted from femurs and tibiae of WT and RBPjκCol1 mutant mice at 2 months of age. Results are representative of three independent experiments. (B) Tomato immunofluorescence (red) on R26-TomatoCol1 knee joint sections shows targeting specificity to osteoblasts and osteocytes using the Col1a1Cre(2.3kb). AC, articular cartilage; M, meniscus; GP: growth plate cartilage; S: synovium. Yellow arrows identify osteoblast cells lining the cortical bone. Results are representative of three independent experiments.

6

Supplementary Figure 6. Osteoblast-specific loss of RBPjκ-dependent Notch signaling does not affect the expression of major extracellular matrix molecules:

IHC analyses of COL2A1 and COL10A1 performed on (A) 2-month-old and (B) 8-month-old WT and RBPjκCol1 mutant knee sections. Scale bars, 200μm. N=4 for WT and N=6 for RBPjκCol1 mutant mice at 2-months of age. N=4 for WT and N=4 for RBPjκCol1 mutant mice at 8- months of age.

Acknowledgments

We thank Dr. Benoit de Crombrugghe and Dr. Henry M. Kronenberg for providing important mouse strains. We would like to gratefully acknowledge the technical expertise and assistance of Sarah Mack, Kathy Maltby, Ashish Thomas, and Michael Thullen within the Histology, Biochemistry, and Molecular Imaging Core and the Biomechanics and Multimodal Tissue Imaging Core in the Center for Musculoskeletal Research at the University of Rochester Medical Center.

Role of the funding source

This work was supported in part by the following United States National Institute of Health grants: R01 grants (AR057022 and AR063071 to MJH), R21 grant (AR059733 to MJH), a P50 Center of Research Translation grant (AR054041 to RJO), a P30 Core Center grant (AR061307), and departmental funds from the Department of Orthopaedic Surgery at Duke University School of Medicine.

Footnotes

Author contributions

Conception and study design: MJH and ZL; Study conduct: ZL, YR, AJM, and CW; Analysis and interpretation of the data: ZL, YR, AJM, CW, MJZ, RJO, and MJH; Drafting and/or editing manuscript: ZL, YR, AJM, CW, MJH, RJO, and MJH; Approving final version of the manuscript: ZL, YR, AJM, CW, MJH, RJO, and MJH.

Competing Interest Statement

The authors declare no conflicts of interest.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

1. Felson DT, Lawrence RC, Dieppe PA, Hirsch R, Helmick CG, Jordan JM, et al. Osteoarthritis: new insights. Part 1: the disease and its risk factors. Ann Intern Med. 2000;133:635–646. [PubMed]
2. Goldring MB, Goldring SR. Osteoarthritis. J Cell Physiol. 2007;213:626–634. [PubMed]
3. Sandell LJ. Etiology of osteoarthritis: genetics and synovial joint development. Nat Rev Rheumatol. 2012;8:77–89. [PubMed]
4. Bos SD, Slagboom PE, Meulenbelt I. New insights into osteoarthritis: early developmental features of an ageing-related disease. Curr Opin Rheumatol. 2008;20:553–559. [PubMed]
5. Reynard LN, Loughlin J. Insights from human genetic studies into the pathways involved in osteoarthritis. Nat Rev Rheumatol. 2013;9:573–583. [PubMed]
6. Wu L, Huang X, Li L, Huang H, Xu R, Luyten W. Insights on biology and pathology of HIF-1alpha/-2alpha, TGFbeta/BMP, Wnt/beta-catenin, and NF-kappaB pathways in osteoarthritis. Curr Pharm Des. 2012;18:3293–3312. [PubMed]
7. Dong Y, Jesse AM, Kohn A, Gunnell LM, Honjo T, Zuscik MJ, et al. RBPjkappa-dependent Notch signaling regulates mesenchymal progenitor cell proliferation and differentiation during skeletal development. Development. 2010;137:1461–1471. [PubMed]
8. Kohn A, Dong Y, Mirando AJ, Jesse AM, Honjo T, Zuscik MJ, et al. Cartilage-specific RBPjkappa-dependent and -independent Notch signals regulate cartilage and bone development. Development. 2012;139:1198–1212. [PubMed]
9. Mead TJ, Yutzey KE. Notch pathway regulation of chondrocyte differentiation and proliferation during appendicular and axial skeleton development. Proc Natl Acad Sci U S A. 2009;106:14420–14425. [PubMed]
10. Mirando AJ, Liu Z, Moore T, Lang A, Kohn A, Osinski AM, et al. RBP-Jkappa-dependent Notch signaling is required for murine articular cartilage and joint maintenance. Arthritis Rheum. 2013;65:2623–2633. [PMC free article] [PubMed]
11. Hilton MJ, Tu X, Wu X, Bai S, Zhao H, Kobayashi T, et al. Notch signaling maintains bone marrow mesenchymal progenitors by suppressing osteoblast differentiation. Nat Med. 2008;14:306–314. [PMC free article] [PubMed]
12. Hayes AJ, Dowthwaite GP, Webster SV, Archer CW. The distribution of Notch receptors and their ligands during articular cartilage development. J Anat. 2003;202:495–502. [PubMed]
13. Dowthwaite GP, Bishop JC, Redman SN, Khan IM, Rooney P, Evans DJ, et al. The surface of articular cartilage contains a progenitor cell population. J Cell Sci. 2004;117:889–897. [PubMed]
14. Kopan R, Ilagan MX. The canonical Notch signaling pathway: unfolding the activation mechanism. Cell. 2009;137:216–233. [PMC free article] [PubMed]
15. Hori K, Sen A, Artavanis-Tsakonas S. Notch signaling at a glance. J Cell Sci. 2013;126:2135–2140. [PubMed]
16. Artavanis-Tsakonas S, Rand MD, Lake RJ. Notch signaling: cell fate control and signal integration in development. Science. 1999;284:770–776. [PubMed]
17. Henry SP, Jang CW, Deng JM, Zhang Z, Behringer RR, de Crombrugghe B. Generation of aggrecan-CreERT2 knockin mice for inducible Cre activity in adult cartilage. Genesis. 2009;47:805–814. [PMC free article] [PubMed]
18. Miao D, He B, Jiang Y, Kobayashi T, Soroceanu MA, Zhao J, et al. Osteoblast-derived PTHrP is a potent endogenous bone anabolic agent that modifies the therapeutic efficacy of administered PTH 1-34. J Clin Invest. 2005;115:2402–2411. [PMC free article] [PubMed]
19. Han H, Tanigaki K, Yamamoto N, Kuroda K, Yoshimoto M, Nakahata T, et al. Inducible gene knockout of transcription factor recombination signal binding protein-J reveals its essential role in T versus B lineage decision. Int Immunol. 2002;14:637–645. [PubMed]
20. Soriano P. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet. 1999;21:70–71. [PubMed]
21. Madisen L, Zwingman TA, Sunkin SM, Oh SW, Zariwala HA, Gu H, et al. A robust and high-throughput Cre reporting and characterization system for the whole mouse brain. Nat Neurosci. 2010;13:133–140. [PMC free article] [PubMed]
22. Hilton MJ, Tu X, Long F. Tamoxifen-inducible gene deletion reveals a distinct cell type associated with trabecular bone, and direct regulation of PTHrP expression and chondrocyte morphology by Ihh in growth region cartilage. Dev Biol. 2007;308:93–105. [PMC free article] [PubMed]
23. Wu Q, Kim KO, Sampson ER, Chen D, Awad H, O'Brien T, et al. Induction of an osteoarthritis-like phenotype and degradation of phosphorylated Smad3 by Smurf2 in transgenic mice. Arthritis Rheum. 2008;58:3132–3144. [PMC free article] [PubMed]
24. Glasson SS, Chambers MG, Van Den Berg WB, Little CB. The OARSI histopathology initiative - recommendations for histological assessments of osteoarthritis in the mouse. Osteoarthritis Cartilage. 2010;18(Suppl 3):S17–23. [PubMed]
25. Kanbe K, Chiba J, Inoue Y, Taguchi M, Yabuki A. SDF-1 and CXCR4 in synovium are associated with disease activity and bone and joint destruction in patients with rheumatoid arthritis treated with golimumab. Mod Rheumatol. 2015:1–15. [PubMed]
26. Kanbe K, Takemura T, Takeuchi K, Chen Q, Takagishi K, Inoue K. Synovectomy reduces stromal-cell-derived factor-1 (SDF-1) which is involved in the destruction of cartilage in osteoarthritis and rheumatoid arthritis. J Bone Joint Surg Br. 2004;86:296–300. [PubMed]
27. Alaaeddine N, Antoniou J, Moussa M, Hilal G, Kreichaty G, Ghanem I, et al. The chemokine CCL20 induces proinflammatory and matrix degradative responses in cartilage. Inflamm Res. 2015;64:721–731. [PubMed]
28. Alaaeddine N, Hilal G, Baddoura R, Antoniou J, Di Battista JA. CCL20 stimulates proinflammatory mediator synthesis in human fibroblast-like synoviocytes through a MAP kinase-dependent process with transcriptional and posttranscriptional control. J Rheumatol. 2011;38:1858–1865. [PubMed]
29. Lisignoli G, Manferdini C, Codeluppi K, Piacentini A, Grassi F, Cattini L, et al. CCL20/CCR6 chemokine/receptor expression in bone tissue from osteoarthritis and rheumatoid arthritis patients: different response of osteoblasts in the two groups. J Cell Physiol. 2009;221:154–160. [PubMed]
30. Tao J, Chen S, Yang T, Dawson B, Munivez E, Bertin T, et al. Osteosclerosis owing to Notch gain of function is solely Rbpj-dependent. J Bone Miner Res. 2010;25:2175–2183. [PMC free article] [PubMed]
31. Tu X, Chen J, Lim J, Karner CM, Lee SY, Heisig J, et al. Physiological notch signaling maintains bone homeostasis via RBPjk and Hey upstream of NFATc1. PLoS Genet. 2012;8:e1002577. [PMC free article] [PubMed]
32. Ustunel I, Ozenci AM, Sahin Z, Ozbey O, Acar N, Tanriover G, et al. The immunohistochemical localization of notch receptors and ligands in human articular cartilage, chondroprogenitor culture and ultrastructural characteristics of these progenitor cells. Acta Histochem. 2008;110:397–407. [PubMed]
33. Grogan SP, Miyaki S, Asahara H, D'Lima DD, Lotz MK. Mesenchymal progenitor cell markers in human articular cartilage: normal distribution and changes in osteoarthritis. Arthritis Res Ther. 2009;11:R85. [PMC free article] [PubMed]
34. Karlsson C, Brantsing C, Egell S, Lindahl A. Notch1, Jagged1, and HES5 are abundantly expressed in osteoarthritis. Cells Tissues Organs. 2008;188:287–298. [PubMed]
35. Hosaka Y, Saito T, Sugita S, Hikata T, Kobayashi H, Fukai A, et al. Notch signaling in chondrocytes modulates endochondral ossification and osteoarthritis development. Proc Natl Acad Sci U S A. 2013;110:1875–1880. [PubMed]
36. Mahjoub M, Sassi N, Driss M, Laadhar L, Allouche M, Hamdoun M, et al. Expression patterns of Notch receptors and their ligands in human osteoarthritic and healthy articular cartilage. Tissue Cell. 2012;44:182–194. [PubMed]
37. Ninov N, Borius M, Stainier DY. Different levels of Notch signaling regulate quiescence, renewal and differentiation in pancreatic endocrine progenitors. Development. 2012;139:1557–1567. [PubMed]
38. Kageyama R, Ohtsuka T, Shimojo H, Imayoshi I. Dynamic Notch signaling in neural progenitor cells and a revised view of lateral inhibition. Nat Neurosci. 2008;11:1247–1251. [PubMed]
39. Shimojo H, Ohtsuka T, Kageyama R. Oscillations in notch signaling regulate maintenance of neural progenitors. Neuron. 2008;58:52–64. [PubMed]
40. Kageyama R, Niwa Y, Shimojo H, Kobayashi T, Ohtsuka T. Ultradian oscillations in Notch signaling regulate dynamic biological events. Curr Top Dev Biol. 2010;92:311–331. [PubMed]
41. Haller R, Schwanbeck R, Martini S, Bernoth K, Kramer J, Just U, et al. Notch1 signaling regulates chondrogenic lineage determination through Sox9 activation. Cell Death Differ. 2012;19:461–469. [PMC free article] [PubMed]
42. Liu Z, Chen J, Mirando AJ, Wang C, Zuscik MJ, O'Keefe RJ, et al. A dual role for NOTCH signaling in joint cartilage maintenance and osteoarthritis. Sci Signal. 2015;8:ra71. [PMC free article] [PubMed]
43. Sugita S, Hosaka Y, Okada K, Mori D, Yano F, Kobayashi H, et al. Transcription factor Hes1 modulates osteoarthritis development in cooperation with calcium/calmodulin-dependent protein kinase 2. Proc Natl Acad Sci U S A. 2015;112:3080–3085. [PubMed]
44. Dreier R. Hypertrophic differentiation of chondrocytes in osteoarthritis: the developmental aspect of degenerative joint disorders. Arthritis Res Ther. 2010;12:216. [PMC free article] [PubMed]
45. Sellam J, Berenbaum F. The role of synovitis in pathophysiology and clinical symptoms of osteoarthritis. Nat Rev Rheumatol. 2010;6:625–635. [PubMed]