Skip to main content
NCBI home page
Endocrinology logoLink to Endocrinology
. 2016 Oct 12;157(12):4602–4614. doi: 10.1210/en.2016-1548

FGF2 High Molecular Weight Isoforms Contribute to Osteoarthropathy in Male Mice

Patience Meo Burt 1, Liping Xiao 1, Caroline Dealy 1, Melanie C Fisher 1, Marja M Hurley 1,✉
PMCID: PMC5133359  PMID: 27732085

Abstract

Humans with X-linked hypophosphatemia (XLH) and Hyp mice, the murine homolog of the disease, develop severe osteoarthropathy and the precise factors that contribute to this joint degeneration remain largely unknown. Fibroblast growth factor 2 (FGF2) is a key regulatory growth factor in osteoarthritis. Although there are multiple FGF2 isoforms the potential involvement of specific FGF2 isoforms in joint degradation has not been investigated. Mice that overexpress the high molecular weight FGF2 isoforms in bone (HMWTg mice) phenocopy Hyp mice and XLH subjects and Hyp mice overexpress the HMWFGF2 isoforms in osteoblasts and osteocytes. Given that Hyp mice and XLH subjects develop osteoarthropathies we examined whether HMWTg mice also develop knee joint degeneration at 2, 8, and 18 mo compared with VectorTg (control) mice. HMWTg mice developed spontaneous osteoarthropathy as early as age 2 mo with thinning of subchondral bone, osteophyte formation, decreased articular cartilage thickness, abnormal mineralization within the joint, increased cartilage degradative enzymes, hypertrophic markers, and angiogenesis. FGF receptors 1 and 3 and fibroblast growth factor 23 were significantly altered compared with VectorTg mice. In addition, gene expression of growth factors and cytokines including bone morphogenetic proteins, Insulin like growth factor 1, Interleukin 1 beta, as well as transcription factors Sex determining region Y box 9, hypoxia inducible factor 1, and nuclear factor kappa B subunit 1 were differentially modulated in HMWTg compared with VectorTg. This study demonstrates that overexpression of the HMW isoforms of FGF2 in bone results in catabolic activity in joint cartilage and bone that leads to osteoarthropathy.

X-linked hypophosphatemia (XLH), the most common form of inherited vitamin D-resistant rickets, which affects 1 in 20 000 people, leads to renal phosphate wasting, inappropriately high levels of circulating fibroblast growth factor 23 (FGF23), rickets, growth retardation, osteomalacia, and osteoarthropathies (1). The most common symptom in adults is joint pain, which can be attributed to the development of enthesopathy and osteoarthritis, both hallmarks of XLH, which include thinning of articular cartilage and sclerosis of subchondral bone (2, 3). The Hyp mouse, a murine homolog of XLH, is known to develop the same signs of humans with XLH, including hypophosphatemia and elevated levels of FGF23 (4). More recently, the Hyp mouse was found to develop enthesopathy, which included mineralization of the tendon and ligament insertion sites, as well as, osteophyte formation (5). Degenerative osteoarthropathy was also prominent in these mice, which was characterized by decreased articular cartilage thickness, increased chondrocyte alkaline phosphatase activity, defective mineralization, and vascular invasion of the cartilage (6).

We have previously developed transgenic mice that phenocopy the Hyp mouse and subjects with XLH, by overexpressing the high molecular weight (HMW) isoforms of fibroblast growth factor 2 (FGF2) under a Col3.6 promoter (HMWTg mice) (7). The FGF2 gene encodes multiple protein isoforms with varying molecular weights. Humans have three HMW FGF2 isoforms (22, 22.5, 24 kDa), and rodents have two HMW FGF2 isoforms (21, 22 kDa). These HMW FGF2 isoforms possess a nuclear localization sequence that enables the proteins to function in an intracrine manner. Humans and rodents also have one low molecular weight (LMW) FGF2 isoform that is exported from cells, which functions in an autocrine and/or paracrine manner (8, 9). The bone phenotype of HMWTg mice was previously fully characterized and it was also determined that HMWTg mice develop dwarfism, decreased bone mineral density, rickets, osteomalacia, and hypophosphatemia, and display increased FGF23 in serum and bone, which are the traits in which the HMWTg mice phenocopy Hyp mice (7). Also, Hyp mice were previously reported to overexpress the HMW FGF2 isoforms (7). Given that the Hyp mouse, as well as, humans with XLH develop osteoarthropathies (2, 6) the aim of our study was to determine whether HMWTg mice also develop joint degeneration.

This study is of particular importance given that recently FGF2 has received considerable attention in the regulation of cartilage and joint homeostasis (10, 11). There is conflicting evidence that demonstrates both catabolic and anabolic potential of FGF2 in the joint (10–13) and no studies, as of yet, addressed the specific FGF2 isoforms in the development of osteoarthropathy.

Materials and Methods

Experimental animals

All animal protocols were approved by the UConn Health Institute of Animal Care and Use Committee.

We have previously described in detail the generation of HMWTg and control/vector transgenic mice (VectorTg) (7). Briefly, the Col3.6-HMW-IRES-green fluorescent protein-sapphire (GFPsaph) construct was created by inserting HMW isoforms of human Fgf2 (22, 23, and 24 kDa) cDNA in previously made Col3.6-CAT-IRES-GFPsaph in place of a chloramphenicol acetyltransferase fragment. This expression construct is able to concurrently overexpress HMW and GFPsaph from a single bicistronic mRNA. The Col3.6-IRES/GFPsaph (Vector), which overexpresses the transgene cassette without the additional FGF2 coding sequences, was created as a control. Col3.6-CAT-IRES-GFPsaph or Col3.6-IRES/GFPsaph construct inserts were released via digestion by AseI and AflII. The Gene Targeting and Transgenic Facility at UCONN Health performed microinjections in the pronuclei of fertilized oocytes. The individual transgenic mouse lines were established by breeding founder mice of the F2 generation of the FVBN strain with wild-type mice. Male homozygous VectorTg and HMWTg mice at 2 (n = 10–11/group), 8 (n = 7–8/group), and 18 (n = 4–7/group) months of age were used in this study. We utilized two independent lines of HMWTg mice, Line 203 (for 2- and 8-month-old mice) and Line 204 (for 18-month-old mice), which were previously extensively characterized (7). Male homozygous mice that overexpress the human LMW, 18-kDa isoform, of FGF2 (LMWTg mice) were created in a similar manner as described above (14) and were compared with the VectorTg controls at 19 months of age (n = 4/group) for development of osteoarthropathy.

Radiology and microcomputed tomography

Digital radiographs of murine knee joints were obtained with a SYSTEM MX 20 from Faxitron X-ray Corporation. Imaging of left knee architecture was performed using ex vivo microcomputed tomography (μCT40, ScanCo Medical AG). The region of subchondral trabecular bone of the epiphysis in the femur and tibia were analyzed by 3-dimensional (3D) microcomputed tomography (μCT) for the following morphometric parameters: bone volume/tissue volume (BV/TV), trabecular thickness (Tb.Th), trabecular spacing (Tb.Sp), and trabecular number (Tb.N).

Histology

Whole knee joints were dissected and fixed in 4% paraformaldehyde for 6 days followed by decalcification with 14% EDTA solution. Specimens were processed for paraffin embedding in a frontal orientation and 7-μm alternate sections were obtained. After sections were deparaffinized and rehydrated, Safranin-O staining of glycosaminoglycans was performed using 0.1% aqueous Safranin-O, followed by counterstaining using Weigert's Iron Hematoxylin and 0.02% aqueous Fast Green. For alkaline phosphatase and von Kossa staining, the knees were formalin fixed, cryomatrix (Thermo Electron Corp.) embedded, and sectioned at 6-μm. Alkaline phosphatase staining was performed by incubating tissue with 100mM Tris-HCl buffer (pH, 8.2) for 10 minutes, followed by incubation with Vector Blue Alkaline Phosphatase Substrate (Vector Laboratories) for 30 minutes.

Immunohistochemistry and RNAscope

Immunohistochemical staining was performed using ImmunoCruz ABC Staining System (Santa Cruz Biotechnology) and DAB Peroxidase Substrate kit (Vector Laboratories). Following deparaffinization and rehydration, sections were incubated at 95°C with 10mM sodium citrate buffer for 10 minutes for antigen retrieval. Once cooled to room temperature for 30 minutes, 3% hydrogen peroxide in water was added to the sections for 15 minutes. After blocking with 10% serum for 1 hour, the slides were incubated with primary antibodies in blocking buffer overnight at 4°C. Details of the following primary antibodies are in Table 1: rabbit anti-MMP13 (Abcam), rabbit anti-ADAMTS5 (Abcam), rabbit anti-FGF-2 (Santa Cruz), rabbit anti-FGFR-1 (Santa Cruz), rabbit anti-FGFR-3 (Santa Cruz), rabbit anti-VEGF (Santa Cruz), anti-FGF-23 (R & D Systems), rabbit anti-NF-kB p65 (Abcam), and antirabbit SOX9 (Abcam). Slides were then washed with tris-buffered saline containing 0.1% Tween 20 and incubated with the appropriate 1:200 biotinylated secondary antibody at room temperature for 30 minutes. Next, slides were washed and developed with DAB substrate solution, followed by counterstaining with Harris hematoxylin. RNAscope was performed using Formalin-Fixed Paraffin-Embedded Sample Preparation and Pretreatment, RNAscope 2.5 HD Detection Reagent, and the probe for mouse Col10a1 according to the manufacturer's instructions (Advanced Cell Diagnostics).

Table 1.

Antibody Table

Peptide/Protein Target RRID Name of Antibody Manufacturer/Catalog Number Species Raised (Monoclonal/Polyclonal) Dilution
MMP13 AB_776416 Anti-MMP13 antibody Abcam, ab39012 Rabbit; polyclonal 1/100
ADAMTS5 AB_2222327 Anti-ADAMTS5 antibody Abcam, ab41037 Rabbit; polyclonal 1/100
FGF-2 AB_631497 FGF-2 (147) Santa Cruz, sc-79 Rabbit; polyclonal 1/50
FGFR-3 AB_2103530 p-FGFR-3 (Tyr 724) Santa Cruz, sc-33 041 Rabbit; polyclonal 1/100
VEGF AB_2212984 VEGF (A-20) Santa Cruz, sc-152 Rabbit; polyclonal 1/100
FGF-23 AB_2104623 Anti-FGF-23 antibody R & D Systems, MAB26291 Rat; monoclonal 1/100
FGFR-1 AB_675545 Flg (C-15) Santa Cruz, sc-121 Rabbit; polyclonal 1/50
NF-κB AB_11160495 Anti-NF-κB p65 (phospho-S536) antibody Abcam, ab131109 Rabbit; polyclonal 1/100
SOX9 AB_945590 SOX9 antibody Abcam, ab59265 Rabbit; polyclonal 1/50
Open in a new tab

RNA isolation and real-time PCR

Following skin and bulk muscle removal from 2-month-old VectorTg and HMWTg mice, total RNA was extracted from whole joints including all structures of the joint, above the growth plate, using TRIzol reagent (14). RNA was reverse transcribed to cDNA using RNA to cDNA EcoDry Premix kit (Clontech Inc., a Takara Bio Company). For real time RT-qPCR analysis a Bio-Rad MyiQ system was used with iTaqTM Universal SYBR Green Supermix (Bio-Rad). The primers used for real-time PCR (Table 2) were synthesized by IDT (Integrated DNA Technologies, Inc.). The relative change in mRNA was normalized to the housekeeping gene β-actin (β-actin), which served as an internal reference for each sample. For each group the mRNA level is expressed as the fold change relative to the first sample.

Table 2.

Primers Used in qRT-PCR

Gene Forward Reverse
β-actin 5′-atggctggggtgttgaaggt-3′ 5′-atctggcaccacaccttctacaa-3'
Mmp-13 5′-ctttggcttagaggtgactgg-3′ 5′-aggcactccacatcttggttt-3′
Adamts5 5′-cctgcccacccaatggtaaa-3′ 5′-ccacatagtagcctgtgccc-3′
Col10a1 5′-gggaccccaaggacctaaag-3′ 5′-gcccaactagacctatctcacct-3′
Vegf 5′-gcacatagagagaatgagcttcc-3′ 5′-ctccgctctgaacaaggct-3′
Fgf2 5′-gtcacggaaatactccagttggt-3′ 5′-cccgttttggatccgagttt-3′
Fgf23 5′-acttgtcgcagaagcatc-3′ 5′-gtgggcgaacagtgtagaa-3′
FGFR1 5′-gactgctggagttaatacca-3′ 5′-ctggtctctcttccagggct-3′
FGFR3 5′-gttctctctttgtagactgc-3′ 5′-agtacctggcagcacca-3′
Sox9 5′-agtacccgcatctgcacaac-3′ 5′-acgaagggtctcttctcgct-3′
Nf-kb1 5′-gaaattcctgatccagacaaaaac-3′ 5′-atcacttcaatggcctctgtgtag-3′
Bmp-2 5′-agcgtcaagccaaacacaaacag-3′ 5′-ggttagtggagttcaggtggtcag-3′
Bmp-4 5′-gccggagggccaagcgtagccctaag-3′ 5′-ctgcctgatctcagcggcacccacatc-3′
Igf-1 5′-gtgagccaaagacacaccca-3′ 5′-acctctgattttccgagttgc-3′
IL-1β 5′-gcaactgttcctgaactcaact-3′ 5′-atcttttggggtccgtcaact-3′
Hif1α 5′-caagatctcggcgaagcaa-3′ 5′-ggtgagcctcataacagaagcttt-3′
Open in a new tab

Determination of articular cartilage thickness

Articular cartilage thickness was determined based on a method previously reported (15) by using Safranin-O stained sections and measuring the mean distance at the thickest point from the articular cartilage surface to the subchondral bone interface across three points per section using ImageJ. Images were taken at 20× from three separate representative sections from each knee, closest to the center of the joint, from different VectorTg and HMWTg mice (n = 3–4/group).

Statistical analysis

Results are presented as means ± SD. Student t test was used to analyze differences between groups. Differences were considered significant at P < .05, unless otherwise stated.

Results

Time course of radiological and morphometric changes in VectorTg and HMWTg knee joints

Digital radiographs were assessed for indications of osteoarthritis (OA) in the bone of VectorTg and HMWTg at 2, 8, and 18 months of age (Figure 1A). All (11 of 11) knee joints of HMWTg mice at 2 months of age showed flattening of the tibial plateau and thinning of femoral subchondral bone. In addition, the knee joints of all (eight of eight) HMWTg mice at 8 months old displayed thinning of subchondral bone, as well as, osteophyte formation mainly on the posterior tibiae. Joint destruction was apparent in 18-month-old HMWTg mice, characterized by osteophyte formation, sclerotic bone development, and narrowing of the patellofemoral space (seven of seven). Conversely, radiographs of VectorTg mice did not display evidence of skeletal joint deterioration at any age.

Figure 1.

Figure 1.

Open in a new tab

Radiographic and MicroCT analysis of subchondral bone in VectorTg and HMWTg mice. A, Sagittal digital x-ray images of HMWTg knees show flattening of tibial plateau (open arrow), thinning of subchondral bone (open arrowhead), osteophyte formation (filled arrowhead), patellofemoral space narrowing (dashed arrow), and sclerotic bone (filled arrow). N = 4–11/group. B, High-resolution μCT images of representative sagittal sections of all HMWTg samples show decreased trabecular bone in the epiphyses, thinning of subchondral bone at age 2 and 8 mo (open arrowheads), flattening of tibial plateau at age 8 mo (open arrow), and thickened sclerotic bone at age 18 mo (filled arrowhead). 3D renderings reveal erosion and pitting of HMWTg knees (arrows) while VectorTg knees maintain a smooth contour. C and D, 3D morphometric parameters revealed a decrease in BV/TV, Tb.N, and Tb.Th and an increase in Tb.Sp in 2- and 8-month-old HMWTg femoral and tibial epiphyses compared with VectorTg. BV/TV, trabecular bone volume; Tb.Sp, trabecular separation; Tb.N, trabecular number; Tb.Th; trabecular thickness. Values are the means ± SD. *, P < .05; **, P < .01; ***, P < .001, compared with VectorTg. N = 3/group.

μCT images of HMWTg knees showed thinning of the subchondral bone of the femur and tibia along with greater loss of trabeculae compared with VectorTg at 2 and 8 months of age (Figure 1B). By 18 months of age, evidence of severe OA was noted in HMWTg knees, which included sclerosis of the femur. Also, 3D surface renderings of HMWTg knees showed pitting, erosion, and deterioration, which increased in intensity with age, whereas the VectorTg subchondral plate retained smooth contour and joint integrity. 3D morphometric parameters calculated from μCT results of the epiphysis of HMWTg femurs and tibiae were significantly different from those of VectorTg mice at 2 and 8 months of age. Specifically, HMWTg epiphyses displayed modifications of the trabecular architecture that are indicative of early OA-like changes (Figure 1, C and D). At 2 months of age the BV/TV of the HMWTg epiphysis was decreased in both the femur and tibia by 35% (P < .01) and 11% (P < .01), respectively, compared with the VectorTg. The trabecular thickness (Tb.Th) of the HMWTg epiphysis was decreased in the femur and tibia at 2 months by 16% (P < .001) and 11% (P < .01), respectively, compared with the VectorTg. Also, at 2 months the femur of HMWTg mice had an increase in trabecular spacing (Tb.Sp) by 16% (P < .05) and a decrease in trabecular number (Tb.N) by 17% (P < .05) compared with VectorTg. Moreover, at 8 months of age, the epiphysis of the femur and tibia of HMWTg had a decrease in BV/TV by 42% (P < .05) and 27% (P < .05), respectively, compared with the VectorTg. The Tb.Sp was increased by 32% (P < .05) and the Tb.N was decreased by 25% (P < .05) in 8-month-old HMWTg femoral epiphyses, compared those of the VectorTg.

Histological changes in VectorTg and HMWTg cartilage

Articular cartilage integrity was determined histologically using Safranin-O staining to examine proteoglycan content in 2, 8, and 18-month-old mice. A dramatic decrease in Safranin-O staining (proteoglycan content) and articular cartilage thickness was found in HMWTg knees at all ages, particularly on the lateral side (Figure 2A, circled area). Osteophytes, formed through endochondral ossification of cartilaginous tissue, were also present in 18-month-old HMWTg knees (Figure 2B). The average tibial articular cartilage thickness was significantly decreased in 2-month-old HMWTg knees compared with VectorTg by 56%, in 8-month-old HMWTg compared with VectorTg by 65%, and in 18-month HMWTg compared with VectorTg by 62% (Figure 2C).

Figure 2.

Figure 2.

Open in a new tab

Proteoglycan content of VectorTg and HMWTg knees at age 2, 8, and 18 mo. A, Safranin-O stained representative images showed a decrease in cartilage thickness in all HMWTg samples. B, 18-month-old HMWTg knees showed severe cartilage loss and osteophyte development of mineralized cartilage (arrow) of femur. Bar = 200 μm. C, Mean lateral tibial articular cartilage thickness of VectorTg and HMWTg. Values are the means ± SD, n = 3–4/group.

We also examined GFP localization in the joints of 1-month-old HMWTg and VectorTg mice as a readout of transgene expression. GFP was expressed by osteocytes and osteoblasts in the bone areas of both strains, but was absent from the articular cartilage and tendons. Notably, GFP was markedly more intense in HMWTg tissues. These results confirmed bone-targeted overexpression of HMW FGF2 isoforms in the HMWTg mice (Figure 3).

Figure 3.

Figure 3.

Open in a new tab

GFP expression in the frontal section of knees of 1-month-old VectorTg and HMWTg mice. VectorTg mice created with the Col3.6-IRES/GFPsaph construct and HMWTg mice created with the Col3.6-HMW Fgf2 isoforms-IRES-GFP construct reveal expression of any GFP in the same location (area of bone), whereas GFP expression was overall increased in HMWTg. The outlined white area is the corresponding box below, which is at a higher magnification. This area of the lateral left side of the knee shows no GFP expression in articular cartilage (dashed arrow) or tendon and meniscus regions. Nuclear stain (blue): To-Pro-3.

Furthermore, to confirm that overexpression of the LMW FGF2 isoform does not lead to joint degeneration, we examined knees of 19-month-old VectorTg and LMWTg mice and found that there was no phenotypic difference in the subchondral bone by x-ray imaging and similar articular cartilage thickness between both groups (Figure 4).

Figure 4.

Figure 4.

Open in a new tab

VectorTg and LMWTg knees at 19 months of age are phenotypically similar. A, Representative digital x-ray images show a comparable subchondral bone appearance in VectorTg and LMWTg knees. N = 4/group. B, Safranin-O stained representative images show a similar cartilage phenotype and cartilage thickness in VectorTg and LMWTg samples. C, There was no difference between the mean lateral tibial articular cartilage thickness of VectorTg and LMWTg samples. Values are the means ± SD, n = 3/group.

Expression of OA markers in VectorTg and HMWTg knees

Total RNA was extracted from knee joints of 2-month-old VectorTg and HMWTg mice and used for measurement of mRNA of genes that are typically up-regulated in OA articular cartilage, including MMP-13 (a collagenase) and type X collagen (hypertrophic chondrocyte marker). Both MMP-13 and type X collagen mRNA expression was significantly increased in HMWTg joints (Figure 5A). In addition, immunohistochemistry was utilized to determine the localization and expression of MMP-13 and ADAMTS-5 (the major aggrecan degrading enzyme in murine cartilage). In 2-month-old HMWTg joints, expression of MMP-13 and ADAMTS-5 was increased in all joint tissues but particularly in the articular cartilage, compared with VectorTg (Figure 5, B and C). MMP-13 expression was also present in the enlarged tendon area of HMWTg knees, but was not present in VectorTg. MMP-13 expression in 8-month-old HMWTg joints was observed within the articular cartilage and in a developing osteophyte area (Figure 5B) and ADAMTS-5 expression in the articular cartilage was also increased compared the VectorTg (Figure 5C). In 18-month-old HMWTg knees, MMP-13 expression was increased overall compared with the VectorTg, most notably in the enlarged tendon areas near the femur (Figure 5B). Similarly, ADAMTS-5 expression was increased in these HMWTg knees compared with the VectorTg knees, particularly along the articular cartilage (Figure 5C). Furthermore, RNAscope was used to examine Col X expression, which showed an increase in labeled articular chondrocytes in HMWTg knees compared with the very few cells, which were labeled in the VectorTg at 2 months of age (Figure 5D).

Figure 5.

Figure 5.

Open in a new tab

Matrix metalloproteinase 13 (MMP-13), a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS-5), and type X collagen (Col X) expression in knees of VectorTg and HMWTg. A, Extraction of total RNA from knees of 2-month-old VectorTg and HMWTg and quantification of Mmp-13, Adamts5 and ColX mRNA by real-time reverse-transcription PCR (qPCR). Values are the means ± SD. *, P < .05; n = 7–8/group. B and C, Representative immunohistochemical staining of VectorTg and HMWTg knees at 2, 8, and 18 mo show increased protein expression of MMP-13 and ADAMTS-5 along articular cartilage (arrows). MMP-13 expression was increased in the enlarged tendon area of HMWTg joints at age 2 mo (filled arrowhead) and in the developing osteophyte region of HMWTg joints at age 8 mo (red arrow). Increased MMP-13 expression at 18 mo within area of tendon was present in HMWTg (filled arrowheads) and absent in VectorTg (open arrowhead). D, RNAscope shows increased Col X expression throughout the articular cartilage in HMWTg (arrows) compared with VectorTg mice. Magnification = 20×. For IHC/RNAscope n = 3–4/group.

Given that HMWTg mice developed an OA phenotype and given that multiple cytokines, growth factors, and oxidative stress and their downstream signaling pathways have been involved in the regulation of OA like changes in the subchondral bone as well as hypertrophic-like changes in osteoarthritic articular chondrocytes (16, 17), we examined gene expression in whole joints from 2-month-old Vector and HMWTg mice. We observed significant increases in the mRNA for bone morphogenetic protein 2 (Bmp2), bone morphogenetic protein 4 (Bmp4), insulin like growth factor 1 (Igf1), interleukin 1 beta (Il-1β), and hypoxia inducible factor 1(Hif1α) compared with the VectorTg (Table 3). There was no increase in TGF-β (Tgfβ1 or Tgfβ3) or TNF-α (Tnfα) or Indian hedgehog (Ihh), (data not shown). We also observed a significant increase in the transcription factors, Sex determining region Y box 9 (Sox9), and nuclear factor kappa B subunit 1 (Nf-κb1) (Figure 6A) in joints from HMWTg compared with Vector. Given that Sox9 and nuclear factor κB (NF-κB) signaling is important in the response to multiple inducers of OA, we also examined P65 subunit of NF-κB by immunohistochemistry and found there was expression throughout the articular cartilage of 2-month-old HMWTg mice, but no expression in the VectorTg (Figure 6B). Also, Sox9 expression was dramatically increased in the cartilage and menisci of HMWTg joints compared with VectorTg (Figure 6C).

Table 3.

Gene Expression in Knee Joints of VectorTg Versus HMWTg Mice

Gene VectorTg HMWTg P Value
Bmp-2 0.95 ± 0.28 1.56 ± 0.59 .003
Bmp-4 1.35 ± 0.50 2.68 ± 1.53 .008
Igf-1 1.52 ± 0.64 2.59 ± 1.62 .041
IL-1β 1.16 ± 0.41 2.08 ± 1.06 .005
Hif1α 1.29 ± 0.47 2.52 ± 1.54 .008
Open in a new tab

Figure 6.

Figure 6.

Open in a new tab

NF-κB and Sox9 mRNA and protein expression in knees of 2-month-old VectorTg and HMWTg. A, HMWTg knee joints have increased expression of NF-κB and Sox9 mRNAs. Values are the means ± SD. *, P < .05; ***, P < .001, n = 7–8/group. B, Representative immunohistochemical staining of HMWTg knees shows phospho-NF-κB p65 expression throughout the articular cartilage (arrows), whereas no expression is present in VectorTg samples. C, Representative images show increased expression of SOX9 along articular cartilage (arrows) and meniscus of HMWTg joints, whereas there is relatively low expression in VectorTg joints. Magnification = 10×; n = 3/group.

Evaluation of characteristics of osteoarthopathies in VectorTg and HMWTg joints

Although vascular endothelial growth factor (VEGF) (Vegf) mRNA levels were similar in both genotypes (data not shown), there was increased VEGF protein expression in 2-month-old HMWTg subchondral bone and articular cartilage (a typically avascular region) compared with VectorTg joints (Figure 7A). Alkaline phosphatase staining was used to visualize the expression and localization of hypertrophic chondrocytes, which showed an increase in HMWTg joints, encroaching upon a greater area of the articular surface compared with that of the VectorTg (Figure 7B). Mineral deposition was determined by von Kossa staining (Figure 7C). VectorTg mice showed mineralization, which corresponded to the alkaline phosphatase stained area of the articular cartilage, whereas, the HMWTg mice showed less mineral deposition in areas that did contain alkaline phosphatase-positive chondrocytes. Also, there was evidence of mineralization on the meniscus in HMWTg mice (Figure 7, B and C).

Figure 7.

Figure 7.

Open in a new tab

Comparison of VEGF, von Kossa, and alkaline phosphatase staining in 2-month-old VectorTg and HMWTg knees. A, HMWTg knee joints show VEGF expression (arrows) in the subchondral bone and at the interface of the articular cartilage and subchondral bone. No cells were labeled for VEGF in VectorTg knee sections. B, Representative alkaline phosphatase staining of knees from HMWTg shows more alkaline phosphatase-positive articular chondrocytes (arrows) compared with VectorTg. C, Corresponding von Kossa stained sections reveal a decrease in mineral deposition despite the presence of alkaline phosphatase-positive chondrocytes in HMWTg knees, relative to VectorTg knees. Magnification = 10×; n = 3/group.

Expression of FGF receptors and ligands in VectorTg and HMWTg knee joints

Given that HMWTg mice overexpress the HMW FGF2 isoforms, which results in increased FGF2 in bone (7) and also increased FGF23 expression in serum and bone (7), we examined total FGF2 and FGF23 expression within the joint. Fgf2 mRNA expression was significantly increased in 2-month-old HMWTg compared with the VectorTg (P < .01) and Fgf23 mRNA expression was increased but did not reach significance (Figure 8A). Immunohistochemistry confirmed that FGF2 protein expression was increased in the articular cartilage, meniscus, and tendons in HMWTg mice compared with VectorTg at 2 and 8 months of age (Figure 8B). Similarly, FGF23 protein expression was markedly increased within the articular cartilage and subchondral bone of HMWTg mice, notably in the developing osteophyte at 8 months (Figure 8C).

Figure 8.

Figure 8.

Open in a new tab

Fgf2 and Fgf23 mRNA and protein expression in knees of VectorTg and HMWTg mice. A, Fgf2 and FGF23 mRNA levels were increased in 2-month-old HMWTg knees compared with VectorTg. Values are the means ± SD. **, P < .01, n = 7–8/group. B, Representative immunohistochemical staining of FGF2 in 2- and 8-month-old HMWTg articular cartilage is increased (arrows). C, FGF23 protein expression is increased in 2-and 8-month-old HMWTg articular cartilage (arrows) and bone (arrowheads). For IHC n = 3–4/group; Bar = 50 μm.

FGF2 and FGF23 signal through the receptors FGFR1 and FGFR3, which have also been implicated in OA disease. FGFR1 and FGFR3 mRNA expression was significantly increased in whole joints of 2-month-old HMWTg compared with the VectorTg mice (Figure 9, A and B). Immunostaining revealed that FGFR1 protein was localized in the articular cartilage and meniscus of the joints, and seemed to be increased in HMWTg knees, compared with those of VectorTg at 2 months of age and even more strikingly at 8 months of age compared with the VectorTg control (Figure 9C). Immunohistochemistry also showed that HMWTg and VectorTg knees had similar expression of FGFR3 protein, particularly within the articular cartilage, which was decreased at 8 months of age compared with 2 months for both genotypes (Figure 9D).

Figure 9.

Figure 9.

Open in a new tab

FGFR1 and FGFR3 expression in knees of HMWTg and VectorTg mice. A and B, FGF2 and FGF23 mRNA level was significantly increased in HMWTg knees at age 2 mo. Values are the means ± SD. *, P ≤ .05; **, P < .01, n = 7–8/group. C, Representative immunohistochemical staining of FGFR1 was more robust in HMWTg cartilage at 2 and 8 mo (arrows). D, Immunohistochemical staining of FGFR3 reveals similar protein expression within articular cartilage of VectorTg and HMWTg at 2 and 8 mo (open arrows). For IHC n = 3–4/group; Bar = 50 μm.

Discussion

Although there have been numerous studies describing FGF2 and its involvement in OA, we provide evidence, for the first time, that the HMW isoforms of FGF2 may be a key component that contributes to joint degeneration in mice. In this study we demonstrated that mice overexpressing the nuclear HMW FGF2 isoforms in bone spontaneously exhibit signs of osteoarthropathy as early as 2 months of age and continue to develop a phenotype that mimic the progression of OA in human XLH subjects (2, 3). This phenotype included skeletal abnormalities of the knee, such as thinning of subchondral bone followed by sclerosis and osteophyte formation, modifications in trabecular architecture such as increased trabecular space and decreased trabecular number, thinning of the articular cartilage, up-regulated expression of cartilage degrading enzymes, and defective mineralization within the joint. Moreover, key FGF family members including FGF2, FGF23, and FGFR1 were overexpressed throughout the knee joint tissue of HMWTg mice.

Conflicting studies have been published about the role of FGF2 in joint homeostasis and degeneration (10–13). This study, in which we show that bone targeted overexpression of FGF2 HMW isoforms is accompanied by joint degeneration in transgenic mice, is consistent with catabolic potential of FGF2 in OA (10–13), which we suggest is probably due to the HMW FGF2 isoforms. HMWTg mice displayed an increase in total FGF2 in articular cartilage, yet the overexpression of the HMW FGF2 isoforms is in osteoblastic lineage cells. This increase in FGF2 within the cartilage may be driven by the overexpression of the HMW FGF2 isoform in the subchondral bone due to the crosstalk of the bone-cartilage unit. Previous reports suggest that products of OA subchondral bone can be secreted into the joint space and accessed by the articular cartilage via synovial fluid (17), and there is evidence to suggest that the HMW FGF2 isoform can be released from the cell (18). Also, given that the Hyp mouse has been found to develop degenerative osteoarthropathy (6) and we reported that these mice have an increase in the HMW FGF2 isoforms in osteocytes (7), as do our HMWTg mice, this suggests that these HMW FGF2 isoforms are contributing to the OA-like phenotype in the Hyp mouse. We posit that the changes within the Hyp mouse joint are due in part to the HMW FGF2 isoforms and not the LMW FGF2 isoforms, given that the knees from our LMWTg mice seemed to be nearly identical to those of the VectorTg control; furthermore, we reported that the LMWFGF2 isoform was not overexpressed in bones of the Hyp mouse (7).

HMWTg mice develop many similar characteristics of osteoarthropathy to those displayed by Hyp mice. Both HMWTg mice and Hyp mice (6) display thinning of articular cartilage without evidence of fraying, a hallmark of osteoarthropathy found in XLH subjects (2, 3). HMWTg and Hyp mice have overall hypomineralization (19, 20), which may be contributing to the observed thin articular cartilage, given that hypomineralization of the subchondral bone, is believed to contribute to the expansion of the tibial plateau in response to increased load in an OA knee, consequently causing the articular cartilage to stretch (21), resulting in decreased thickness. HMWTg joints had an increase in alkaline phosphatase and type X collagen staining, particularly in the area of the deep zone, as well as, throughout the cartilage, suggesting there could be an advancement of the tidemark and subchondral bone, a common OA-like characteristic (22, 23). Although there was increased alkaline phosphatase staining throughout the articular cartilage in HMWTg mice, the von Kossa staining did not correspond to the alkaline phosphatase-positive chondrocytes, which was similarly observed in Hyp mice (6), implying that defective mineralization is occurring. Vascular invasion of articular cartilage usually accompanies the advancement of the tidemark, which was evident in Hyp mice (6), as well as HMWTg mice, which displayed VEGF expression at the interface of the articular cartilage and subchondral bone. Angiogenesis was present in the articular cartilage, a typically avascular tissue, is another hallmark of joint degeneration and may contribute to endochondral osteophyte formation in HMWTg and Hyp joints. Generally, HMWTg and Hyp mice share many of the same phenotypic traits of osteoarthropathy that are common to humans with XLH.

Although there have been no studies investigating the subchondral bone trabecular architecture in the Hyp mouse, there is increasing evidence that articular cartilage and subchondral bone act together as a functional unit (17, 24). We reported modifications in subchondral bone of HMWTg mice that are consistent with many osteoarthritic models (24, 25), including thinning of bone in younger ages and sclerosis in older mice, which is prominent in XLH subjects (3). Also, the BV/TV and trabecular number and thickness were decreased, whereas the trabecular spacing was increased within the epiphysis of the femur and tibia. Other studies have shown that these alterations of the subchondral bone lead to an imbalance in stress distribution on the articular cartilage, ultimately leading to its degeneration (26). Thus, the decrease in articular cartilage thickness that was observed in HMWTg mice could be due to the modification of the underlying bone caused by the overexpression of the HMW FGF2 isoforms in that area.

A differing characteristic between HMWTg and Hyp knee joints is the localization and level of expression of articular cartilage matrix degradative enzymes, which are typically up-regulated in degenerated joints. However, MMP13 is reduced in the joints of Hyp mice even though they display signs of OA (6). In contrast, unlike Hyp mice, MMP13 expression was significantly increased in the articular cartilage of our HMWTg mice at all ages, suggesting these mice develop a more prototypical form of osteoarthropathy than Hyp mice (6). This elevation of MMP-13 may be attributed to the increase in FGF2 expression we observed in HMWTg mice, particularly in the articular cartilage, given that it has been shown that FGF2 stimulates MMP-13 activation (12). HMWTg mice displayed increased ADAMTS5 protein expression within the articular cartilage and areas of the joint compared with VectorTg at all ages. However, at 2 months of age Adamts5 mRNA expression was not significantly increased between HMWTg and VectorTg mice (data not shown). The lack of change in Adamts5 mRNA may be due to the age at which the RNA was isolated (2-month-old mice in our study). It should be noted that other studies have reported that Adamts5 mRNA expression is not changed in early-stage human OA cartilage but is significantly increased in late-stage degenerative joint disease (27).

The data from our study suggests that HMW FGF2 isoform overexpression in osteoblastic lineage cells is contributing to the up-regulation of cartilage degradative enzymes within the joint. We therefore propose that the overexpression of the HMW FGF2 isoforms may be contributing to osteoarthropathy in HMWTg mice by shifting catabolic over anabolic activity in the joint, causing an imbalance in cartilage homeostasis.

FGFR1 is known to have an important role in OA in humans and mice (28–30). We found that FGFR1 gene and protein expression were increased in HMWTg knees, which is similar to other studies in which FGFR1 expression is increased in human osteoarthritic chondrocytes (29). Our observation is also consistent with reported effects of inhibition of FGFR1 signaling in murine articular cartilage, which attenuates OA progression (30). Given that binding of FGF2 to FGFR1 leads to a cascade of events that are typically catabolic in articular chondrocytes, including activation of NF-κB, which we determined was increased in HMWTg joints, and MAPK pathways (31) it is possible that FGF2 overexpression in our model stimulates matrix degradation and blocks natural anabolic functions, causing the premature and severe osteoarthropathy we have observed. FGF2 also binds to FGFR3 with high affinity (29) and activation of FGFR3 has been shown to have anabolic effects in articular cartilage (32, 33). However, we found that similar patterns and levels of FGFR3 protein were present in the articular cartilage of HMWTg and VectorTg mice. Similarly, other studies have shown that FGFR3 expression is unchanged in mice that underwent surgical-induction of OA (34). However, FGFR3 is decreased in human osteoarthritic chondrocytes in vitro (29). These studies emphasize the inconsistent and potentially distinct roles of FGFR1 and FGFR3 in joint homeostasis and degeneration. Given that FGF2 binds both FGFR1 and FGFR3, which have contrasting roles in cartilage homeostasis, the FGFR1/FGFR3 ratio is likely essential in the biological outcome of FGF2. Although we found an increase in FGFR3 gene expression in HMWTg joints, the ratio of FGFR1: FGFR3 gene expression was significantly greater than that found in VectorTg (data not shown). This was also seen in human OA articular chondrocytes compared with healthy chondrocytes (29). However, the increased FGFR3 gene expression observed in this study could also reflect the inclusion of extracts used for this analysis. Indeed, it has been found by microarray analysis that FGFR3 expression is significantly increased in the subchondral bone of rats with degenerative joint disease (34). In addition, FGFR3 gene expression in our study may be increased due to an early repair response in HMWTg mice, given that the qPCR data are only for 2-month-old mice, and FGFR3 has been reported to promote anabolic activities in articular cartilage (32). Future studies examining FGFR3 gene expression in older mice and in specific tissue types are needed to fully resolve the relationships between FGFR1, FGFR3, and FGF2 in the joint.

HMWTg mice phenocopy most traits in XLH subjects, which includes increased FGF23 levels in serum and bone (7) and now similar osteoarthropathies (2, 3). Our laboratory has previously shown that the HMW isoforms of FGF2 modulate and colocalize with FGF23 in osteoblast and osteocytes in HMWTg mice (7). Thus, HMW FGF2 isoforms may be contributing to OA through FGF23 signaling, given that FGF23 expression is up-regulated in human osteoarthritic chondrocytes (35), which has been determined to drive MMP-13 expression (36), and we also demonstrated that FGF23 protein and MMP-13 mRNA and protein expression was increased within the joint, including the cartilage of HMWTg mice. Previous studies have shown that the development of OA could be due to the activation of hypertrophic differentiation of articular chondrocytes (37) and multiple signaling pathways are responsible including growth factors, such as BMPS and IGF-I, cytokines such as IL-1β and transcription factors Hif, Nf-κB, and Sox9 (16). Furthermore, although Sox9 is a marker for chondroprogenitor cells, studies have shown there is an increase in Sox9 expression in human chondrocytes of early-stage OA (38), as well as, murine osteoarthritic cartilage, particularly in areas of developing osteophytes (39). Given that we found a significant increase in the expression of genes of these pathways in HMWTg joints, the OA-like phenotype may be caused by changes in the subchondral bone as well as hypertrophic differentiation of chondrocytes due to modulation of these factors and signaling molecules.

In conclusion, transgenic mice with bone-targeted overexpression of HMW FGF2 (HMWTg mice) developed OA-like characteristics that were not simply restricted to a single component of the joint. Although changes in articular cartilage occurred, modifications within the subchondral bone were concurrently happening, likely due to the targeted overexpression of the HMW FGF2 isoforms in bone in this model. Additional studies will be necessary to decipher the exact mechanism that causes this phenotype. Overall, HMWTg mice will serve as a novel model for joint degeneration, particularly in the context of osteoarthropathy observed with XLH, which could be used in future studies aimed at better understanding the pathophysiology of osteoarthropathy and OA, and may offer new insight into how to treat this disabling disorder.

Acknowledgments

This work was supported by National Institutes of Health Grant RO1 DK098566.

Disclosure Summary: The authors have nothing to disclose.

Footnotes

Abbreviations:
3D
three-dimensional
μCT
microcomputed tomography
BV/TV
bone volume/tissue volume
FGF2
fibroblast growth factor 2
FGF23
fibroblast growth factor 23
HMW
high molecular weight
LMW
low molecular weight
NF-κB
nuclear factor κB
OA
osteoarthritis
Tb.N
trabecular number
Tb.Sp
trabecular spacing
Tb.Th
trabecular thickness
VEGF
vascular endothelial growth factor
XLH
X-linked hypophosphatemia.

References

  • 1. Econs MJ, Francis F. Positional cloning of the PEX gene: New insights into the pathophysiology of X-linked hypophosphatemic rickets. Am J Physiol Renal Physiol. 1997;273:F489–F498. [DOI] [PubMed] [Google Scholar]
  • 2. Hardy DC, Murphy WA, Siegel BA, Reid IR, Whyte MP. X-linked hypophosphatemia in adults: Prevalence of skeletal radiographic and scintigraphic features. Radiology. 1989;171(2):403–414. [DOI] [PubMed] [Google Scholar]
  • 3. Reíd IR, Hardy DC, Murphy WA, Teitelbaum SL, Bergfeld MA, Whyte MP. X-linked hypophosphatemia: A clinical, biochemical, and histopathologic assessment of morbidity in adults. Medicine. 1989;68(6):336–352. [PubMed] [Google Scholar]
  • 4. Liu S, Guo R, Simpson LG, Xiao ZS, Burnham CE, Quarles LD. Regulation of fibroblastic growth factor 23 expression but not degradation by PHEX. J Biol Chem. 2003;278(39):37419–37426. [DOI] [PubMed] [Google Scholar]
  • 5. Liang G, Katz LD, Insogna KL, Carpenter TO, Macica CM. Survey of the enthesopathy of X-linked hypophosphatemia and its characterization in Hyp mice. Calcif Tissue Int. 2009;85(3):235–246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Liang G, Vanhouten J, Macica CM. An atypical degenerative osteoarthropathy in Hyp mice is characterized by a loss in the mineralized zone of articular cartilage. Calcif Tissue Int. 2011;89(2):151–162. [DOI] [PubMed] [Google Scholar]
  • 7. Xiao L, Naganawa T, Lorenzo J, Carpenter TO, Coffin JD, Hurley MM. Nuclear isoforms of fibroblast growth factor 2 are novel inducers of hypophosphatemia via modulation of FGF23 and KLOTHO. J Biol Chem. 2010;285(4):2834–2846. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Delrieu I. The high molecular weight isoforms of basic fibroblast growth factor (FGF-2): An insight into an intracrine mechanism. FEBS Lett. 2000;468(1):6–10. [DOI] [PubMed] [Google Scholar]
  • 9. Touriol C, Bornes S, Bonnal S, et al. Generation of protein isoform diversity by alternative initiation of translation at non-AUG codons. Biol Cell. 2003;95(3–4):169–178. [DOI] [PubMed] [Google Scholar]
  • 10. Orito K, Koshino T, Saito T. Fibroblast growth factor 2 in synovial fluid from an osteoarthritic knee with cartilage regeneration. J Orthop Sci. 2003;8(3):294–300. [DOI] [PubMed] [Google Scholar]
  • 11. Chia SL, Sawaji Y, Burleigh A, et al. Fibroblast growth factor 2 is an intrinsic chondroprotective agent that suppresses ADAMTS-5 and delays cartilage degradation in murine osteoarthritis. Arthritis Rheumatol. 2009;60:2019–2027. [DOI] [PubMed] [Google Scholar]
  • 12. Wang X, Manner PA, Horner A, Shum L, Tuan RS, Nuckolls GH. Regulation of MMP-13 expression by RUNX2 and FGF2 in osteoarthritic cartilage. Osteoarthritis Cartilage. 2004;12(12):963–973. [DOI] [PubMed] [Google Scholar]
  • 13. Im HJ, Li X, Muddasani P, et al. Basic fibroblast growth factor accelerates matrix degradation via a neuro-endocrine pathway in human adult articular chondrocytes. J Cell Physiol. 2008;215(2):452–463. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
  • 14. Xiao L, Liu P, Li X, et al. Exported 18-kDa isoform of fibroblast growth factor-2 is a critical determinant of bone mass in mice. J Biol Chem. 2009;284(5):3170–3182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Shepard JB, Jeong JW, Maihle NJ, O'Brien S, Dealy CN. Transient anabolic effects accompany epidermal growth factor receptor signal activation in articular cartilage in vivo. Arthritis Res Ther. 2013;15:1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Zhong L, Huang X, Karperien M, Post JN. The regulatory role of signaling crosstalk in hypertrophy of MSCs and human articular chondrocytes. Int J Mol Sci. 2015;16(8):19225–19247. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Westacott CI, Webb GR, Warnock MG, Sims JV, Elson CJ. Alteration of cartilage metabolism by cells from osteoarthritic bone. Arthritis Rheumatol. 1997;40:1282–1291. [DOI] [PubMed] [Google Scholar]
  • 18. Taverna S, Ghersi G, Ginestra A, et al. Shedding of membrane vesicles mediates fibroblast growth factor-2 release from cells. J Biol Chem. 2003;278(51):51911–51919. [DOI] [PubMed] [Google Scholar]
  • 19. Xiao L, Esliger A, Hurley MM. Nuclear fibroblast growth factor 2 (FGF2) isoforms inhibit bone marrow stromal cell mineralization through FGF23/FGFR/MAPK in vitro. J Bone Miner Res. 2013;28(1):35–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Miao D, Bai X, Panda D, McKee M, Karaplis A, Goltzman D. Osteomalacia in hyp mice is associated with abnormal phex expression and with altered bone matrix protein expression and deposition. Endocrinology. 2001;142(2):926–939. [DOI] [PubMed] [Google Scholar]
  • 21. Kurosawa H, Fukubayashi T, Nakajima H. Load-bearing mode of the knee joint: Physical behavior of the knee joint with or without menisci. Clin Orthop Relat Res. 1980;149(149):283–290. [PubMed] [Google Scholar]
  • 22. Oegema TR, Jr, Carpenter RJ, Hofmeister F, Thompson RC., Jr The interaction of the zone of calcified cartilage and subchondral bone in osteoarthritis. Microsc Res Tech. 1997;37(4):324–332. [DOI] [PubMed] [Google Scholar]
  • 23. Rees JA, Ali SY. Ultrastructural localisation of alkaline phosphatase activity in osteoarthritic human articular cartilage. Ann Rheum Dis. 1988;47(9):747–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Hayami T, Pickarski M, Zhuo Y, Wesolowski GA, Rodan GA, Duong LT. Characterization of articular cartilage and subchondral bone changes in the rat anterior cruciate ligament transection and meniscectomized models of osteoarthritis. Bone. 2006;38(2):234–43. [DOI] [PubMed] [Google Scholar]
  • 25. Altman R, Asch E, Bloch D, et al. Development of criteria for the classification and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Diagnostic and Therapeutic Criteria Committee of the American Rheumatism Association. Arthritis Rheum. 1986;29(8):1039–49. [DOI] [PubMed] [Google Scholar]
  • 26. Burr DB. The importance of subchondral bone in osteoarthrosis. Curr Opin Rheumatol. 1998;10(3):256–262. [DOI] [PubMed] [Google Scholar]
  • 27. Bau B, Gebhard PM, Haag J, Knorr T, Bartnik E, Aigner T. Relative messenger RNA expression profiling of collagenases and aggrecanases in human articular chondrocytes in vivo and in vitro. Arthritis Rheum. 2002;46(10):2648–2657. [DOI] [PubMed] [Google Scholar]
  • 28. Ellman MB, Yan D, Ahmadinia K, Chen D, An HS, Im HJ. Fibroblast growth factor control of cartilage homeostasis. J Cell Biochem. 2013;114(4):735–742. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Yan D, Chen D, Cool SM, et al. Fibroblast growth factor receptor 1 is principally responsible for fibroblast growth factor 2-induced catabolic activities in human articular chondrocytes. Arthritis Res Ther. 2011;13(4):R130. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
  • 30. Weng T, Yi L, Huang J, et al. Genetic inhibition of fibroblast growth factor receptor 1 in knee cartilage attenuates the degeneration of articular cartilage in adult mice. Arthritis Rheum. 2012;64(12):3982–3992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Muddasani P, Norman JC, Ellman M, van Wijnen AJ, Im HJ. Basic fibroblast growth factor activates the MAPK and NFkappaB pathways that converge on Elk-1 to control production of matrix metalloproteinase-13 by human adult articular chondrocytes. J Biol Chem. 2007;282(43):31409–31421. [DOI] [PubMed] [Google Scholar]
  • 32. Davidson D, Blanc A, Filion D, et al. Fibroblast growth factor (FGF) 18 signals through FGF receptor 3 to promote chondrogenesis. J Biol Chem. 2005;280(21):20509–20515. [DOI] [PubMed] [Google Scholar]
  • 33. Li X, Ellman MB, Kroin JS, et al. Species-specific biological effects of FGF-2 in articular cartilage: Implication for distinct roles within the FGF receptor family. J Cell Biochem. 2012;113(7):2532–2542. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Zhang R, Fang H, Chen Y, et al. Gene expression analyses of subchondral bone in early experimental osteoarthritis by microarray. PLoS One. 2012;7(2):e32356. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Orfanidou T, Iliopoulos D, Malizos KN, Tsezou A. Involvement of SOX-9 and FGF-23 in RUNX-2 regulation in osteoarthritic chondrocytes. J Cell Mol Med. 2009;13(9B):3186–3194. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Bianchi A, Guibert M, Cailotto F, et al. Fibroblast growth factor 23 drives MMP13 expression in human osteoarthritic chondrocytes in a klotho-independent manner. Osteoarthritis Cartilage. 2016;24(11):1961–1969. [DOI] [PubMed] [Google Scholar]
  • 37. Dreier R. Hypertrophic differentiation of chondrocytes in osteoarthritis: The developmental aspect of degenerative joint disorders. Arthritis Res Ther. 2010;12:1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Zhang Q, Ji Q, Wang X, et al. SOX9 is a regulator of ADAMTSs-induced cartilage degeneration at the early stage of human osteoarthritis. Osteoarthritis Cartilage. 2015;23(12):2259–2268. [DOI] [PubMed] [Google Scholar]
  • 39. Zhou S, Wang Z, Tang J, et al. Exogenous fibroblast growth factor 9 attenuates cartilage degradation and aggravates osteophyte formation in post-traumatic osteoarthritis. Osteoarthritis Cartilage. 2016. [Epub ahead of print]. [DOI] [PubMed] [Google Scholar]

Articles from Endocrinology are provided here courtesy of The Endocrine Society

RESOURCES