Tibial plateaus were dissected away from the remaining cartilage and trabecular bone under sterile conditions from OA patients who had undergone total knee replacement surgery as previously described 111240. A total of 64 patients (aged 71.5 ± 9.9 years) classified as having OA according to the recognized clinical criteria of the American College of Rheumatology were included in this study 41. OA grade ranged from moderate to severe in these patients. None of the patients had received medication that would interfere with bone metabolism, including corticosteroids, for six months before surgery. A total of 16 subchondral bone specimens of tibial plateaus from normal individuals (aged 62.2 ± 18.9 years) were collected at autopsy within 12 h of death. These were used following the establishment that they had not been on any medication that could interfere with bone metabolism or had any bone metabolic disease. Individuals showing abnormal cartilage macroscopic changes and/or subchondral bone plate sclerosis were not included in the normal group. All human materials were acquired following a signed agreement by patients undergoing knee surgery or their relatives for the specimens collected at autopsy following the Centre Hospitalier de l'Université de Montréal (CHUM) ethical committee guidelines.

Isolation of subchondral bone plate and the cell cultures were prepared as we recently described 42. At confluence, cells were passaged once at 25,000 cells/cm² and grown for five days in HamF12/DMEM media (Sigma-Aldrich, Oakville, Ontario, Canada) containing 10% FBS before specific assays. These cells were incubated with the same media containing 0.5% FBS. After 24 hours of preconditioning, cells were incubated for either an additional 48 hours in HamF12/DMEM media containing 0.5% FBS and the indicated treatments for the determination of phenotypic markers, or they were incubated for an additional 24 hours in the same media in presence or absence of increasing doses of leptin and the indicated treatments for the determination of the expression of leption or OB-Rb, or they were incubated for 15 minutes with increasing doses of leptin in preparation for Western blot analysis of p42/44 and p38. For the determiniation of phenotypic markers, cells were either treated with 1 μg/ml recombinant human leptin (rhleptin, Calbiochem, San Diego, California, USA), 10 μg/ml recombinant human leptin R/Fc chimera (R&D Systems, Minneapolis, MN, USA) that neutralizes the activity of rhleptin, 100 μM Tyrphostin (AG490, Sigma-Aldrich), 75 μM piceatannol (Pce, Sigma-Aldrich), or the vehicle. Supernatants were collected at the end of the incubation and kept at -80°C prior to assays. Cells were either prepared for SDS-PAGE separation or RT-PCR experiments. Cells prepared for SDS-PAGE separation were lysed with RIPA buffer (50 mM Tris HCl pH 7.4, 1% NP-40, 0.5% Na-deoxicholate, 0.1% SDS, 150 mM NaCl with the following inhibitors: 10 μg/ml aprotinin, 10 μg/ml leupeptin, 10 μg/ml pepstatin, 10 μg/ml O-phenatroline, 1 mM Na-orthovanadate, 1 mM DTT), and kept at -80°C prior to assays. Protein determination was performed by the bicinchoninic acid method 43.

Phenotypic features of Ob were determined by evaluating 1,25(OH)₂D₃-dependent (50 nM) alkaline phosphatase activity and osteocalcin release, and by measuring the release of the carboxy-terminal propeptide of collagen type 1 (CICP) in cells treated or not for their last 48 hours of culture with recombinant human leptin R/Fc chimera to neutralize the activity of leptin, 100 μM tyrphostin (AG490) or 75 μM piceatannol (Pce), inhibitors of leptin signaling, or with siRNA directed against leptin or OB-Rb (see below). Alkaline phosphatase activity was determined on cell aliquots by substrate hydrolysis using p-nitrophenylphosphate (PNPP), and osteocalcin release was determined in cell supernatants using an EIA as previously described 1112. CICP was determined using a selective ELISA (Quidel Corporation, Cedarlane, Hornby, Ontario, Canada) in conditioned media from confluent OA Ob incubated in HAMF12/DMEM media containing 0.5% bovine serum albumine (BSA). CICP release was then reported as ng per cellular proteins. Transforming growth factor-β1 (TGF-β1) was measured in supernatants using a highly specific Quantikine ELISA assay from R&D Systems (Minneapolis, MN, USA). The sensitivity of the assay is 7 pg/ml and is a very specific assay that does not cross react with related cytokines/growth factors when tested at saturating concentrations. Cellular proliferation was assessed using two complementary approaches: the BrdU cell proliferation assay as described in the system's manual from Calbiochem (San Diego, California, USA) and MTT assay as described by Zhao et al 44. Cells were plated at 10,000 cells/cm² in 96-well plates in Ham F12/DMEM media containing 10% FBS. After overnight attachment, cells were fed Ham F12/DMEM media containing 0.5% FBS for 24 hours prior to stimulation with or without increasing doses of recombinant human leptin as indicated for another 24 hours of incubation.

For RT-PCR assays, total cellular RNA from normal and OA Ob was extracted with the TRIzol™ reagent (Invitrogen, Burlington, Ontario, Canada) according to the manufacturer's specifications and treated with the RNA-free™ Dnase Treatment and Removal kit (Ambion, Austin, TX, USA) to ensure complete removal of chromosomal DNA. The RNA was quantitated using the RiboGreen RNA quantification kit (Molecular Probes, Eugene, OR, USA). The RT reactions were primed with random hexamers with 1 μg of total RNA in a 100 μl final reaction volume followed by PCR amplification as previously described 40 using 20 pmol of each specific PCR primers (see below). The amplification of all mRNA species was performed separately from GAPDH mRNA amplification to avoid substrate depletion. After amplification, DNA was analyzed on an agarose gel and visualized by ultraviolet detection.

Real-time quantification of leptin and GAPDH mRNA was performed in the GeneAmp 5700 Sequence Detection System (Applied Biosystems, Foster City, CA, USA) with the 2× Quantitect SYBR Green PCR Master Mix (Qiagen, Missisauga, Ontario, Canada) used according to the manufacturer's specifications. Primers used were: 5'-GGCTTTGGCCCTATCTTTTC-3' (sense) and 5'-GGATAAGGTCAGGATGGGGT-3' (antisense) for Lep1; 5'-CCTCATCAAGACAATTGTCACC-3' (sense) and 5'-CAGCATGTCCTGCAGAGACC-3' (antisense) for Lep2; 5'-GCCAGAGACAACCCTTTGTTAAA-3' (sense) and 5'-TGGAGAACTCTGATGTCCGTGAA-3' (antisense) for OB-Rb; 5'-CAGAACATCATCCCTGCCTCT-3' (sense) and 5'-GCTTGACAAAGTGGTCGTTGAG-3' (antisense) for GAPDH. Amplicons were 197, 376, 417 and 319 bp, respectively. In brief, 100 ng of the cDNA obtained from the RT reactions were amplified in a total volume of 50 μl consisting of 1× Master mix, uracil-N-glycosylase (UNG, 0.5 Unit, Epicentre Technologies, Madison, WI, USA) and the gene-specific primers which were added at a final concentration of 200 nM. The tubes were first incubated for two minutes at 50°C (UNG reaction), then at 95°C for 15 minutes (UNG inactivation and polymerase activation) followed by 40 cycles consisting each of denaturation (94°C for 15 seconds), annealing (60°C for 30 seconds), extension (72°C for 30 seconds) and data acquisition (77°C for 15 seconds) steps. The data were collected and processed with the GeneAmp 5700 SDS software and given as threshold cycles (Ct), corresponding to the PCR cycle at which an increase in reporter fluorescence above baseline signal can first be detected. When comparing normal and OA basal expression levels, the Ct were converted to the number of molecules and the values for each sample calculated as the ratio of the number of molecules of the target gene/number of molecules of GAPDH.

We used a siRNA technique to transiently inhibit leptin or OB-Rb expression in OA Ob. SiRNA were obtained from Dharmacon (Lafayette, CO, USA) and we followed the manufacturer's directions for their preparation. Briefly, OA Ob were split at 100,000 cells/ml. Leptin or OB-Rb siRNA (a set of four different siRNA per gene) or scramble RNA (basal condition) was added to OA Ob at a final concentration of 100 ng/ml with 6 μl Hi-perfect (Quiagen, Missisauga, ON, Canada) per 100 μl total volume in BGJb media without serum for one hour on Day 0 and Day 3. Cells were then fed BGJb media with 10% FBS containing 50 nM 1,25(OH)₂D₃ until Day 7, with media changes every two days. Cells were harvested in either ALPase buffer to perform ALP and protein determination or in TRIzol to prepare for RT-PCR to detect changes in leptin and OB-Rb levels. Supernatants were kept for the determination of osteocalcin.

The cell extracts were loaded on polyacrylamide gels and separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under reducing condition 45. Loading of the protein was adjusted according to the cellular protein concentration of each specimen. The proteins were then electrophoretically transferred onto Polyvinylidene Fluoride (PVDF) membranes (Boehringer Mannheim, Penzberg, Germany), and immunoblotting was performed as described in the ECL Plus Western blotting detection system's manual (Amersham Pharmacia Biotech, Piscataway, NJ, USA). Rabbit anti-leptin receptor at a dilution of 1:1,000 (Cedarlane, Hornby, Ontario, Canada), rabbit anti-human actin at a dilution of 1:10,000 (Sigma-Aldrich), rabbit anti p42/44 at a dilution of 1:5,000 (Cell Signaling Technology, Beverly, MA, USA), rabbit anti-phosphorylated p42/44 (Thr202/Tyr204) at a dilution of 1:5,000 (Cell Signaling Technology), rabbit anti p38 at a dilution of 1:2,000 (Cell Signaling Technology), and anti-phosphorylated p38 at a dilution of 1:1,000 (Cell Signaling Technology) as primary antibodies, and goat anti-rabbit IgG as secondary antibodies at a dilution of 1:20,000 (Upstate Biotechnology, Lake Placid, NY, USA) were used for the assays.

Densitometry analysis of western blot films was performed on a Macintosh Mac OS 9.1 computer using the public domain NIH Image program developed at the U.S. National Institutes of Health with the Scion Image 1.63 program 46.

Leptin was evaluated in Ob-conditioned media. Confluent Ob were cultured for 48 h in HAMF12/DMEM media containing 0.5% FBS. At the end of the incubation, their conditioned-media were concentrated five-fold using Amicon Ultra-4 filters (Ultracil-10 k, Millipore Corporation, Bedford, MA, USA) with a cutoff of 10 kDa. Samples were centrifuged at 1,000 g for 15 minutes at 4°C. The concentrated conditioned media were then tested for leptin using a selective high sensitivity ELISA (R&D Systems). The sensitivity of the assay was 7.8 pg/ml and the intra-assay precision is 3.2 ± 0.2%.

All quantitative data are expressed as mean ± SEM. Statistical analysis was performed by an ANOVA analysis of variance for dose-response experiments, followed by adequate subtests when statistical significance was reached. A non parametric Mann-Whitney U statistical test was performed for all other experiments and P values < 0.05 were considered statistically significant.

We first questioned if human OA osteoblasts (Ob) expressed leptin compared to normal Ob using real-time RT-PCR with two different set of primers, one described by Dumond et al 37 for rat samples and adapted to the human sequence, and the other by Gordeladze et al 16 for primary human osteoblasts. Using both sets of primers we detected leptin expression in OA Ob (Figure 1A). We next evaluated if OA Ob produced variable levels compared to normal Ob. Using real-time RT-PCR we observed that OA Ob produced about approximately five-fold more leptin mRNA than normal Ob using one set of primers (Figure 1B). Since leptin has been shown to promote its own expression 47, we next determined if this could be the case in OA Ob. Indeed, leptin dose-dependently stimulated its own expression (Figure 1C), yet this was also the case for OB-Rb expression (Figure 2B). As Ob expressed leptin, we next evaluated the capacity of Ob to synthesize leptin. As shown in Figure 1D, OA Ob released about approximately two-fold more leptin than normal Ob under basal condition when measured using a very selective ELISA.

In order to determine if OA Ob could respond to leptin, we next evaluated the presence of the long, signaling competent, form of the leptin receptor (OB-Rb). As shown in Figure 2A using real-time RT-PCR, OA Ob expressed slightly less OB-Rb than normal Ob although this did not reach significance. Exogenous leptin at high concentrations significantly stimulated OB-Rb expression in OA Ob (Figure 2B). In addition, OB-Rb mRNA levels were increased by both TGF-β1 and HGF in OA Ob (not illustrated), and this increased expression was reflected at the protein level by Western blot analysis (Figure 2C). Similar Western blot results were obtained with OA chondrocytes (not illustrated).

Since OA Ob expressed both leptin and leptin receptors, we tested if these cells could respond to exogenous leptin and we first determined the effect of leptin on cell proliferation. Figure 3A and 3B show that leptin dose-dependently (1 ng/ml to 10 μg/ml) stimulated cell proliferation and this effect plateaued at 100 ng/ml leptin when assessing proliferation using BrdU incorporation or MTT assay respectively. We next evaluated if the effect of leptin on cell proliferation was via the Erk 1/2 MAPK pathway as we previously showed with insulin-like growth factor 1 45. Indeed, in response to exogenous leptin, phospho p42/44 MAPK levels rose (Figure 3C). This effect was again dose-dependent and also plateaued around 100 ng/ml (Figure 3D). In addition, we evaluated the role of leptin on the p38 pathway. Again, leptin dose-dependently stimulated phospho p38 levels (Figure 3E) and this effect was significant at doses as low as 1 μg/ml (Figure 3F).

Leptin influences the synthesis of phenotypic markers and inflammatory mediators in a number of cells and in particular can increase phenotypic markers in primary human Ob 36. Because OA Ob responded to exogenous leptin, we then questioned if the endogenous elevated leptin production observed in OA Ob could be responsible for the abnormal phenotypic markers of these cells. Hence, we measured alkaline phosphatase activity, osteocalcin release and the production of CICP under basal condition and in the presence leptin or of a recombinant human leptin R/Fc chimera (anti-Rb) that neutralizes the activity of leptin. First, in preliminary assays we tested if exogenous leptin, the recombinant leptin R/Fc chimera or the antagonist of leptin signaling AG490 would alter alkaline phosphatase activity in normal Ob. Indeed, leptin addition to normal Ob stimulated vitamin D₃-dependent alkaline phosphatase activity, however, neither anti-Rb nor AG490 had any effect on this activity (Figure 4A), indicating no cytotoxic effects of these treatments on normal Ob. Hence, we next tested their effect on OA Ob and compared it to basal levels of these phenotypic markers in normal Ob run in parallel. Here, the inhibition of leptin signaling in OA Ob in response to 100 μM AG490 or 75 μM Pce, selective inhibitors of leptin intracellular signaling, reduced ALPase and CICP to values similar to normal Ob (Figure 4B and 4D), whereas the effect of these inhibitors on osteocalcin secretion could not be tested since they interfered with the EIA method. These inhibitors did not promote any significant cell death as assessed by total protein content and cell count by trypan blue exclusion (not illustrated). In addition, anti-Rb inhibited all these activities in OA Ob (Figure 4B to 4D). We then questioned if exogenous leptin could promote these activities. The addition of exogenous leptin to OA Ob enhanced vitamin D₃-dependent alkaline phosphatase activity (Figure 4B) as in normal Ob (Figure 4A), but it failed to further stimulate osteocalcin release (Figure 4B) or collagen type 1 production (Figure 4C) above their already elevated values in OA Ob.

As another key feature of OA Ob that distinguishes them from normal Ob is their enhanced production of TGF-β1 12, and because leptin has been shown to stimulate TGF-β1 synthesis in other cells, we evaluated if high levels of TGF-β1 in OA Ob could be due to a response to endogenous leptin via a paracrine/autocrine stimulation. As shown in Figure 4E, TGF-β1 levels in OA Ob were elevated compared to normal Ob and the presence of AG490 or Pce reduced by approximately 50% and approximately 60% the endogenous levels of TGF-β1 in OA Ob, reducing them to near normal values.

Last, using siRNA techniques, we next evaluated if inhibiting leptin or OB-Rb would abrogate the response of OA Ob to endogenous leptin production. Indeed, as shown in Figure 5A, siRNA against leptin reduced alkaline phosphatase activity about 60% compared to a scrambled RNA. A similar observation could be made for osteocalcin (Figure 5B). Likewise, inhibiting OB-Rb expression using siRNA techniques also reduced ALP and OC about 60% in OA Ob (Figure 5A and 5B). Figures 5C and 5D show that specific siRNA inhibition reduced leptin and OB-Rb expression 60 and 55% respectively in these cells compared to a scrambled RNA.