Collagenase-P was purchased from Roche Applied Science (Indianapolis, IN, USA). Pronase was from Calbiochem (Gibbstown, NJ, USA). DMEM/Ham's F-12 (1:1), antibiotics, fetal bovine serum, and PicoGreen double-stranded DNA assay reagent were from Invitrogen (Carlsbad, CA, USA). IGF-I was from Austral Biologicals (San Ramon, CA, USA). [³⁵S]sulfate was from GE Healthcare (Piscataway, NJ, USA). Antibodies and their sources were as follows: insulin receptor substrate (IRS-1) (Ser(P)-312 and total) was from Upstate Biotechnology, Inc. (Lake Placid, NY, USA); Akt (Ser(P)-473 and total), ERK1/2 (Thr(P)-202/Tyr(P)-204 and total) and mitogen-activated protein kinase kinase (MEK) inhibitor (U0126) were from Cell Signaling Technology (Danvers, MA, USA). Recombinant eNAMPT/visfatin/PBEF (endotoxin levels less then < 0.1 EU/μg protein) was from BioVision (Mountain View, CA, USA).

Human ankle cartilage was obtained from tissue donors within 48 hours of death through the National Disease Research Interchange (Philadelphia, PA, USA) and the Gift of Hope Organ and Tissue Donor Network (Elmhurst, IL, USA) in accordance with institutional protocols. Only tissue from donors without a history of known arthritis was used. The tissue was graded on a scale of 0 to 4 for evidence of morphological changes, as previously described 21 . All tissue for this study was either grade 0 or 1. Tissues from a total of 40 donors ranging from 40 to 90 years old were used in the experiments. Cells from at least three independent donors were used in each experiment.

Chondrocytes were isolated under aseptic conditions by sequential enzymatic digestion at 37°C using pronase 2 mg/ml in serum-free DMEM/F-12/antibiotics for 1 hour followed by overnight digestion with collagenase-P at 0.25 mg/ml in DMEM/F-12 (5% fetal bovine serum). Viability of isolated cells was determined using trypan blue and cells were counted using a hemocytometer. Monolayer cultures were established by plating cells in six-well plates at 2 × 10⁶ cells/ml in DMEM/F-12 medium supplemented with 10% fetal bovine serum. Cells were maintained for approximately 3 to 5 days with feedings every 2 days until they reached 100% confluency prior to experimental use.

The [³⁵S]sulfate incorporation assay was performed to measure PG synthesis. Chondrocytes in culture were made serum-free and pretreated with eNAMPT (0 to 5 μg/ml) overnight followed by 24-hour stimulation with IGF-I (50 ng/ml). The medium was then replaced with fresh serum-free medium 1 hour prior to incubation with [³⁵S]sulfate for an additional 4 hours. The [³⁵S]sulfate incorporation was measured using the Alcian blue precipitation method 15 and normalized to DNA content. DNA was measured using the PicoGreen double-stranded DNA assay according to the manufacturer's protocol.

Normal human chondrocytes cultured in serum-free DMEM/Ham's F-12 supplemented with 1% mini ITS plus ascorbate were treated with or without eNAMPT (5 μg/ml) overnight followed by IGF-1 (50 μg/ml) for an additional 24 hours. After incubation, media were removed and cell layers were extracted according to the manufacturer's protocol and analyzed for collagen II levels using an ELISA kit (MD Biosciences Inc., St Paul, MN, USA).

Total RNA was extracted using TRIzol (Invitrogen) according to the manufacturer's protocol. Total RNA (2 μg) was used to synthesize cDNA using oligo(dT)15 as the reverse primer. Equivalent amounts of cDNA were used for real-time PCR in a 25 μl reaction mixture with 12.5 μl of 2× SYBR Green PCR Mastermix (SA Bioscience, Frederick, MD, USA) and 1 μl specific primer pairs. Reactions were run in triplicate with 40 cycles of amplification on an ABI Prism 7000 real-time PCR machine (Applied Biosystems, Foster City, CA, USA). The sequences of primers used were as follows: TATA box-binding protein, sense (5'-TGCACAGGAGCCAAGAGTGAA-3') and antisense (5'-CACATCACAGCTCCCCACCA-3'); and collagen II, sense (5'-TGCTGCCCAGATGGCTGGAGGA-3') and antisense (5'-TGCCTTGAAATCCTTGAGGCCC-3') 22 . The expression level of collagen II was normalized relative to the expression of TATA box-binding protein measured in parallel samples.

Confluent human chondrocyte monolayers were made serum-free overnight before treating with purified recombinant human eNAMPT (0 to 5 μg/ml) overnight followed by stimulation with IGF-1 (50 ng/ml) for 0 to 60 minutes for signaling studies. In some experiments, cells were pretreated with 10 μM MEK inhibitor (U0126) for 30 minutes followed by treatment with eNAMPT or IGF-1. We have previously shown that treatment of cells with MEK inhibitor (U0126) did not affect chondrocyte viability 23 . After incubation, cells were washed with PBS and lysed with lysis buffer that contained 20 mM Tris (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM tetrapyrophosphate, 1 mM glycerol phosphate, 1 mM Na₃VO₄, 1 μl/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride. Lysates were centrifuged to remove insoluble material, and the soluble protein concentration was determined with BCA reagent (Thermo Scientific, Rockford, IL, USA). Samples containing equal amounts of total protein were separated by SDS-PAGE, transferred to nitrocellulose, and probed for signaling proteins. Immunoreactive bands were detected using the ECL system (GE Healthcare). All immunoblotting experiments were repeated at least three times with similar results.

Data were expressed as the mean ± standard deviation and subjected to analysis of variance using StatView 5.0 software (SAS Institute, Cary, NC, USA). P ≤ 0.05 was considered significant.

We examined the effect of eNAMPT on IGF-1-stimulated PG synthesis. Pretreatment of chondrocytes with eNAMPT overnight, followed by IGF-1 stimulation for an additional 24 hours, inhibited IGF-1-induced PG synthesis. Inhibition by eNAMPT occurred in a dose-dependent manner with maximum inhibition observed at 5 μg/ml (Figure 1A). Interestingly, overnight treatment of chondrocytes with eNAMPT alone inhibited basal PG synthesis (Figure 1B).

IGF-1 is known to promote synthesis of collagen type II, a major component of the cartilage matrix. Since we found that eNAMPT inhibits IGF-1-stimulated PG synthesis, we were interested to examine the effect of eNAMPT on collagen type II production. Our data showed that pretreatment of chondrocytes with 5 μg/ml eNAMPT inhibited both basal and IGF-1-stimulated collagen type II expression and synthesis (Figure 2A, B).

Since our data showed that pretreatment of chondrocytes with eNAMPT inhibited IGF-1-mediated PG synthesis and collagen production, we wanted to examine the effect of eNAMPT on IGF-1 signaling. IGF-1-mediated activation of AKT has been shown essential for PG synthesis and collagen type II synthesis 23 24 . Stimulation of normal chondrocytes with IGF-1 resulted in phosphorylation of the IGF-1 receptor (Y-1135/36) and the downstream signaling molecules, including IRS-1 (Y-612; activating site) and AKT (S-473) (Figure 3A, lane 2), whereas eNAMPT alone did not stimulate phosphorylation of IGF-1 receptor, IRS-1 or AKT (Figure 3A, lane 3) but did stimulate a robust and sustained phosphorylation of ERK1/2 (at least for 60 minutes) compared with transient phosphorylation (15 minutes) with IGF-1 (Figure 3B). Pretreatment of chondrocytes with eNAMPT (dosage 0 to 5 μg/ml) followed by IGF-1 stimulation did not change the phosphorylation status of the IGF-1 receptor; however, eNAMPT inhibited the IGF-1-mediated phosphorylation of IRS-1 (Y-612) and downstream AKT (serine-473) (Figure 3A, lanes 4 to 6).

In addition, pretreatment of chondrocytes with eNAMPT stimulated increased phosphorylation of IRS-1 at the serine-312 residue (Figure 4A), which is inhibitory to IGF-1 signaling. Pretreatment of chondrocytes with the MEK inhibitor (U1206) inhibited eNAMPT-induced phosphorylation of IRS-1 at the serine-312 residue (Figure 4A) and restored phosphorylation of IRS-1 (Y-612) and AKT (serine-473) equal to the level stimulated by IGF-1 alone (Figure 4B). Since phosphorylation and activation of AKT are important steps in IGF-1-stimulated PG synthesis and collagen expression, we quantified the relative levels of phosphorylated AKT to total AKT from the dataset presented in Figure 4B (see Figure 4C). Our data showed that pretreatment of cells with eNAMPT followed by IGF-1 stimulation decreased AKT phosphorylation by 40%; however, treatment with MEK inhibitor restored IGF-1-induced AKT phosphorylation to 100% (Figure 4C). Taken together, these results suggest that eNAMPT does not directly inhibit the IGF-1 receptor, but activates a separate signaling pathway that results in ERK activation, which then inhibits the IGF-1-mediated activation of IRS-1 and AKT in chondrocytes through serine phosphorylation of IRS-1.