The experiments for this study were performed with human chondrocytes from knee cartilage of OA patients who had undergone endoprothetic surgery and knee cartilage harvested from cadaver donors with healthy (termed non-OA) cartilage. We used 16 OA-patients (6 male and 10 female) and 10 non-OA donors (6 male and 4 female) in the study. The average age of the OA patients was 62 years (+/- 9 years) and for the non-OA donors was 30 years (+/- 11 years). The use of human tissue was approved by the ethics committee at the University of Regensburg (Az: 14-101-0189, email: ethikkommission@klinik.ukr.de). The cartilage was mechanically crushed and digested by collagenase type 2 (Worthington, Lakewood, USA) for 16 h at 37°C. The chondrocytes were then seeded at a density of 12,000 cells per cm² in T175 vials (Corning, Corning, USA) and cultivated with 25 ml culture medium consisting of DMEM-F12 Ham (Sigma-Aldrich, St. Louis, USA), 10% FCS (Sigma-Aldrich, St. Louis, USA) and 1% Pen/Strep (Sigma-Aldrich, St. Louis, USA) in an incubator (Thermo Fisher Scientific, Waltham, USA) at 37°C and 5% CO₂. After 5 days, the culture medium was refreshed for the first time. The adherent chondrocytes were expanded up to a confluence of 70 - 80% by changing medium twice a week and either frozen in liquid nitrogen for storage or directly used for further experiments either in monolayer culture (2D) or in a fibrin gel culture system (3D). All experiments with OA chondrocytes were performed at passage 2 and with non-OA chondrocytes at passage 3.

Chondrocytes were stimulated in all experimental set ups with 10^-8 M and 10^-10 M of either SP (Sigma-Aldrich, St. Louis, USA; #S6883) or αCGRP (Bachem, Bubendorf, Switzerland; #4013281) diluted in PBS for indicated time points.

Fibrin gels are a three-dimensional (3D) culture systems in which the cells are embedded in a fibrin matrix stabilizing the chondrogenic phenotype. Cells are supplied by diffusion of the culture medium through this matrix. OA and non-OA chondrocytes were cultivated in fibrin gels for a period of 7 days. According to Leyh et al. 2x10 (6) cells each were suspended in 10 µl fibrinogen (Sigma-Aldrich, St. Louis, USA) and 18 µl thrombin (Baxter, Deerfield, USA) and pipetted dropwise into the middle of a well of a 24-well plate (Sarstedt, Newton, USA) (11). Cultivation was performed with daily medium change of 500 µl chondrogenic medium [composition according to (12)] including the neuropeptides SP or αCGRP in a concentration of 10^-8M and 10^-10M diluted in PBS or PBS as a control for the neuropeptide effects. The fibrin gel cultures were used for the isolation of RNA and the supernatants were used for ELISA-based analysis.

For protein analysis, chondrocytes were cultured in 6-Well plates (20.000-30.000 cells per well) for 2-3 days and treated with SP (Sigma-Aldrich, St. Louis, USA; #S6883), αCGRP (Bachem, Bubendorf, Switzerland; #4013281), GR 82334 (10^-6 M, Tocris Bioscience, Bristol, UK; #1670), SB268262 (10µM, Tocris Bioscience, Bristol, UK; #4314), U73122 (5 µM, Tocris Bioscience, Bristol, UK; #1268) or a combination thereof. Afterwards, cells were washed with PBS, detached with trypsin–EDTA (Merck, Darmstadt, Germany) and after washing again with PBS, harvested and lysed. Total cell lysates were prepared with RIPA buffer (Invitrogen/Thermo Fisher, Waltham, MA, USA) containing proteinase inhibitor (Roche, Munich, Germany) and phosphatase inhibitor (Roche, Munich, Germany). Protein concentration was quantified with the BCA assay (Invitrogen/Thermo Fisher, Waltham, MA, USA) and lysate aliquots containing 25–50 µg of total protein (depending on the protein of interest) were boiled for 5 min with SDS-sample buffer containing β-mercapto-ethanol (Merck, Darmstadt, Germany) and subjected to a 10–15% SDS-PAGE. After electrophoretic separation, the proteins were transferred to nitrocellulose membranes (Bio-Rad, Hercules, CA, USA) and blocked with 5% dried milk (Carl Roth, Karlsruhe, Germany) and subsequently incubated with primary antibodies for 16 h at 4°C or 1 h at room temperature ( Table 1 ):

After washing, the membranes were incubated with rabbit HRP (horseradish peroxidase)-coupled secondary antibody (1:10,000, Jackson Immuno Research, West Grove, PA, USA). Proteins were detected using ECL detection reagents (Invitrogen/Thermo Fisher, Waltham, MA, USA). Western blot signals were analyzed densitometrically using Photoshop CS3 and the ratio between the protein of interest and ß-actin or between the phosphorylated and not phosphorylated proteins was calculated.

All gene expression analysis were performed with RNA isolated from fibrin gels (3D). For the isolation of RNA from fibrin gels, the MasterPure Complete RNA Purification Kit (Epicentre, Madison, USA, distributed by Biozym, Hessisch Oldendorf, Germany) was used according to Leyh et al. (11, 13). The RNA concentration was determined using a NanoDrop Spectrophotometer (260/280nm) (Thermo Fisher Scientific, Waltham, USA). To generate single-stranded cDNA, RNA was reverse transcribed with an AffinityScript QPCR cDNA Synthesis Kit (Stratagene, San Diego, CA, USA) and PCR was performed with the Mx3005P QPCR System from Agilent Technologies using Brilliant II SYBER Green qPCR Mastermix (Agilent Technologies, Santa Clara, CA, USA). Gene expression in chondrocytes were analyzed relatively according to the formula 2^-ΔΔΔCT using the MX3005P QPCR system (Agilent Technologies, Santa Clara, USA) and were analyzed with the MxPro QPCR software for Mx3000P and Mx3005P QPCR systems (Stratagene, La Jolla, USA). The results were calibrated to the gene expression in untreated control cells, and normalized to TBP and 18s. Primer sequences are shown in Table 2 .

The CellTiter-Blue Cell Viability Assay (Promega, Germany) was performed to assay viability of chondrocytes stimulated with SP or αCGRP for 1, 3 or 7 days in 2D cell culture and for 7 days in 3D cell culture (fibrin gels). 10.000 cells/well were seeded into 96-well plates for 1 day of SP or αCGRP stimulation, 8.000 cells/well were seeded into 96-well plates for 3 days of SP or αCGRP stimulation and 7.000 cells/well were seeded into 48-well plates for 7 days of SP or αCGRP stimulation. Chondrocytes were daily treated with SP- or αCGRP-containing medium. Fibrin gels were prepared with 500.000 chondrocytes embedded in 2,5 µl Fibrinogen and 4,5 µl Thrombin, and treated for 7 days with SP or αCGRP with daily medium changes and addition of neuropeptides. After indicated stimulation time, the amount of resazurin reduced to resorufin (pink and highly fluorescent) was determined at 579/584 nm with a Tecan ELISA reader (Maennedorf, Switzerland).

For assaying the cell growth, equal numbers of cells (20,000-30,000) were seeded in 6-well plates in triplicates and treated with SP or αCGRP each day. The cells of one well were detached each day and counted with a cell counter (Cedex, Roche, Germany) for up to 6 days. The doubling time was calculated using the following formula:

For assaying apoptosis, equal numbers of cells (5000) were seeded in black 96-well plates in triplicates and treated with SP or αCGRP for 1 day. Caspase-3/7 enzymatic activity was measured as an indicator of apoptosis using the Apo-ONE Homogeneous Caspase-3/7 assay (Promega, Fitchburg, WI, USA) according to the manufacturer’s instructions. A non-fluorescent caspase substrate (Z-DEVD-R110), added to the HTB94 cells, was thereby cleaved into fluorescent molecules with an emission maximum at 521 nm and measured in a microplate reader (Tecan, Männedorf, Switzerland).

For assaying senescence, equal numbers of cells (5000) were seeded in 96-well plates in triplicates and treated with SP or αCGRP for 1 day. For the determination of cellular senescence, we used the Mammalian SA-β-Galactosidase Assay Kit (Thermo Fisher Scientific, Waltham, USA) according to manufacturer’s instruction. Thereby, we measured the senescence-associated β-galactosidase (SA-β-gal) activity by using a fluorometric substrate, producing a bright yellow color with a peak absorbance at 420 nm that can be quantified using a microplate reader (Tecan, Männedorf, Switzerland).

For measuring the adhesion of chondrocytes on cell culture dishes, 1000 cells were seeded (per well of a 96-well plate) together with SP or αCGRP. Culture medium and non-adhered cells were removed 30 min later. Cells were washed with PBS and fixed with 1% Glutaraldehyd (Merck, Darmstadt, Germany) for 30 min at rt. Crystal violet solution (in 0,02% aqua dest.) (Carl Roth, Karlsruhe, Germany) was added to the wells at rt for 15 min. After removing of the staining solution, wells were washed with deionized H₂O and crystal violet was extracted from the cells by adding 70% ethanol for 3 h. Plates were shaken gently at room temperature for 15 min and 100 µl of the extracted crystal violet were transferred to a flat-bottom 96-well plates. Absorbance was measured at 590 nm in a microplate reader (Tecan, Männedorf, Switzerland).

The amount of IL6 or IL1ß in the cell culture supernatant of the fibrin gel cultures was measured according to manufacturer’s instruction using a human IL-6 ELISA (Ray Biotech, Norcross, GA, US) or a human IL-1ß ELISA (R&D, Minneapolis, MN, US). Chondrocytes in fibrin gels were daily treated with SP or αCGRP for 7 days and supernatant samples were harvested on the last day. Standards and samples are pipetted into the wells of an antibody pre-coated plate and IL-6 or IL1ß present in a sample or standard is bound by the immobilized antibody. The standard curve is used to determine the concentration of the samples. Standards and samples are measured at 450 nm in a microplate reader (Tecan, Männedorf, Switzerland).

Chondrocytes were cultured in fibrin gels and treated daily with SP or αCGRP for 7 days. Fibrin gels were homogenized and digested with pepsin (10 mg/ml in 0.05 M acetic acid containing 0.4 M NaCl, Sigma-Aldrich, St. Louis, MO, USA) for 48 h at 4°C and further digested with elastase (1 mg/ml in Tris-buffered saline (TBS) pH 8; Serva Electrophoresis, Heidelberg, Germany) for 24 h at 4°C. Samples were stored at -20°C until they were analyzed for GAG (glycosaminoglycan) or collagen type II content by ELISA.

Collagen type II content of digested fibrin gels were measured with specific sandwich ELISA, which recognize the native conformation of collagen type II chains (Chondrex, Redmond, WA, USA), according to the manufacturer’s instructions and measured in an ELISA-reader (Tecan, Männedorf, Switzerland) at 490 nm. All measurements were performed in triplets.

GAG content of digested fibrin gels were measured by DMMB (dimethyl methylene blue) (Merck, Darmstadt, Germany) assay. DMMB specifically stains chondroitin sulphate. Aliquots (25 µl) of the sample or of a standard dilution series (1:2) of known concentrations of chondroitin sulphate A (Merck, Darmstadt, Germany), used for calibration, were mixed with 250 μl DMMB staining solution (2ml formic acid and 2g sodium formate were dissolved in distilled water, pH 3.0 adjusted with HCl; 18 mg DMMB solved in 5 mL ethanol and added) and measured in an ELISA-reader (Tecan, Männedorf, Switzerland) at 595 nm. All measurements were performed in triplicate.

Gelatin zymography was performed to analyze MMP-2 or -9 activity with chondrocytes cultured either in 2D monolayer or 3D fibrin gels. For monolayer culture 50.000 cells/well were seeded into 6-well plates. Chondrocytes were cultured for 3 days with daily addition of fresh medium containing SP or αCGRP. For the last 24 hours, cells were cultured in serum-free medium containing SP or αCGRP. Fibrin gels were prepared with 500.000 chondrocytes embedded in 2,5 µl Fibrinogen and 4,5 µl Thrombin, and treated for 7 days with SP or αCGRP with daily medium changes and addition of neuropeptides Equal amount of supernatants were mixed with 5x loading buffer (20% glycerol, 2% SDS, 2mM EDTA, 0.02% bromophenol blue, 20 mM Tris/HCl, pH 8.0) and subjected to electrophoresis (10% SDS-polyacrylamide gels supplemented with 1% gelatin). Subsequently, gels were washed twice for 30 min in Triton-X 100 (2,5%) (Sigma-Aldrich, US) and in distilled water, and were developed for 1 day at 37°C in Tris/HCl (50mM), pH 8.5, containing CaCl2 (5mM). Finally, the gels were stained with Coomassie Brilliant Blue R250 (Serva, Germany) to detect protease activity. Signals were analyzed densitometrically using Photoshop CS3.

50.000 cells/well were seeded into 6-well plates. Chondrocytes were cultured for 3 days with daily addition of fresh medium containing SP or αCGRP. For the last 24 hours, cells were cultured in serum-free medium containing SP or αCGRP with or without addition of IL1ß (1ng/µl) for 24h (Thermo Fisher Scientific, Waltham, USA) known as an inducer of proteases. Fibrin gels were prepared with 500.000 chondrocytes embedded in 2,5 µl Fibrinogen and 4,5 µl Thrombin, and treated for 7 days with SP or αCGRP and with or without addition of IL1ß (1ng/µl) for 24h with daily medium changes and addition of neuropeptides. Equal amount of supernatants were mixed with 5x loading buffer (20% glycerol, 2% SDS, 2mM EDTA, 0.02% bromophenol blue, 20 mM Tris/HCl, pH 8.0) and subjected to electrophoresis. MM13 was detected using anti-rabbit polyclonal antibody to MMP13 (1:6000, 1 hour RT in 5% dry milk, Abcam, #ab39012) and rabbit horseradish peroxidise-coupled secondary antibody; 1:10.000, Jackson Immuno Research, US). Protein bands were detected with ECL detection reagents (Thermo Fisher Scientific, US).

For assaying the intracellular cyclic AMP (cAMP) level equal numbers of cells (20,000-30,000) were seeded in 6-well plates in triplicate. After preincubation with Rolipram (Sigma-Aldrich, St. Louis, USA) and following treatment with forskolin (Sigma-Aldrich, St. Louis, USA), SP or αCGRP, cAMP was measured using the cAMP ELISA Kit (Colorimetric) (Cell Biolabs Inc., San Diego, CA, USA) according to the manufacturer’s instruction. The cAMP-ELISA is a competitive enzyme immunoassay, using an anti-Rabbit IgG polyclonal coating antibody being adsorbed onto a microtiter plate. Cyclic AMP present in the sample or standard competes with Peroxidase cAMP Tracer for plate binding, in the presence of Rabbit Anti-cAMP Polyclonal Antibody. Following incubation and wash steps, a Peroxidase cAMP Tracer bound to the plate is detected with addition of substrate solution. The colored product formed is inversely proportional to the amount of cAMP present in the sample. The reaction is terminated by addition of a Stop solution and absorbance is measured at 450 nm. A standard curve is prepared from cAMP standard and sample concentration is then determined.

Statistical analysis was performed using Prism 6 (GraphPad Software Inc., San Diego, CA, USA). Results are presented in boxplots (median, min to max) or columns (mean, SD). Each assay was performed in replicates and repeated at least in three independent experiments. One sample t-tests were used to determine if the mean of the different groups (experimental group vs. control) were significantly different. Exact P-values were calculated. A value of p ≤ 0.05 was considered statistically significant. For the semi-quantitative analysis of Western blot intensity (or if specifically denoted), unpaired t-tests were used.

We analyzed the concentration of SP and αCGRP in the culture supernatants of chondrocytes from OA patients (OA-CH) and healthy donors (non-OA-CH). Whereas for SP no differences were found between both groups (OA-CH: 22,91 ± 4,762 pg/ml vs. non-OA-CH: 24,85 ± 3,218 pg/ml) ( Figure 1A ), the level of αCGRP was increased by trend in the supernatant of OA-CH (191,3 ± 93,77 pg/ml) compared to non-OA-CH (88,79 ± 36,00 pg/ml), although not statistically significantly ( Figure 1B ). The protein expression of neurokinin-1 receptor (NK1R), the receptor with the highest affinity for SP, was similar for OA- and non-OA-CH ( Figures 1C, D ). The receptor for αCGRP (CGRP-R) consists of two transmembrane protein subunits, the calcitonin receptor-like receptor (CRLR) unit and the receptor activity-modifying protein 1 (RAMP1) unit, with the latter subunit providing the specificity for αCGRP. The signal of the band corresponding to RAMP1 was significantly weaker in OA-CH, whereas the band signal intensity of CRLR in OA-CH was similar to that in non-OA-CH ( Figures 1C, D ).

OA and non-OA chondrocytes produced and secreted SP and αCGRP in comparably low concentrations (pg/ml). Non-OA chondrocytes showed higher expression of the αCGRP receptor subunit RAMP-1 compared to OA chondrocytes.

First, we analyzed effects of SP and αCGRP treatment on viability of OA- and non-OA chondrocytes at different culture time points. Both concentrations of SP reduced viability of OA-CH at the first day of culture but not at later culture time points compared to untreated controls. Also, both concentrations of αCGRP reduced viability of non-OA-CH at day 7 - the end point- of culture compared to untreated controls ( Figures 2A–C ). Viability of chondrocytes kept in 3D fibrin gel culture was not affected at any time point ( Figure 2D ).

We compared the proliferation rate of OA- and non-OA-CH to their respective untreated control cells by counting the cell numbers after 3 and 7 days of stimulation with SP or αCGRP in two different concentrations (10^-8M or 10^-10M). After 3 days of stimulation with 10^-8M SP, OA-CH had proliferated less compared with control cells ( Figure 3A ). This effect was lost after 7 days, but by then, stimulation with 10^-8M αCGRP leads to a significantly reduced proliferation rate in OA- and non-OA-CH ( Figure 3B ).

Apoptosis was also influenced by the two neuropeptides. Stimulation with SP (10^-8M), and also with αCGRP (10^-8M and 10^-10M) for 1 day, increased caspase 3/7 activity in OA-CH, indicating enhanced apoptosis in those cells. In contrast, 10^-10M αCGRP has beneficial effects on non-OA-CH by decreasing apoptosis significantly ( Figure 3C ).

As apoptosis is often closely related to cellular senescence, we measured the senescence-associated (SA)-ß-galactosidase activity as an indicator for senescence-related processes in OA- and non-OA-CH after 1 day of addition of SP and αCGRP. OA-CH respond with an enhanced SA-ß-galactosidase activity to addition of 10^-8M SP and αCGRP ( Figure 3D ). Even though there is no universal marker for cell senescence, most senescent cells increase expression of p16 (p16^ink4A), a cell cycle inhibitor that targets cyclin-dependent kinases (CDKs). Thus, we analyzed the gene expression of p16 after 7 days of incubation with both neuropeptides. αCGRP exerted opposite effects on OA- and non-OA-CH. Whereas in non-OA-CH, αCGRP reduced the expression of p16, in OA-CH αCGRP induced gene expression of this senescence marker ( Figure 3E ) in comparison with untreated control cells.

Chondrocyte adhesion is critical in cell based regenerative strategies for treatment of focal cartilage defects. As SP and αCGRP might be of interest for therapeutic approaches, we investigated the adhesion behavior of OA- and non-OA-CH during stimulation with the two neuropeptides. Both neuropeptides decreased the number of adherent non-OA-CH at 30 minutes after seeding; SP in a low concentration (10^-10M) and αCGRP in a high concentration (10^-8M) ( Figure 3F ). As we have observed direct effects of SP and αCGRP on different cellular metabolic properties like viability, proliferation, apoptosis, senescence and adhesion in OA- and non-OA-CH, we analyzed the metabolic activity of SP- or αCGRP-treated cells using a real-time metabolic bioanalyzer (Seahorse XF). No significant differences between any groups were detected ( Supplementary Figure S1 ).

Both neuropeptides affected metabolism of OA- and non-OA chondrocytes differently. SP stimulation mostly affected OA-chondrocytes whereas in non-OA chondrocytes, SP only modulated adhesion capability. Contrary, both OA- and non-OA chondrocytes responded to αCGRP stimulation.

To detect the influence of SP and αCGRP on chondrogenic/mesenchymal marker genes, we analyzed the expression levels of SOX9, ACAN, BMP4, VEGFR, ITGA11, COL9A1 and COL2A1 ( Figures 4A–G ). The OA- and non-OA-CH were encapsulated in fibrin gels for 7 days with daily SP or αCGRP addition and the gene expression levels were compared with the respective untreated control chondrocytes. We measured decreased gene expression of SOX9, ACAN ( Figures 4A, B ) and ITGA11 ( Figure 4E ) in OA-CH after addition of 10^-8 and 10^-10M SP, and of COL9A1 and COL2A1 in the presence of 10^-10M SP ( Figures 4F, G ). In contrast, the gene expression levels of BMP-4 ( Figure 4C ), VEGFR ( Figure 4D ), ITGA11 ( Figure 4E ) and COL9A1 ( Figure 4F ) were increased in non-OA-CH after 10^-8M αCGRP addition. In contrast to the effects on non-OA-CH, the gene expression of COL2A1 ( Figure 4G ), COL9A1 ( Figure 4F ) and ITGA11 ( Figure 4E ) gene expression were lower in OA-CH after 10^-8M αCGRP addition. COL9A1 gene expression in OA-CH was also decreased by 10^-10M αCGRP stimulation ( Figure 4F ). The biosynthesis of collagen type II was not affected ( Supplementary Figure S2 ) in either chondrocyte group after SP or αCGRP addition, but a decreased glycosaminoglycan (GAG) content in OA-CH containing fibrin gels after incubation with 10^-10M αCGRP was observed ( Figure 4H ).

Stimulation with SP affected only gene expression in OA-chondrocytes (reduction of anabolic gene expression of SOX9, ACAN, ITGA11, COL9A1, COL2A1) but not in non-OA chondrocytes. Stimulation with αGRRP induced gene expression (BMP4, VEGFR, ITGA11, COL9A1) in non-OA chondrocytes but reduced gene expression (ITGA11, COL9A1, COL2A1) in OA-chondrocytes. However, protein synthesis of collagen II remained unaffected as did mostly GAG synthesis except in OA chondrocytes after αCGRP stimulation.

SP and αCGRP effects are often related to inflammatory processes in various tissues and can influence the expression of pro-inflammatory cytokines. We therefore analyzed gene expression of TNFα, IL6, IL1ß, ICAM and COX2 ( Figures 5A–E ) and protein concentration of IL6 and IL1ß in culture supernatants ( Figures 5F, G ) after stimulation with SP or αCGRP. In OA-CH, 10^-8M SP resulted in an increased gene expression of TNFα, IL6 and IL1ß ( Figures 5A–C ). The gene expression of IL1ß was also increased when OA-CH were subjected to a lower dose of SP (10^-10M) ( Figure 5C ). The gene expression of the transmembrane protein ICAM-1 (CD54), a cell surface glycoprotein known for regulating leukocyte recruitment from circulation to sites of inflammation, was reduced in OA-CH after treatment with 10^-8M SP ( Figure 5D ). In non-OA-CH, 10^-8M and 10^-10M SP increased the gene expression of COX2 ( Figure 5E ), a key mediator of pro-inflammatory pathways, and 10^-8M SP increased the gene expression level of IL6 in non-OA-CH compared to control cells ( Figure 5B ). We also measured protein levels in cells with altered IL6 or IL1ß gene expression. IL6 protein amount was increased in the supernatant of non-OA-CH that had been exposed to 10^-8M SP, in comparison with controls ( Figure 5F ). However, in the OA-CH group, neither the IL6 ( Figure 5F ) nor IL1ß ( Figure 5G ) concentrations in the supernatants were affected by 10^-8M SP treatment. Notably, αCGRP did not affect any of the analyzed inflammatory markers neither in OA- nor in non-OA-CH ( Supplementary Figures S3A–E ). Notably, the IL1ß concentration in non-OA-CH supernatants was below the ELISA detection limit. In addition, the expression and activity of different MMPs was analyzed in OA- CH after SP stimulation. MMP-9 was not detectable at all and MMP-13 was mostly only detectable after IL1ß stimulation in 3D culture only and not affected by SP stimulation. Activity of pro- and active forms of MMP-2 was not affected by SP stimulation neither in 2D- nor in 3D culture ( Supplementary Figures S4A–E ).

Expression of pro-inflammatory genes remained unaffected by αCGRP stimulation but was induced by SP stimulation (except ICAM gene expression which was reduced) in both chondrocyte groups. In addition, SP stimulation increased IL6 concentration in culture supernatant of non-OA CH.

G-protein-coupled receptors (GPCR) like NK1-R and CLRL/Ramp1 (CGRP-R) are able to activate different signaling pathways via different alpha-subunits. To elucidate the signaling mechanism in OA- and non-OA-CH after SP and αCGRP treatment, we examined the level of the second messenger cyclic adenosine-mono-phosphate (cAMP), which is known to be activated after assembling of GPCRs with Gαs subunits. To do this, we pre-incubated the cells with Rolipram, a selective inhibitor of type 4 cyclic nucleotide phosphodiesterases (PDE4), which mediates cAMP degradation. In non-OA-CH a high concentration (10^-8M) of SP and αCGRP provoked a comparable increase in the cAMP-level ( Figure 6A ). In contrast, in OA-CH, SP did not induce alterations in the cAMP level, but αCGRP increased the amount of cAMP significantly when applied in a concentration of 10^-10M, and by trend also at 10^-8M ( Figure 7A ). It is noteworthy that in non-OA-CH a markedly stronger cAMP response was generated by forskolin treatment, which represents the internal positive treatment control (non-OA-CH: cAMP mean after forskolin: 2658 pM (+/-1574); OA-CH: cAMP mean after forskolin: 699 pM (+/-183)).

αCGRP stimulation provoked increased cAMP concentrations in non-OA and OA, whereas SP only induced cAMP level in non-OA-CH.

The activation of the ERK and AKT signaling pathways is a critical step in the regulation of chondrocyte gene expression in a chronic inflammatory state. Here, we analyzed if SP or αCGRP can induce the phosphorylation of ERK or AKT in OA- and non-OA-CH. In general, non-OA-CH developed a stronger response to the treatment with SP and αCGRP than OA-CH ( Figures 6 and 7 ). Both neuropeptides induced phosphorylation of ERK and AKT within 5-15 minutes without alteration of the unphosphorylated ERK and AKT protein concentration in non-OA-CH ( Figures 6B–D ).

We hypothesized that the receptors for SP and αCGRP (NK1-R and CGRP-R) can assemble not only with Gαs but also with the αq subunit (Gq) in chondrocytes, as this is often the case when ERK and AKT pathways are activated. Alpha q proteins couple to G proteins of GPCR to activate beta-type phospholipase C (PLC-β) enzymes which modulate the formation of phosphoinositol -3-phosphate (PIP3) which in turn is responsible for anchoring AKT/PKB at the membrane and thus modulates signaling. We therefore used a specific PLC-ß inhibitor (U73122) in combination with SP and αCGRP to test this hypothesis. The phosphorylation of ERK and AKT can be blocked when non-OA-CH are treated with SP and αCGRP together with U73122 ( Figures 6E–G ), indicating an involvement of PLC-ß in this signaling pathway. In OA-CH, the induction of ERK phosphorylation via SP and αCGRP treatment was detected within 5 minutes ( Figure 7B ), but the ratio of phospho-ERK to total ERK was lower compared with that in non-OA-CH in the presence of SP ( Figure 7C ). The phosphorylation of ERK via αCGRP in OA-CH was comparable with that seen in non-OA-CH ( Figure 7C ). In contrast, the AKT pathway could not be activated via SP or αCGRP treatment in OA-CH ( Figures 7B, D ). Treatment of OA-CH with the specific PLC-ß inhibitor U73122 in combination with SP and αCGRP lead to an inhibition of ERK phosphorylation ( Figures 7E, F ), similar to the effect seen in non-OA-CH. The unchanged levels of phospho-AKT in control, SP- and αCGRP-treated OA-CH decreased slightly after PLC-ß inhibition ( Figures 7E, G ).

To further show that the induction of ERK and AKT phosphorylation is SP- and αCGRP-specific, we used NK1-R antagonist (GR 82,334) and CGRP-R antagonist (SB 273779) for treatment of a human chondrocyte cell line (C28/I2), which reacts comparably to non-OA-CH to SP and αCGRP treatment. In both cases, the receptor antagonists were able to reduce the phosphorylation of ERK and AKT, when applied together with SP or αCGRP ( Supplementary Figures S5A, B ).

Both neuropeptides induced phosphorylation of ERK1/2 in OA- and in non-OA CH whereas AKT-phosphorylation was only induced by both neuropeptides in non-OA-CH.